Governance, Legislation, Tech & Democracy

In The Trump Era, Deregulation Has Become A Foreign Policy Directive

(Bruna Santos – Tech  Policy Press – 18 July 2025) On July 15th, the Trump administration launched an investigation into Brazil under the guise of examining “unfair” trading practices. The probe seeks to determine whether practices adopted by the Brazilian government and the recent Supreme Court ruling on the country’s platform liability regime amount to “unreasonable or discriminatory and burden or restrict US commerce.”. It’s the latest instance of the Trump administration repeating a tactic already deployed against the European Union and Canada, and follows threats of tariffs and letters to the Brazilian Supreme Court. The Office of the US Trade Representative’s probe into Brazil’s trade practices focuses on digital trade and electronic payment methods, as well as alleged attacks on American social media companies and other supposed unfair trading practices that would harm US companies, workers, farmers, and technology innovators. – https://www.techpolicy.press/in-the-trump-era-deregulation-has-become-a-foreign-policy-directive/

Brazil Has a Bridge to Defending the Internet

(James Görgen – Tech Policy Press – 18 July 2025) Last month, Brazil’s Federal Supreme Court (STF) approved a landmark legal precedent, declaring Article 19 of the Marco Civil da Internet (MCI) progressively and partially unconstitutional. The article, part of the cornerstone law governing the use of the Internet in Brazil, had stipulated that an Internet application provider could only be held legally responsible if it failed to remove content from its platform following a judicial order. The court’s actions follow a volatile period in Brazilian politics, during which a far right movement sought to overturn the result of a presidential election that saw the defeat of Jair Bolsonaro, who is now on trial over his role in an attempted coup d’état. Given the potential implications for how Brazil engages in platform regulation and determines civil liability for online speech, this trial put a Sword of Damocles over the head of the Brazilian far right, which used the lack of civil liability of digital platforms to aggressively distribute disinformation on the Internet. The decision has also reverberated far beyond São Paulo. Perhaps former President Bolsonaro’s most powerful ally, United States President Donald Trump is now attempting to use trade policy to intervene in the trial, and in Brazilian digital policy, part of a campaign underway since before he started his second term in office. On July 9, Trump sent a letter to President Luiz Inácio Lula da Silva raising tariffs on Brazilian products exported to the US to 50%. Trump also directed his trade representative to open a Section 301 investigation, accusing Brazil of adopting measures that jeopardize the unbridled operation of US tech platforms within Brazil. Trump’s intervention is a clear attempt to protect far right interests in Brazil and around the world, exposing the degree to which “free speech” arguments have been co-opted by the far right to preserve the ability to mislead and incite violence. – https://www.techpolicy.press/brazil-has-a-bridge-to-defending-the-internet/

What the Debate Over Differing Approaches to Online Freedom of Expression in the US and Brazil is Missing

(Fernanda Buril – Tech Policy Press – 18 July 2025) Last month, Brazil’s Supreme Court decided to expand Big Tech’s liability for online content—a move that drew praise from President Lula and concerns from Google. Meanwhile, in the United States, an aggressive push for deregulation is underway, despite the last-minute removal of a controversial moratorium on state AI laws from a federal budget bill. Leaders on both sides claim to be protecting democracy with their pro or anti-regulation stances. Just last week, tensions between the two countries escalated when President Donald Trump announced new tariffs on the Latin American country citing, among other things, what he called the Brazilian Supreme Court’s “Censorship Orders to US Social Media platforms.”. Trump’s attempt to directly influence other countries’ regulatory decisions can undermine their ability to find solutions that best fit their contexts. With technology advancing so quickly and affecting the information environment so profoundly, we need to move on from the philosophical debate on whether constraints on online content are good and focus instead on identifying when and to what extent they are necessary. When they are needed, we need to ensure regulations are locally tailored, adaptable, and complemented by longer-term strategies that will ideally render them obsolete. – https://www.techpolicy.press/what-the-debate-over-differing-approaches-to-online-freedom-of-expression-in-the-us-and-brazil-is-missing/

The Trump Administration’s AI Action Plan Is Coming. Here’s What to Look For

(Joshua Geltzer – Just Security – 18 July 2025) The Trump administration’s long-anticipated “AI Action Plan” is expected to be publicly released next week. In response to a Request for Information issued by the White House’s Office of Science and Technology Policy, stakeholders submitted over 10,000 public comments laying out their views on what should—and should not—be included.  That volume of feedback underscores the level of public interest in the administration’s approach to AI and the far-reaching consequences of U.S. government policy on this front. – https://www.justsecurity.org/117234/trump-ai-action-plan-what-to-look-for/

How cultural heritage can go green through digital preservation

(DigWatch – 18 July 2025) A Europe-wide survey has exposed how cultural heritage institutions (CHIs) protect humanity’s legacy digitally while unintentionally harming the environment. Instead of focusing only on efficiency, the European Climate Action Community recommends a shift towards environmentally sustainable and regenerative digital preservation. Led by the Environmental Sustainability Practice Task Force, the survey collected input from 108 organisations across 24 EU countries. While 80% of CHIs recognise environmental responsibility, just 42% follow formal environmental strategies and a mere 14% measure carbon footprints. – https://dig.watch/updates/how-cultural-heritage-can-go-green-through-digital-preservation

Time for Britain’s CMA to strike hard – or risk losing the cloud competition fight

(The Register – 18 July 2025) The UK’s ambition to become a global AI superpower hinges on a vibrant and competitive cloud market. The next few days will show if its competition regulator really appreciates both the pace of change and the scale of remedies needed to achieve both of these things. As the Competition and Markets Authority (CMA) highlighted in its provisional findings earlier this year, this crucial foundation is currently shaky. Amazon Web Services (AWS) and Microsoft Azure collectively dominate around 80 percent of UK cloud spending, exhibiting classic and adverse signs of entrenched market power. Their practices – from high exit fees and vendor lock-in, to preferential licensing and limited technical interoperability – are stifling competition and threatening the very innovation the government seeks to foster. The CMA’s investigation has shone a spotlight on these concerning practices, recognising that a truly competitive cloud landscape is essential for nurturing a flourishing AI innovation ecosystem. If businesses are locked into a handful of providers, their ability to experiment, scale, and develop cutting-edge AI applications will be severely constrained. – https://www.theregister.com/2025/07/18/comment_on_cma/

EU cloud gang wins Microsoft concessions, but fair software licensing group brands them ‘stalling tactic’

(The Register – 18 July 2025) A trade group of European cloud providers has claimed a small victory in bringing lower prices and more flexibility in deploying Microsoft software on their infrastructure, though the Coalition for Fair Software Licensing has blasted it as a “stalling tactic” by the software giant. Cloud Infrastructure Services Providers in Europe (CISPE) said the agreement meant that members could offer Microsoft software to their customers on a pay-as-you-go basis via Microsoft’s Cloud Solution Program (CSP-H), allowing stronger privacy for customers of European cloud providers. However, the agreement has failed to remove the technical tie-in between Entra ID (formerly Azure Active Directory) and Microsoft 365, restricting users in their choice of ID management when deploying Microsoft software in the cloud. – https://www.theregister.com/2025/07/18/cispe_microsoft_concessions/

As companies race to add AI, terms of service changes are going to freak a lot of people out

(The Register – 18 July 2025) WeTransfer this week denied claims it uses files uploaded to its ubiquitous cloud storage service to train AI, and rolled back changes it had introduced to its Terms of Service after they deeply upset users. The topic? Granting licensing permissions for an as-yet-unreleased LLM product. Agentic AI, GenAI, AI service bots, AI assistants to legal clerks, and more are washing over the tech space like a giant wave as the industry paddles for its life hoping to surf on a neural networks breaker. WeTransfer is not the only tech giant refreshing its legal fine print – any new product that needs permissions-based data access – not just for AI – is going to require a change to its terms of service. – https://www.theregister.com/2025/07/18/llm_products_terms_of_service/

Making Media Pluralism Work in the Age of Algorithms

(Urbano Reviglio – Tech Policy Press – 17 July 2025) Last September, over 60 civil society organizations and academics signed a call published in Le Monde advocating for “algorithmic pluralism” on social networks. Drawing from Francis Fukuyama’s 2021 “middleware” idea, the main proposal is to create the conditions for a consumer-facing market of algorithmic systems in social media, allowing users to choose which one to employ and thus shape their online experiences according to their tastes, interests, and moods. By choosing their customized feeds, users could outsource content curation to third-party providers, be it companies, newspapers, or even individual users and communities. In theory, such a market could spur competition, prompting platforms to refine their currently applied engagement-driven algorithms into more meaningful, interest-based curation systems. This idea is not new, and has been openly debated for several years now. Though it represents a promising path, it is widely acknowledged as a challenging one. Even in the best-case scenario, where most users enjoy many and more personalized experiences, it seems unlikely this would directly ensure a more pluralistic media environment. Traditional news consumption and algorithmic-driven news consumption are two fundamentally different experiences. Algorithmic systems behave differently with different users, and change across platforms and over time, eventually influencing not only what users are exposed to, thus affecting news discoverability and users’ worldviews, but even information behavior — how users consume news. – https://www.techpolicy.press/making-media-pluralism-work-in-the-age-of-algorithms/

Netherlands urges EU to reduce reliance on US cloud providers

(DigWatch – 17 July 2025) The Dutch government has released a policy paper urging the European Union to take coordinated action to reduce its heavy dependence on non-EU cloud providers, especially from the United States. The document recommends that the European Commission introduce a clearer and harmonized approach at the EU level. Key proposals include creating a consistent definition of ‘cloud sovereignty,’ adjusting public procurement rules to allow prioritizing sovereignty, promoting open-source technologies and standards, setting up a common European decision-making framework for cloud choices, and ensuring sufficient funding to support the development and deployment of sovereign cloud technologies. – https://dig.watch/updates/netherlands-urges-eu-to-reduce-reliance-on-us-cloud-providers

Meta faces fresh EU backlash over Digital Markets Act non-compliance

(DigWatch – 17 July 2025) Meta is again under EU scrutiny after failing to fully comply with the bloc’s Digital Markets Act (DMA), despite a €200 million fine earlier this year. The European Commission says Meta’s current ‘pay or consent’ model still falls short and could trigger further penalties. A formal warning is expected, with recurring fines likely if the company does not adjust its approach. The DMA imposes strict rules on major tech platforms to reduce market dominance and protect digital fairness. While Meta claims its model meets legal standards, the Commission says progress has been minimal. – https://dig.watch/updates/meta-faces-fresh-eu-backlash-over-digital-markets-act-non-compliance

EU presses pause on probe of X as US trade talks heat up

(Ars Technica – 17 July 2025) The European Commission has stalled one of its investigations into Elon Musk’s X for breaking the bloc’s digital transparency rules, while it seeks to conclude trade talks with the US. Brussels was expected to finalise its probe into the social media platform before the EU’s summer recess but will miss this deadline, according to three officials familiar with the matter. They noted that a decision was likely to follow after clarity emerged in the EU-US trade negotiations. “It’s all tied up,” one of the officials added. The EU has several investigations into X under the bloc’s Digital Services Act, a set of rules for large online players to police their platforms more aggressively. – https://arstechnica.com/tech-policy/2025/07/eu-presses-pause-on-probe-of-x-as-us-trade-talks-heat-up/

Texas Just Created A New Model for State AI Regulation

(Matthew Ferraro, Anna Z. Saber – Tech Policy Press – 17 July 2025) Texas Governor Greg Abbott (R) last month signed into law the Texas Responsible Artificial Intelligence Governance Act, or TRAIGA (HB 149), joining Colorado as only the second state to adopt a comprehensive AI governance law. TRAIGA brings forth a new approach to AI regulation, both by limiting Texas’ ability to punish companies with prohibitions on only a few intentional harms and by expanding the state’s investigatory powers. When the law goes into force on January 1, 2026, the net effect will likely subject many private enterprises to regulatory review and few to actual punishment. Given the size and economic heft of Texas, the law will likely have substantial ramifications for the AI industry. – https://www.techpolicy.press/texas-just-created-a-new-model-for-state-ai-regulation/

Pennsylvania criminalises malicious deepfakes under new digital forgery law

(DigWatch – 17 July 2025) Governor Shapiro has enacted a new statute enhancing Pennsylvania’s legal stance on AI-generated content by defining deceptive deepfakes as digital forgery. The law criminalises creating and distributing such content, mainly when used for deceit, highlighting a proactive response to deepening online threats. The legislation differentiates between uses of deepfakes: non-consensual impersonation will result in misdemeanour charges, while cases involving fraudulent intent, such as financial scams or political manipulation, are now classified as third-degree felonies. – https://dig.watch/updates/pennsylvania-criminalises-malicious-deepfakes-under-new-digital-forgery-law

Pediatricians Urge Carney Government to Prioritize New Online Safety Law

(Charlotte Moore Hepburn, Ashley Vandermorris, Alene Toulany, Rachel Mitchell – Tech Policy Press – 17 July 2025) Children and youth are being hurt online. As the new Carney government develops its online safety legislation, we’re calling on the Prime Minister to hold social media platforms accountable in keeping our kids safe online. Imagine cars without seatbelts. Cribs with dangerous drop sides. Toys with hidden choking hazards. Reckless, irresponsible, and dangerous. Yet this is the reality children and youth face every day in the digital world. With no pediatric-specific safeguards, children and youth are left to navigate toxic algorithms, addictive platforms, and predatory content with no basic safety standards designed with their well-being in mind. Too many kids are being hurt online. As doctors, we see the negative impact on our patients. We see parents struggling to protect their children. And we hear from kids themselves that they are experiencing harm. Canadians overwhelmingly agree that the government should address this generational hazard. – https://www.techpolicy.press/pediatricians-urge-carney-government-to-prioritize-new-online-safety-law/

ADB urges urgent reforms to bridge Pakistan’s digital divide

(DigWatch – 17 July 2025) The Asian Development Bank (ADB) has urged Pakistan to act quickly to close its widening digital gap by adopting urgent reforms. Its recommendations include accelerating delayed spectrum auctions and launching 5G services, significantly expanding fibre-optic infrastructure, especially in underserved areas, and reforming the telecom tax regime to attract private investment and lower consumer costs. The ADB also calls for gender-responsive and income-sensitive digital inclusion programs, greater investment in digital literacy, and targeted subsidies to help low-income and marginalised communities afford internet-enabled devices and connectivity. – https://dig.watch/updates/adb-urges-urgent-reforms-to-bridge-pakistans-digital-divide

How to govern AI in agriculture responsibly: risks, tools and solutions

(Rozita Dara – OECD.AI – 16 July 2025) The integration of data and artificial intelligence (AI) into agriculture and food systems has accelerated in recent years. Innovations such as soil sensors, satellite imagery, AI-powered crop management tools, disease monitoring and prediction, supply chain traceability systems, and weather forecasting models are transforming traditional farming and food production. For example, precision agriculture allows farmers to apply water, fertilisers, and pesticides only where needed, reducing environmental impacts and lowering costs. AI models forecast harvest times, predict pest outbreaks, and monitor crop health with remarkable accuracy, enabling proactive management. Blockchain technology tracks food from farm to plate, increasing consumer confidence in the products they consume. In short, AI and data-driven innovations are transforming how we produce, process, and distribute food, promising a more sustainable and efficient future for global food systems. – https://oecd.ai/en/wonk/how-to-govern-ai-in-agriculture-responsibly-risks-tools-and-solutions

YouTuber faces jail time for showing off Android-based gaming handhelds

(Ars Technica – 16 July 2025) There are countless Android-powered gaming handhelds, but they go beyond the usual slate of Android games by offering console emulation support. The problem is the game ROMs on these devices, which are not entirely legal. Italian YouTuber Once Were Nerd is learning how seriously some rightsholders are taking game piracy after agents from the country’s Guardia di Finanza showed up to confiscate his consoles. He now says the investigation could lead to criminal charges and the end of his channel. Once Were Nerd has produced YouTube content covering a plethora of gaming topics, including Android-based handheld game machines from the likes of Powkiddy and TrimUI. These devices usually run an older version of Android that has been heavily modified for gaming, featuring built-in emulation support for retro consoles like SNES, Nintendo 64, PlayStation Portable, GameCube, and more. They’ve become quite popular as the cost of mobile hardware has come down, making it possible to buy what is essentially an updated PSP or Game Boy Advance for $100 or less. Recently, Once Were Nerd attracted the attention of Italy’s Ministry of Economy and Finance, which is tasked with policing copyright in the country. In the video first spotted by Android Authority (which has an AI-generated English language track), the YouTuber explains that Guardia di Finanza appeared at his door in April with a search warrant. – https://arstechnica.com/gadgets/2025/07/youtuber-faces-jail-time-for-showing-off-android-based-gaming-handhelds/

Mapping the AI economy: Which regions are ready for the next technology leap

(Mark Muro and Shriya Methkupally – Brookings – 16 July 2025) Artificial intelligence has emerged as a general purpose technology with far-reaching consequences for industries, places, and people. AI systems promise to drive productivity by automating routine tasks or augmenting work, allowing humans to focus on higher-value activities. The technology is also accelerating the pace of discovery and innovation by analyzing vast datasets and identifying patterns humans might miss. And for that matter, AI enables more efficient resource allocation through intelligent forecasting and optimization. As such, AI could heavily influence the nation’s ability to achieve its larger goals, whether it be through faster drug development, personalized learning, or “virtual employees” optimizing supply chain complexities. With that in mind, it matters a lot whether and which U.S. cities and regions are prepared to facilitate AI development in high-quality ways, and are therefore demonstrating a readiness to truly benefit from future AI build-out. – https://www.brookings.edu/articles/mapping-the-ai-economy-which-regions-are-ready-for-the-next-technology-leap/

Partners or Provocateurs? Private-Sector Involvement in Offensive Cyber Operations

(Sezaneh Seymour, Brandon Wales – Lawfare – 16 July 2025) As the scale and sophistication of cyber threats from state and criminal actors grow, U.S. officials are reevaluating the long-standing policy that reserves offensive cyber operations as an exclusively governmental function. In this new Lawfare research report, we examine the risks and benefits of expanding private-sector participation in such operations. Rather than endorsing a specific policy change, we present a structured framework to guide a focused discussion among policymakers. The framework is built on three interdependent factors. First, it requires defining clear policy objectives, such as augmenting government capacity or disrupting adversary infrastructure. Second, it addresses the scope of authorized activities, clarifying what actions are permissible, who may be targeted, and where they may be attacked. Finally, it tackles the complex legal and liability considerations, including the potential legal authorities for such actions and the unresolved question of who bears responsibility when operations harm innocent third parties. By systematically addressing these questions, we aim to help policymakers clarify goals and mitigate the significant risks of escalation and diplomatic fallout before altering the rules of cyber offense. – https://www.lawfaremedia.org/article/partners-or-provocateurs–private-sector-involvement-in-offensive-cyber-operations

The Cybersecurity Patchwork Quilt Remains Incomplete

(Jim Dempsey – Lawfare – 16 July 2025) Earlier this year, the chief information security officer for JPMorgan Chase, Patrick Opet, published an open letter on software security. Three points stood out. The first was not news to anyone following cybersecurity and its policy debates: Software providers have prioritized rapid feature development over robust security. This is exactly the point, in almost the same words, that President Biden made in March 2023 when calling for legal reform to shift liability onto those developers that fail to take reasonable precautions to secure their software. Second, Opet focused on a problem that has not received much policy attention: The rapid shift to the software-as-a-service (SaaS) delivery model is creating a major, new form of vulnerability. The SaaS model, Opet warned, is “fundamentally reshaping how companies integrate services and data,” breaking down barriers between internal and external resources and rendering some traditional approaches to cybersecurity ineffective. Third, Opet called for collective action, asking other businesses to join with him in demanding better software security, implicitly admitting that even a large and security-savvy software customer such as JPMorgan Chase could not solve the problem alone. Opet did not mention legal or regulatory responses, but his letter screams “market failure.” Moreover, as a customer, the federal government is in exactly the same position as JPMorgan Chase and other large businesses. It is eager to take advantage of the efficiency and rapid innovation that SaaS and other cloud-based services offer, but it has not yet been successful in leveraging its procurement power to insist on better security. On June 6, President Trump issued an executive order that began to chart the outlines of a cybersecurity policy. Much of the initial news coverage focused on Trump’s repeal of measures ordered by President Biden in the closing days of his administration. In fact, the Biden initiatives repealed by Trump on government software procurement and digital identity were modest. While pushing even incremental actions though the Biden administration took huge effort, and while the drafters of the Biden order could take pride in their work, the provisions on procurement and identity would have done little to advance the cause of software security or cybersecurity in general. What received less attention was President Trump’s preservation and adoption of a majority of the elements in the Biden order as his own. And what received almost no attention at all was what both the Trump and Biden orders left unresolved: the glaring deficiencies that remain in critical infrastructure cybersecurity (most notably in the drinking water, health care, emergency services, and telecommunications sectors) and the costs that software developers impose on American industry and American taxpayers by continuing to produce insecure software. – https://www.lawfaremedia.org/article/the-cybersecurity-patchwork-quilt-remains-incomplete

San Francisco deploys AI assistant to 30,000 staff

(DigWatch – 16 July 2025) San Francisco has equipped almost 30,000 city employees, from social workers and healthcare staff to administrators, with Microsoft 365 Copilot Chat. The large-scale rollout followed a six-month pilot where workers gained up to five extra hours a week handling routine tasks, particularly in 311 service lines. Copilot Chat helps streamline bureaucratic functions, such as drafting documents, translating over 40 languages, summarising lengthy reports, and analysing data. The goal is to free staff to focus more on serving residents directly. – https://dig.watch/updates/san-francisco-deploys-ai-assistant-to-30000-staff

Trump launches $70 billion AI and energy investment plan

(DigWatch – 16 July 2025) President Donald Trump has announced a $70 billion initiative to strengthen America’s energy and data infrastructure to meet growing AI-driven demand. The plan was revealed at Pittsburgh’s Pennsylvania Energy & Innovation Summit, with over 60 primary energy and tech CEOs in attendance. The investment will prioritise US states such as Pennsylvania, Texas, and Georgia, where energy grids are increasingly under pressure due to rising data centre usage. Part of the funding will come from federal-private partnerships, alongside potential reforms led by the Department of Energy. – https://dig.watch/updates/trump-launches-70-billion-ai-and-energy-investment-plan

To Make Sure AI Advances Democracy, First Ask, ‘Who Does It Serve?’

(Tech Policy Press – 16 July 2025) “When looms weave by themselves, man’s slavery will end.“ I happened upon that ancient line from Aristotle in 1964, in a New York Times article on automation and employment. It has shaped my imagination and career ever since. I learned to program during a summer job at Bell Labs, and was thrilled at how computers could end certain forms of drudgery. Then, around 1970, three experiences brought the future into focus. At a conference, I saw Doug Engelbart reprise his famous 1968 “Mother of All Demos” on augmenting human intelligence as contrasted to early notions of artificial intelligence, and also, I clicked on an implementation of Ted Nelson’s seemingly magical hyperlinks. Third, I delved into early concepts for augmenting human collaboration, as enabled by Murray Turoff’s computerized Delphi conferencing. In those early days of the digital technology revolution, I saw glimpses of the future of computing and networking that Steve Jobs later described as “bicycles for our minds.” My interests in epistemology, psychology, history, media, and economics added sociotechnical dimensions to that vision. Around 1990, working on financial market data news feeds, I saw how individual traders could select analytics filters. Then, around 2003, I proposed designs for an ecosystem of tools for collaborating on open innovation. – https://www.techpolicy.press/to-make-sure-ai-advances-democracy-first-ask-who-does-it-serve/

AI and the world of work: A real-time snapshot of how national businesses are adopting AI in Israel

(Daniel Roash – OECD.AI – 15 July 2025) Artificial Intelligence is no longer a futuristic idea—it’s a force changing how businesses operate worldwide. Since the explosion of generative AI in late 2022, businesses are rushing to experiment with AI tools, exploring how these technologies can make operations more efficient, increase productivity, and reshape workflows. However, as AI hype grows, a fundamental challenge arises: how can we accurately measure AI adoption, and what might that reveal about its impact on businesses, workers, and the broader economy? Without reliable data, it is difficult for decision-makers to know whether AI is driving meaningful change or merely being used in limited ways with minimal impact. This is not a trivial question. Experimenting with free AI tools and deeply embedding sophisticated, paid AI systems should yield very different outcomes in terms of productivity, job creation, and job loss. Yet capturing the complexity of these scenarios in comparable data is challenging. AI evolves rapidly, its applications vary widely across industries, and the line between augmentation and automation is not always clear. Recognizing this challenge, the Israel Central Bureau of Statistics (CBS), after consultation with experts from the Bank of Israel, the Israel Innovation Authority, and the Israel Democracy Institute, a process that was facilitated by the Permanent Mission of Israel to the OECD, pioneered a dynamic, repeatable survey model designed to capture a real-time snapshot of how national businesses are adopting AI. This agile approach digs deeper than just asking whether businesses use AI. It enquires into how AI integrates into operations, whether it replaces tasks previously performed by employees, and whether businesses rely on free tools or invest in more advanced systems. The result is a living, evolving data source that helps policymakers, economists, and business leaders understand how AI is reshaping the economy in practice. – https://oecd.ai/en/wonk/ai-and-the-world-of-work-a-real-time-snapshot-of-how-national-businesses-are-adopting-ai-in-israel

Swiss to Release Open, Multilingual LLM Model

(AI Insider – 15 July 2025) Switzerland will release its first fully public large language model (LLM) in late summer 2025, developed by ETH Zurich, EPFL, and the Swiss National Supercomputing Centre (CSCS) to promote transparency, multilingualism, and open AI innovation. Funded by the ETH Board and trained on the “Alps” supercomputer powered by over 10,000 NVIDIA Grace Hopper Superchips, the Swiss LLM is designed for sovereign AI infrastructure and will be released under an Apache 2.0 license with full access to code, training data, and model weights. Featuring multilingual capabilities across more than 1,000 languages and two model sizes (8B and 70B parameters), the model targets broad adoption in science, government, education, and industry, with researchers emphasizing compliance with Swiss and EU regulations, ethical data sourcing, and transparent documentation. – https://theaiinsider.tech/2025/07/15/swiss-to-release-open-multilingual-llm-model/

Want Accountable AI in Government? Start with Procurement

(Nari Johnson, Elise Silva, Hoda Heidari – Tech Policy Press – 15 July 2025) In 2018, the public learned that the New Orleans Police Department had been using predictive policing software from Palantir to decide where to send officers. Civil rights groups quickly raised alarms about the tool’s potential for racial bias. But the deeper issue wasn’t just how the technology worked, but the processes that shaped its adoption by the city. Who approved its use? Why was it hidden from the public? Like New Orleans, all US cities rely on established public procurement processes to contract with private vendors. These regulations, often written into law, typically apply to every government purchase, whether it’s school buses, office supplies, or artificial intelligence systems. But this case exposed a major loophole in the city’s procurement rules: because Palantir donated the software for free, the deal sidestepped the city’s usual oversight processes. No money changed hands, so the agreement didn’t trigger standard checks such as a requirement for city council debate and approval. The city didn’t treat philanthropic gifts like traditional purchases, and as a result, key city officials and council members had no idea the partnership even existed. Inspired by this story and several others across the US, our research team, made up of scholars from Carnegie Mellon University and the University of Pittsburgh, decided to investigate the purchasing processes that shape critical decisions about public sector AI. – https://www.techpolicy.press/want-accountable-ai-in-government-start-with-procurement/

The UK’s Opportunity to Lead on Social Media Transparency

(Mark Scott – Tech Policy Press – 15 July 2025) In the world of online safety rulemaking, most attention has focused on the European Union’s Digital Services Act. But just across the English Channel, the United Kingdom’s Online Safety Act (OSA), a set of rules for social media, video-sharing, and internet messaging companies to remove illegal content like terrorist material and online financial fraud, is now well underway. That rulebook, which includes potential fines of up to 10 percent of a firm’s global revenue, just got a double revamp. – https://www.techpolicy.press/the-uks-opportunity-to-lead-on-social-media-transparency/

US House passes NTIA cyber leadership bill after Salt Typhoon hacks

(DigWatch – 15 July 2025) The US House of Representatives has passed legislation that would officially designate the National Telecommunications and Information Administration (NTIA) as the federal lead for cybersecurity across communications networks. The move follows last year’s Salt Typhoon hacking spree, described by some as the worst telecom breach in US history. The National Telecommunications and Information Administration Organization Act, introduced by Representatives Jay Obernolte and Jennifer McClellan, cleared the House on Monday and now awaits Senate approval. – https://dig.watch/updates/us-house-passes-ntia-cyber-leadership-bill-after-salt-typhoon-hacks

AI’s future in banking depends on local solutions and trust

(DigWatch – 15 July 2025) According to leading industry voices, banks and financial institutions are expected to play a central role in accelerating AI adoption across African markets. Experts at the ACAMB stakeholders’ conference in Lagos stressed the need for region-specific AI solutions to meet Africa’s unique financial needs. Niyi Yusuf, Chairman of the Nigerian Economic Summit Group, highlighted AI’s evolution since the 1950s and its growing influence on modern banking. – https://dig.watch/updates/ais-future-in-banking-depends-on-local-solutions-and-trust

Asia’s humanities under pressure from AI surge

(DigWatch – 15 July 2025) Universities across Asia, notably in China, are slashing liberal arts enrollments to expand STEM and AI programmes. Institutions like Fudan and Tsinghua are reducing intake for humanities subjects, as policymakers push for a high-tech workforce. Despite this shift, educators argue that sidelining subjects like history, philosophy, and ethics threatens the cultivation of critical thinking, moral insight, and cultural literacy, which are increasingly necessary in an AI-saturated world. They contend that humanistic reasoning remains essential for navigating AI’s societal and ethical complexities. – https://dig.watch/updates/asias-humanities-under-pressure-from-ai-surge

AI Isn’t Responsible for Slop. We Are Doing It to Ourselves

(José Marichal – Tech Policy Press – 15 July 2025) Our social media feeds are increasingly being overrun by AI slop, so much so that Fast Company’s Mark Sullivan has dubbed it the ”AI Slop Summer.” Critics are drawing attention to the looming dangers of AI driven content, as in a memorable John Oliver segment. Google’s new video AI generator, Veo3, is being used to produce racist and antisemitic videos that are then posted on social media. YouTube has taken notice of this phenomenon; on July 15 it announced that the YouTube Partner Program will exclude AI slop from being monetized. This content isn’t only banal—it can make our toxic public sphere even worse. But while the very real dangers of AI slop are often framed as large tech companies imposing dangerous tools on an unsuspecting public, perhaps we should also consider why we are so receptive to low quality content. – https://www.techpolicy.press/ai-isnt-responsible-for-slop-we-are-doing-it-to-ourselves/

As energy demands for AI increase, so should company transparency

(Josie Stewart, Brooke Tanner, and Nicol Turner Lee – Brookings – 14 July 2025) The rise of commercially available AI systems has led to alarming headlines: each time ChatGPT writes an email, it’s like dumping out a bottle of water, or every prompt answered by a chatbot is equivalent to powering a light bulb for about 20 minutes. As the technology becomes more integrated into our daily lives, researchers are attempting to quantify AI’s impact on the environment. While concerns over AI’s energy demand and its increasing ubiquity grow, successful efforts among policymakers to track or regulate the industry’s footprint have not. That is largely due to the lack of relevant data and reporting mechanisms from companies in the tech and energy sectors. – https://www.brookings.edu/articles/as-energy-demands-for-ai-increase-so-should-company-transparency/

Why Technology Won’t Save Us Unless We Change Our Behavior

(Frenk van Harreveld – Tech Policy Press – 14 July 2025) We can design greener tech, smarter AI, and healthier systems—but unless people use them, trust them, and stick with them, they won’t matter. Climate change, overstretched healthcare systems, and the rise of artificial intelligence are among the greatest challenges of our time. We often turn to technology for solutions: cleaner energy, more efficient healthcare, safer algorithms. But innovation is only half the story. The other half is us—and our behavior. Even the most promising technology fails if people don’t use it, understand it, or trust it. Green products have to be purchased and applied. Preventive health tools only work if lifestyles change. AI systems can boost efficiency, but only if users engage critically and responsibly. More often than not, the bottleneck is not in what we can build, but in what people actually do. – https://www.techpolicy.press/why-technology-wont-save-us-unless-we-change-our-behavior/

Vatican urges ethical AI development

(DigWatch – 13 July 2025) At the AI for Good Summit in Geneva, the Vatican urged global leaders to adopt ethical principles when designing and using AI. The message, delivered by Cardinal Pietro Parolin on behalf of Pope Leo XIV, warned against letting technology outpace moral responsibility. – https://dig.watch/updates/vatican-urges-ethical-ai-development – original text: https://www.vatican.va/roman_curia/secretariat_state/parolin/2025/documents/rc_seg-st_20250708_parolin-messaggio-aiforgood_en.html

How the EU’s Voluntary AI Code is Testing Industry and Regulators Alike

(Ramsha Jahangir – Tech Policy Press – 13 July 2025) Following months of negotiation and just weeks before the first legal deadlines under the EU AI Act take effect, the European Commission published the final Code of Practice on General-Purpose AI. The Code is voluntary and intended to help companies demonstrate compliance with the AI Act. It sets out detailed expectations around transparency, copyright, and measures to mitigate systemic risks. Signatories will need to publish summaries of training data, avoid unauthorized use of copyrighted content, and establish internal frameworks to monitor risks. Companies that sign on will see a “reduced administrative burden” and greater legal clarity, the Commission said. At the same time, both European and American tech companies have raised concerns about the AI Act’s implementation timeline, with some calling to “stop the clock” on the AI Act’s rollout. – https://www.techpolicy.press/how-the-eus-voluntary-ai-code-is-testing-industry-and-regulators-alike/

How US States Are Shaping AI Policy Amid Federal Debate and Industry Pushback

(Cristiano Lima-Strong – Tech Policy Press – 13 July 2025) In the United States, state legislatures are key players in shaping artificial intelligence policy, as lawmakers attempt to navigate a thicket of politics surrounding complex issues ranging from AI safety, deepfakes, and algorithmic discrimination to workplace automation and government use of AI. The decision by the US Senate to exclude a moratorium on the enforcement of state AI laws from the budget reconciliation package passed by Congress and signed by President Donald Trump over the July 4 weekend leaves the door open for more significant state-level AI policymaking. – https://www.techpolicy.press/how-us-states-are-shaping-ai-policy-amid-federal-debate-and-industry-pushback/

New AI traffic control promises faster commutes in Dubai

(DigWatch – 13 July 2025) Dubai’s Roads and Transport Authority (RTA) has rolled out a next-generation traffic signal system using AI and digital twin technology to optimise city traffic. The system, UTC-UX Fusion, dynamically adjusts real-time signals based on predictive traffic models. UTC-UX Fusion is expected to ease congestion during peak hours and at busy intersections by simulating road scenarios before implementation. RTA reports the system can reduce journey times by up to 20 percent on key routes. – https://dig.watch/updates/new-ai-traffic-control-promises-faster-commutes-in-dubai

The HAIP Reporting Framework: Feedback on a quiet revolution in AI transparency

(Arisa Ema, Fumiko Kudo, Toshiya Jitsuzumi – OECD.AI – 11 July 2025) AI is transforming our world, but who gets to look under the hood? In a world where algorithms influence elections, shape job markets, and generate knowledge, transparency is no longer just a “nice-to-have”—it’s the foundation of trust. This is one of the pressing challenges the Hiroshima AI Process (HAIP) addresses. HAIP is a G7 initiative launched in 2023 that aims to establish a global solution for safe and trustworthy AI. As part of this effort, it has developed, with the OECD, a voluntary reporting framework that invites AI developers to disclose how they align with international guidelines for responsible AI. Let’s look at some early insights from interviews with 11 of the first 19 participating organisations and a multistakeholder meeting held in Tokyo in June 2025. The findings reveal a picture that is both promising and complex, with lessons for the future of global AI governance. – https://oecd.ai/en/wonk/the-haip-reporting-framework-feedback-on-a-quiet-revolution-in-ai-transparency

A Proposed Scheme for International Diplomacy on AI Governance

(Judit Bayer – Tech Policy Press – 11 July 2025) Organizations and agreements related to AI governance are proliferating, and yet the vision of a consensual and binding governance agreement among the major powers remains out of sight. Even between the US and the EU – seen as robust allies for many decades – the dividing lines in AI policy have dramatically deepened. Nonetheless, new lines of alliance are dynamically developing around both major Atlantic powers. The EU’s impact on global regulation has come into doubt, especially as its AI Act appears to be losing support even among EU officials. This moment of uncertainty presents the opportunity to re-evaluate European priorities for AI governance, based on the changing geopolitical landscape. It suggests thinking and planning as if from scratch, going back to the drawing board to calibrate a novel approach towards a global AI governance agreement. – https://www.techpolicy.press/a-proposed-scheme-for-international-diplomacy-on-ai-governance/

June 2025 Tech Litigation Roundup

(Melodi Dinçer – Tech Policy Press – 11 July 2025) This roundup gathers and briefly analyzes tech-related cases across a variety of legal issues. The Tech Justice Law Project (TJLP) tracks these and other tech-related cases in the US, federal, state, and international courts in this regularly updated litigation tracker. – https://www.techpolicy.press/june-2025-tech-litigation-roundup/

Geostrategies

Denmark launches quantum initiative with world‑leading computer

(DigWatch – 18 July 2025) Positioning itself at the forefront of quantum innovation, Denmark has made a €80 million commitment led by the Novo Nordisk Foundation and EIFO. The Quantum initiative, operating under the new entity QuNorth, aims to deploy a system named Magne, expected to deliver unprecedented computational power when it becomes operational in late 2026 or early 2027. Magne will boast approximately 50 logical qubits, enabling tasks beyond the reach of classical computers. Built by Atom Computing and powered by Microsoft’s quantum stack, it marks one of the first Level 2 quantum systems with integrated error correction. – https://dig.watch/updates/denmark-launches-quantum-initiative-with-world%e2%80%91leading-computer

LEO becomes Africa’s first AI chatbot to enable remittances

(DigWatch – 18 July 2025) United Bank for Africa (UBA) has introduced cross-border payments via its AI chatbot, LEO, enabling instant transfers across African nations in local currencies. However, this marks the first time an AI-powered chatbot has facilitated such transactions on the continent. The feature relies on the Pan-African Payment and Settlement System, developed by Afreximbank, which allows seamless fund transfers between African countries supported by their central banks. It eliminates the need for hard currencies and aims to reduce fees while supporting financial inclusion. – https://dig.watch/updates/leo-becomes-africas-first-ai-chatbot-to-enable-remittances

Costa Rica rolls out first standalone 5G network with Ericsson

(DigWatch – 18 July 2025) Costa Rica has launched its first 5G Standalone (SA) network, marking a significant milestone in the country’s digital transformation. The deployment is part of a six-year agreement between Ericsson and Liberty Latin America. The network features fully native 5G infrastructure using Ericsson’s dual-mode 5G Core and advanced Radio Access Network technology. Over 1,400 sites are live, reaching around 3.7 million users with faster speeds, lower latency, and greater reliability. – https://dig.watch/updates/costa-rica-rolls-out-first-standalone-5g-network-with-ericsson

Europe’s quantum ambitions meet US private power and China’s state drive

(DigWatch – 17 July 2025) Quantum computing could fundamentally reshape technology, using quantum bits (qubits) instead of classical bits. Qubits allow complex calculations beyond classical computing, transforming sectors from pharmaceuticals to defence. Europe is investing billions in quantum technology, emphasising technological sovereignty. Yet, it competes fiercely with the United States, which enjoys substantial private investment, and China, powered by significant state-backed funding. The UK began quantum initiatives early, launching the National Quantum Programme 2014. It recently pledged £2.5 billion more, supporting start-ups like Orca Computing and Universal Quantum, alongside nations like Canada, Israel, and Japan. Europe accounted for eight of the nineteen quantum start-ups established globally in 2024, including IQM Quantum Computers and Pasqal. Despite Europe’s scientific strengths, it only captured 5% of global quantum investments, versus 50% for the US. – https://dig.watch/updates/europes-quantum-ambitions-meet-us-private-power-and-chinas-state-drive

Lee Jae-myung’s AI aspirations for South Korea may be too ambitious

(Afeeya Akhand – The Strategist – 17 July 2025) During his campaign, South Korean President Lee Jae-myung pledged 100 trillion Korean won (about A$112 billion) over a five-year period to turn the country into one of the top three AI powers in the world. However, faced with practical barriers such as funding constraints and talent shortages, Lee’s administration may fall short of its AI aspirations. South Korea’s transformation into a global AI powerhouse is a signature policy of Lee’s administration thus far. The policy focuses on developing the country’s AI capabilities, including a Korean-language large language model (LLM), by leveraging the expertise of five South Korean corporations, which are yet to be selected by the government. Lee’s AI push serves a dual purpose. First, in response to domestic economic woes, the government seeks to boost the country’s tech industry and accordingly consolidate global exports of AI technologies. Lee’s policy will build on the international efforts of South Korean tech companies such as KT and Naver, which are currently developing LLMs tailored to the cultural and linguistic nuances of the Thai and Saudi Arabian markets respectively. – https://www.aspistrategist.org.au/lee-jae-myungs-ai-aspirations-for-south-korea-may-be-too-ambitious/

India’s Digital Infrastructure Is Going Global. What Kind of Power Is It Building?

(Anuradha Sajjanhar – Tech Policy Press – 16 July 2025) During June’s high-stakes meetings in London around the India–UK Free Trade Agreement, India’s Commerce Minister, Piyush Goyal, made a striking declaration for a bold blueprint of UK-India collaboration: beyond tariff lines and visas, the deal aims to show how “the world can benefit from [India’s] skilled talent, cost-effective solutions, and growing capabilities in AI and emerging technologies.” By pitching Aadhaar-based systems, UPI-style payments (United Payments Interface) and CoWIN-style (Covid Vaccine Intelligence Network) certification as part of the partnership, Goyal signaled that India sees its digital governance model not only as domestic policy, but as exportable infrastructure and diplomatic leverage on the world stage. The question is whether this infrastructure is a universal engine for development, or a new template of technocratic statecraft. Since 2014, Prime Minister Narendra Modi’s government has dramatically expanded its digital governance infrastructure. Platforms like Aadhaar (a biometric ID system), UPI (a payments interface), and the India Stack (a set of public APIs) now form the backbone of how services are delivered, populations are tracked, and the state is imagined. These systems are increasingly promoted as models for other countries, especially in the Global Majority. With support from global development institutions and philanthropies, India’s approach is being exported as Digital Public Infrastructure (DPI): low-cost, open-source, and scalable. Yet beyond the language of inclusion and innovation, this technology demands a deeper set of questions. What kind of state is being built through digital systems? What histories of power do these infrastructures carry? And as other countries adopt India’s model, what kinds of politics are they also inheriting? – https://www.techpolicy.press/indias-digital-infrastructure-is-going-global-what-kind-of-power-is-it-building/

Johnson Electric Partners with Shanghai Mechanical & Electric for Humanoid Robotics Business

(AI Insider – 16 July 2025) Johnson Electric and Shanghai Mechanical & Electrical Industry Co. (SMEIC) have announced two joint ventures in China focused on humanoid robotics, backed by a combined investment of approximately $21 million (150 million yuan). The Shanghai-based venture will handle sales, customer support, business development, research support, and after-sales service for humanoid robotics solutions across China, while the Shenzhen-based venture will specialize in engineering, design, and manufacturing of robotic hardware modules and integrated systems. SMEIC contributes its industrial manufacturing expertise, while Johnson Electric provides robotics technologies, aiming to create scalable production capabilities and customized engineering support to meet evolving customer demands in China’s growing robotics sector. – https://theaiinsider.tech/2025/07/16/johnson-electric-partners-with-shanghai-mechanical-electric-for-humanoid-robotics-business/

Oracle commits billions to expand AI infrastructure in Europe

(DigWatch – 16 July 2025) Oracle has confirmed a $3 billion investment in its AI and cloud infrastructure across Germany and the Netherlands over the next five years. The move aims to boost its capacity in Europe as demand for advanced computing services continues to rise. The company plans to invest $2 billion in Germany and $1 billion in the Netherlands, joining other major tech firms ramping up data centre infrastructure. Oracle’s strategy reflects broader market trends, with companies like Meta and Amazon committing large sums to meet AI-driven cloud needs. – https://dig.watch/updates/oracle-commits-billions-to-expand-ai-infrastructure-in-europe

Europe to launch Eurosky to regain digital control

(DigWatch – 16 July 2025) Europe is taking steps to assert its digital independence by launching the Eurosky initiative, a government-backed project to reduce reliance on US tech giants. Eurosky seeks to build European infrastructure for social media platforms and promote digital sovereignty. The goal is to ensure that the continent’s digital space is governed by European laws, values, and rules, rather than being subject to the influence of foreign companies or governments. To support this goal, Eurosky plans to implement a decentralised content moderation system, modelled after the approach used by the Bluesky network. – https://dig.watch/updates/europe-to-launch-eurosky-to-regain-digital-control

EXA to boost European connectivity with new fibre route and subsea cable

(DigWatch – 16 July 2025) EXA Infrastructure has launched a strategic 1,200 km high-capacity fibre route connecting London, Frankfurt, Amsterdam, and Brussels (FLAP cities), featuring the first new subsea cable in the North Sea corridor in 25 years. The new deployment includes 1,085 km of low-loss terrestrial fibre and a 115 km subsea segment using ultra-low-loss G.654C cable, running between Margate (UK) and Ostend (Belgium). The project also introduces two new landing stations, EXA’s 21st and 22nd globally, enhancing its infrastructure across the UK, Belgium, and the Netherlands. These efforts complement EXA’s prior investments in the Channel Tunnel route, including upgrades to in-line amplifier (ILA) facilities and modern, high-fibre-count cables. The new route is part of EXA’s broader push to improve Europe’s digital infrastructure with ultra-low latency, high-bandwidth, and scalable fibre paths between key hubs. – https://dig.watch/updates/exa-to-boost-european-connectivity-with-new-fibre-route-and-subsea-cable

Latin America struggling to join the global AI race

(DigWatch – 14 July 2025) Currently, Latin America is lagging in AI innovation. It contributes only 0.3% of global startup activity and attracts a mere 1% of worldwide investment, despite housing around 8% of the global population. Experts point to a significant brain drain, a lack of local funding options, weak policy frameworks, and dependency on foreign technology as major obstacles. Many high‑skilled professionals emigrate in search of better opportunities elsewhere. – https://dig.watch/updates/latin-america-struggling-to-join-the-global-ai-race

Malaysia enforces trade controls on AI chips with US origin

(DigWatch – 14 July 2025) Malaysia’s trade ministry announced new restrictions on the export, transshipment and transit of high-performance AI chips of US origin. Effective immediately, individuals and companies must obtain a trade permit and notify authorities at least 30 days in advance for such activities. The restrictions apply to items not explicitly listed in Malaysia’s strategic items list, which is currently under review to include relevant AI chips. The move aims to close regulatory gaps while Malaysia updates its export control framework to match emerging technologies. – https://dig.watch/updates/malaysia-enforces-trade-controls-on-ai-chips-with-us-origin

Indonesia opens AI centre with global tech partners

(DigWatch – 14 July 2025) Indonesia has inaugurated a National AI Centre of Excellence in Jakarta in partnership with Indosat Ooredoo Hutchison, NVIDIA and Cisco. The centre is designed to fast-track the adoption of AI and build digital talent to support Indonesia’s ambitions for its 2045 digital vision. Deputy Minister Nezar Patria said the initiative will help train one million Indonesians in AI, networking and cybersecurity by 2027. Officials and industry leaders stressed the importance of human capability in maximising AI’s potential. – https://dig.watch/updates/indonesia-opens-ai-centre-with-global-tech-partners

Spain awarded €12.3 million in contracts to Huawei

(Security Affairs – 14 July 2025) The Spanish Ministry of the Interior has awarded €12.3 million ($14.3 million) contracts to manage and store judicially authorized wiretaps used by law enforcement and intelligence agencies, raising concerns about potential Chinese government access due to the company’s ties to Beijing. Between 2021 and 2025, Spain awarded centralized contracts for digital storage of court-ordered wiretaps to Huawei, using its OceanStor 6800 V5 servers. These systems store and classify intercepted communications for law enforcement, adhering to National Security Framework standards and ICT Security Guidelines set by the National Cryptologic Center (CCN-STIC). The contracts, part of routine procurement processes managed by the General Directorate for the Rationalization and Centralization of Procurement, were publicly tendered and are listed on the State Procurement Platform. – https://securityaffairs.com/179884/intelligence/spain-awarded-e12-3-million-in-contracts-to-huawei.html

Iran seeks at least three cloud providers to power its government

(The Register – 14 July 2025) The Information Technology Organization of Iran (ITOI), the government body that develops and implements IT services for the country, is looking for suppliers of cloud computing. The org recently posted a notification of its desire to evaluate, grade, and rank cloud players to assess their suitability to host government services. At the end of the exercise, the organization hopes to have a panel of at least three cloud operators capable of handling government services. – https://www.theregister.com/2025/07/14/iran_cloud_panel_evaluation/

Terrorism

Assessing Terrorist Use of Virtual Asset Intermediaries

(RUSI – 14 July 2025) A research briefing by Allison Owen examines intermediary services that convert virtual assets to fiat currency in cases related to three groups: Hamas, Hezbollah and ISIS. This research brief is based, in part, on data provided by blockchain analytics company Crystal Intelligence, which was gathered through its internal investigations. – https://www.rusi.org/explore-our-research/publications/external-publications/assessing-terrorist-use-virtual-asset-intermediaries

Assessment of the Global Terrorism Threat Landscape in Mid-2025

(Soufan Center – 11 July 2025) Al-Qaeda’s franchise groups, affiliates, and regional branches have ebbed and flowed in strength, but at the moment, both al-Shabaab in Somalia and Jama’a Nusrat ul-Islam wa al-Muslimin (JNIM) in the Sahel are on the ascent. Terrorist groups like the Islamic State Khorasan Province (ISKP) benefit from leveraging artificial intelligence to facilitate their propaganda, thus freeing up more bandwidth for militants to plan external operations. It will soon be the rule, not the exception, that terrorists use emerging technologies in one or multiple aspects of the attack planning cycle. The litany of challenges is occurring against a backdrop of budget cuts related to counterterrorism capacity, which could make the U.S. more vulnerable to a major terrorist attack on U.S. soil perpetrated by any number of hostile actors. – https://thesoufancenter.org/intelbrief-2025-july-11/

Security

Knightscope Joins Forces with Palantir to Advance Public Safety AI-Driven Robotics

(AI Insider – 18 July 2025) Knightscope has signed a two-year agreement with Palantir Technologies to join the FedStart program, accelerating its entry into the U.S. federal security market. The partnership provides Knightscope with FedRAMP High and DoD Impact Level 5 accredited infrastructure, onboarding services, and Authority to Operate (ATO) support for deploying its autonomous security robots in federal settings. With access to Palantir-managed AWS GovCloud and compliance resources, Knightscope aims to position itself to support national security, public safety, and infrastructure protection amid increasing U.S. focus on autonomous systems leadership. – https://theaiinsider.tech/2025/07/18/knightscope-joins-forces-with-palantir-to-advance-public-safety-ai-driven-robotics/

xAI Faces Criticism from AI Researchers Over Safety Practices and Grok Model Issues

(AI Insider – 18 July 2025) A wave of criticism from leading AI safety experts has been directed at xAI, the artificial intelligence startup founded by Elon Musk, following a series of controversial incidents involving its chatbot, Grok. Researchers from organizations including OpenAI and Anthropic are calling out what they describe as reckless and opaque safety practices at xAI, diverging from established industry norms. The backlash follows several high-profile incidents involving Grok, including antisemitic responses and inappropriate character personas, prompting growing concern over the platform’s safety guardrails. The launch of Grok 4, xAI’s latest frontier model, lacked any published safety report or system card, a move researchers say undermines transparency and accountability. – https://theaiinsider.tech/2025/07/18/xai-faces-criticism-from-ai-researchers-over-safety-practices-and-grok-model-issues/

Most US teens use AI companion bots despite risks

(DigWatch – 18 July 2025) A new national survey shows that roughly 72% of American teenagers, aged 13 to 17, have tried AI companion apps such as Replika, Character.AI, and Nomi, with over half interacting with them regularly. Although some teens report benefits like practising conversation skills or emotional self-expression, significant safety concerns have emerged. Around 34% have been left uncomfortable by the bot’s behaviour, and one-third have turned to AI for advice on serious personal issues. Worryingly, nearly a quarter of users disclosed their real names or locations in chats. – https://dig.watch/updates/most-us-teens-use-ai-companion-bots-despite-risks

Afghan data breach prompts secret UK relocation

(DigWatch – 18 July 2025) A serious data breach involving nearly 19,000 Afghans who sought relocation to the UK has come to light following a High Court ruling. The incident occurred in February 2022 when a UK Special Forces HQ official mistakenly emailed a spreadsheet containing personal details to an unauthorised recipient. Names, contact details and family information of those who feared Taliban reprisals due to their ties to British forces were exposed. – https://dig.watch/updates/afghan-data-breach-prompts-secret-uk-relocation

EU helps Vietnam prepare for cyber emergencies

(DigWatch – 18 July 2025) The European Union and Vietnam have conducted specialised cyber‑defence training to enhance the resilience of key infrastructure sectors such as power, transportation, telecoms and finance. Participants, including government officials, network operators and technology experts, engaged in interactive threat-hunting exercises and incident simulation drills designed to equip teams with practical cyber‑response skills. – https://dig.watch/updates/eu-helps-vietnam-prepare-for-cyber-emergencies

Hackers hide malware using DNS TXT records

(DigWatch – 18 July 2025) Hackers are increasingly exploiting DNS records to deliver malware undetected, according to new research from DomainTools. Instead of relying on typical delivery methods such as emails or downloads, attackers now hide malicious code within DNS TXT records, part of the Domain Name System, often overlooked by security systems. The method involves converting malware into hexadecimal code, splitting it into small segments, and storing each chunk in the TXT record of subdomains under domains like whitetreecollective.com. – https://dig.watch/updates/hackers-hide-malware-using-dns-txt-records

Google Sues 25 Chinese Entities Over BADBOX 2.0 Botnet Affecting 10M Android Devices

(The Hacker News – 18 July 2025) Google on Thursday revealed it’s pursuing legal action in New York federal court against 25 unnamed individuals or entities in China for allegedly operating BADBOX 2.0 botnet and residential proxy infrastructure. “The BADBOX 2.0 botnet compromised over 10 million uncertified devices running Android’s open-source software (Android Open Source Project), which lacks Google’s security protections,” the tech giant said. “Cybercriminals infected these devices with pre-installed malware and exploited them to conduct large-scale ad fraud and other digital crimes.” – https://thehackernews.com/2025/07/google-sues-25-chinese-entities-over.html

From Backup to Cyber Resilience: Why IT Leaders Must Rethink Backup in the Age of Ransomware

(The Hacker News – 18 July 2025) With IT outages and disruptions escalating, IT teams are shifting their focus beyond simply backing up data to maintaining operations during an incident. One of the key drivers behind this shift is the growing threat of ransomware, which continues to evolve in both frequency and complexity. Ransomware-as-a-Service (RaaS) platforms have made it possible for even inexperienced threat actors with less or no technical expertise to launch large-scale, damaging attacks. And these attacks don’t just encrypt data now. They exfiltrate sensitive information for double and triple extortion, alter or delete backups, and disable recovery infrastructure to block restoration efforts. This is especially critical for small and midsize businesses (SMBs), which are increasingly targeted due to their leaner defenses. For an SMB generating $10 million in annual revenue, even a single day of downtime can cost $55,076, without factoring in the long-term impact on customer trust and brand reputation. While also considering the mounting pressure to meet compliance mandates, tightening regulations in sectors like finance and healthcare, and the evolving standards set by cyber insurance providers, it’s no longer enough to simply back up critical data. Organizations need a cyber resilience strategy that enables them to maintain operations even during major disruptions. Let’s examine where traditional backup strategies fall short and how SMBs can build true cyber resilience to keep their businesses running when it matters most. – https://thehackernews.com/2025/07/how-cyber-resilience-helps-it-defend-against-ransomwa.html

New Phobos and 8base ransomware decryptor recover files for free

(Bleeping Computer – 18 July 2025) The Japanese police have released a Phobos and 8-Base ransomware decryptor that lets victims recover their files for free, with BleepingComputer confirming that it successfully decrypts files. Phobos is a ransomware-as-a-service operation that launched in December 2018, enabling other threat actors to join as affiliates and utilize their encryption tool in attacks. In exchange, any ransom payments were split between the affiliate and the operators. While the ransomware operation did not receive as much media attention as other ransomware operations, Phobos is considered one of the most widely distributed ransomware operations, responsible for many attacks on businesses worldwide. – https://www.bleepingcomputer.com/news/security/new-phobos-ransomware-decryptor-lets-victims-recover-files-for-free/

Hackers scanning for TeleMessage Signal clone flaw exposing passwords

(Bleeping Computer – 18 July 2025) Researchers are seeing exploitation attempts for the CVE-2025-48927 vulnerability in the TeleMessage SGNL app, which allows retrieving usernames, passwords, and other sensitive data. TeleMessage SGNL is a Signal clone app now owned by Smarsh, a compliance-focused company that provides cloud-based or on-premisses communication solutions to various organizations. – https://www.bleepingcomputer.com/news/security/hackers-scanning-for-telemessage-signal-clone-flaw-exposing-passwords/

Peering into the ‘Double Black Box’ of National Security and AI

(Faiza Patel – Lawfare – 18 July 2025) Ashley Deeks’s “The Double Black Box” appears at an opportune time. National security agencies are racing to find ways to integrate machine learning (ML) and other forms of artificial intelligence (AI) into their work, most of which is conducted in classified settings. In this context, Deeks seeks to answer the critical question: “[H]ow do we hold the executive branch accountable for its use of AI in national security settings?”. The book amply demonstrates Deeks’s central thesis: that the opaque nature of AI combined with the secrecy endemic to national security programs creates a “double black box.” The result, Deeks explains, is that the traditional actors that are meant to serve as checks on classified executive branch policymaking—such as Congress, the courts, executive branch lawyers, and inspectors general—will have an increasingly hard time doing so. She pulls together a useful overview of several known uses of AI and ML tools in national security activities, highlighting two instances in which public pushback led to a change in agency plans. This is followed by a synopsis of the most common critiques of AI: the use of biased data that leads to incorrect outcomes; the difficulty determining whom to hold accountable when AI-based decisions turn out to be wrong; the tendency of human decision-makers to defer to automated outcomes; and the lack of transparency about how these systems produce decisions or results. – https://www.lawfaremedia.org/article/peering-into-the–double-black-box–of-national-security-and-ai

Chinese Hackers Target Taiwan’s Semiconductor Sector with Cobalt Strike, Custom Backdoors

(The Hacker News – 17 July 2025) The Taiwanese semiconductor industry has become the target of spear-phishing campaigns undertaken by three previously undocumented Chinese state-sponsored threat actors. “Targets of these campaigns ranged from organizations involved in the manufacturing, design, and testing of semiconductors and integrated circuits, wider equipment and services supply chain entities within this sector, as well as financial investment analysts specializing in the Taiwanese semiconductor market,” Proofpoint said in a report published Wednesday. The activity, per the enterprise security firm, took place between March and June 2025. They have been attributed to three China-aligned clusters it tracks as UNK_FistBump, UNK_DropPitch, and UNK_SparkyCarp. – https://thehackernews.com/2025/07/chinese-hackers-target-taiwans.html

UNC6148 deploys Overstep malware on SonicWall devices, possibly for ransomware operations

(Security Affairs – 17 July 2025) Google’s Threat Intelligence Group warns that a threat actor tracked as UNC6148 has been targeting SonicWall SMA appliances with new malware dubbed Overstep. Active since at least October 2024, the group uses a backdoor and user-mode rootkit to potentially enable data theft, extortion, or ransomware attacks. While these activities suggest financial motives, researchers have not yet confirmed them definitively. “GTIG assesses with high confidence that UNC6148 is leveraging credentials and one-time password (OTP) seeds stolen during previous intrusions, allowing them to regain access even after organizations have applied security updates.” reads the report published by Google. “Evidence for the initial infection vector was limited, as the actor’s malware is designed to selectively remove log entries, hindering forensic investigation; however, it is likely this was through the exploitation of known vulnerabilities.”. Google’s Threat Intelligence Group (GTIG) assesses with moderate confidence that UNC6148 used a zero-day RCE vulnerability to deploy OVERSTEP malware on SonicWall SMA appliances. – https://securityaffairs.com/180035/hacking/unc6148-deploys-overstep-malware-on-sonicwall-devices-possibly-for-ransomware-operations.html

Global operation targets NoName057(16) pro-Russian cybercrime network

(Europol – 16 July 2025) Between 14 and 17 July, a joint international operation, known as Eastwood and coordinated by Europol and Eurojust, targeted the cybercrime network NoName057(16). Law enforcement and judicial authorities from Czechia, France, Finland, Germany, Italy, Lithuania, Poland, Spain, Sweden, Switzerland, the Netherlands and the United States took simultaneous actions against offenders and infrastructure belonging to the pro-Russian cybercrime network. The investigation was also supported by ENISA, as well as Belgium, Canada, Estonia, Denmark, Latvia, Romania and Ukraine. The private parties ShadowServer and abuse.ch also assisted in the technical part of the operation. – https://www.europol.europa.eu/media-press/newsroom/news/global-operation-targets-noname05716-pro-russian-cybercrime-network

Retailer Co-op: Attackers snatched all 6.5M member records

(The Register – 16 July 2025) Co-op Group’s chief executive officer has confirmed that all 6.5 million of the organization’s members had their data stolen during its April cyberattack – Scattered Spider is believed to be behind the digital heist. Shirine Khoury-Haq confirmed the scale of the attack to the BBC Breakfast show on Wednesday, adding that the member file is what the attackers copied, but were thwarted before they could deploy ransomware. “The good thing was because we did block them, they could not erase what they did,” she said. “So we could monitor every mouse click, we saw every piece of code that they had written, we knew everywhere they went in our systems, and we were able to relay that back to the authorities.” – https://www.theregister.com/2025/07/16/coop_data_stolen/

Hyper-Volumetric DDoS Attacks Reach Record 7.3 Tbps, Targeting Key Global Sectors

(The Hacker News – 16 July 2025) Cloudflare on Tuesday said it mitigated 7.3 million distributed denial-of-service (DDoS) attacks in the second quarter of 2025, a significant drop from 20.5 million DDoS attacks it fended off the previous quarter. “Overall, in Q2 2025, hyper-volumetric DDoS attacks skyrocketed,” Omer Yoachimik and Jorge Pacheco said. “Cloudflare blocked over 6,500 hyper-volumetric DDoS attacks, an average of 71 per day.”. In Q1 2025, the company said an 18-day sustained campaign against its own and other critical infrastructure protected by Cloudflare was responsible for 13.5 million of the attacks observed during the time period. Cumulatively, Cloudflare has blocked nearly 28 million DDoS attacks, surpassing the number of attacks it mitigated in all of 2024. – https://thehackernews.com/2025/07/hyper-volumetric-ddos-attacks-reach.html

UNC6148 Backdoors Fully-Patched SonicWall SMA 100 Series Devices with OVERSTEP Rootkit

(The Hacker News – 16 July 2025) A threat activity cluster has been observed targeting fully-patched end-of-life SonicWall Secure Mobile Access (SMA) 100 series appliances as part of a campaign designed to drop a backdoor called OVERSTEP. The malicious activity, dating back to at least October 2024, has been attributed by the Google Threat Intelligence Group (GTIG) to a hacking crew it tracks as UNC6148. The number of known victims is “limited” at this stage. The tech giant assessed with high confidence that the threat actor is “leveraging credentials and one-time password (OTP) seeds stolen during previous intrusions, allowing them to regain access even after organizations have applied security updates.”. “Analysis of network traffic metadata records suggests that UNC6148 may have initially exfiltrated these credentials from the SMA appliance as early as January 2025.” – https://thehackernews.com/2025/07/unc6148-backdoors-fully-patched.html

Newly Emerged GLOBAL GROUP RaaS Expands Operations with AI-Driven Negotiation Tools

(The Hacker News – 15 July 2025) Cybersecurity researchers have shed light on a new ransomware-as-a-service (RaaS) operation called GLOBAL GROUP that has targeted a wide range of sectors in Australia, Brazil, Europe, and the United States since its emergence in early June 2025. GLOBAL GROUP was “promoted on the Ramp4u forum by the threat actor known as ‘$$$,'” EclecticIQ researcher Arda Büyükkaya said. “The same actor controls the BlackLock RaaS and previously managed Mamona ransomware operations.”. It’s believed that GLOBAL GROUP is a rebranding of BlackLock after the latter’s data leak site was defaced by the DragonForce ransomware cartel back in March. It’s worth mentioning that BlackLock in itself is a rebrand of another RaaS scheme known as Eldorado. The financially motivated group has been found to lean heavily on initial access brokers (IABs) to deploy the ransomware by weaponizing access to vulnerable edge appliances from Cisco, Fortinet, and Palo Alto Networks. Also put to use are brute-force utilities for Microsoft Outlook and RDWeb portals. – https://thehackernews.com/2025/07/newly-emerged-global-group-raas-expands.html

Abacus dark web drug market goes offline in suspected exit scam

(Bleeping Computer – 15 July 2025) Abacus Market, the largest Western darknet marketplace supporting Bitcoin payments, has shut down its public infrastructure in a move suspected to be an exit scam. Exit scams occur when the operator of a marketplace decides to vanish with the money they hold in escrow for various transactions between platform users. Blockchain intelligence firm TRM Labs reports that Abacus shutting down so abruptly has all the indications of either an exit scam or a covert law enforcement operation dismantling the activity. – https://www.bleepingcomputer.com/news/security/abacus-dark-web-drug-market-goes-offline-in-suspected-exit-scam/

Belk hit by May cyberattack: DragonForce stole 150GB of data

(Security Affairs – 15 July 2025) The infamous Ransomware group DragonForce claimed responsibility for the May disruptive attack on US department store chain Belk. The ransomware gang claimed it had stolen 156 gigabytes of data from Belk. Belk, Inc. is a major American department store chain, founded in 1888 in Monroe, North Carolina, and currently headquartered in Charlotte. Operating around 300 locations across 16 states, Belk offers apparel, footwear, home furnishings, jewelry, beauty products, and more. Belk suffered a cyberattack between May 7 and 11, 2025, where an unauthorized party accessed corporate systems and stole some internal documents. – https://securityaffairs.com/179958/data-breach/belk-hit-by-may-cyberattack-dragonforce-stole-150gb-of-data.html

NSA: Volt Typhoon was ‘not successful’ at persisting in critical infrastructure

(Jonathan Greig – The Record – 15 July 2025) Senior cybersecurity officials at the National Security Agency and FBI said the agencies have been successful in addressing some of the Chinese cyber campaigns targeting critical infrastructure in the U.S. During the International Conference on Cyber Security at Fordham University in New York City on Tuesday, experts spoke at length about Beijing’s so-called Typhoon campaigns — which have involved Chinese government and private sector groups launching attacks on U.S. government agencies and companies. Kristina Walter, director of the NSA’s Cybersecurity Collaboration Center, focused on Volt Typhoon, an effort by Chinese actors to preposition themselves on U.S. critical infrastructure for disruptive or destructive cyberattacks in the event of a kinetic conflict centered around Taiwan. – https://therecord.media/china-typhoon-hackers-nsa-fbi-response

Google says ‘Big Sleep’ AI tool found bug hackers planned to use

(Jonathan Greig – The Record – 15 July 2025) Google said a large language model it developed to find vulnerabilities recently discovered a bug that hackers were preparing to use. Late last year, Google announced an AI agent called Big Sleep — a project that evolved out of work on vulnerability research assisted by large language models done by  Google Project Zero and Google DeepMind. The tool actively searches and finds unknown security vulnerabilities in software. On Tuesday, Google said Big Sleep managed to discover CVE-2025-6965 — a critical security flaw that Google said was “only known to threat actors and was at risk of being exploited.” – https://therecord.media/google-big-sleep-ai-tool-found-bug

The Security Stakes in the Global Quantum Race

(Argyri Panezi – Just Security – 15 July 2025) The quantum era is around the corner. Major tech companies are announcing impressive breakthroughs in quantum advantage, quantum error correction, and quantum networking. Competing quantum chips are also reaching new heights, from IBM’s Condor breaking the 1,000-qubit barrier in December 2023 signaling the ability to dramatically expand computational power, to Google’s Willow, presented in December 2024, and Microsoft’s Majorana 1  announcement in February 2025 – a breakthrough that remains contested. The quantum race is international, with competition between major players in the West and the East. Public investments in quantum technologies have surged globally, reaching $42 billion in 2023. China leads with more than $15 billion in investments, followed by Germany, the United Kingdom, the United States, and South Korea. Similar to the global race for AI leadership, quantum technology has geopolitical dimensions. Commentators are drawing comparisons between the quantum race and the earlier, nuclear and space races. Should policymakers anticipate that the quantum race will pose major security and safety risks, as with nuclear power? If this proves to be the case, then the international community can expect security implications of analogous magnitude. However, by coordinating and acting early, governments have an opportunity to prevent harmful competition, anticipate societal impacts, and build inclusive governance frameworks that support responsible and equitable development and adoption of quantum technologies. – https://www.justsecurity.org/116473/security-stakes-global-quantum-race/

North Korean XORIndex malware hidden in 67 malicious npm packages

(Bleeping Computer – 15 July 2025) North Korean threat actors planted 67 malicious packages in the Node Package Manager (npm) online repository to deliver a new malware loader called XORIndex to developer systems. The packages collectively count more than 17,000 downloads and were discovered by researchers at package security platform Socket, who assess them to be part of the continued Contagious Interview operation. Socket researchers say that the campaign follows threat activity detected since April. Last month, the same actor infiltrated npm with 35 packages that dropped information stealers and backdoors onto developers’ devices. – https://www.bleepingcomputer.com/news/security/north-korean-xorindex-malware-hidden-in-67-malicious-npm-packages/

Are Cyber Defenders Winning?

(Jason Healey, Tarang Jain – Lawfare – 14 July 2025) On June 6, President Trump signed an executive order to “reprioritize cybersecurity efforts to protect America,” outlining a rough agenda “to improve the security and resilience of the nation’s information systems and networks.” As the administration develops a new cybersecurity strategy, it is essential that it understand and respond to a shifting trend in cyberspace: After a decades-long slump, defenders may finally be gaining the advantage. In the 1970s, computers could be kept secure simply by being in locked rooms. But when these computers were connected to networks, attackers gained the advantage. Despite decades of defensive innovations since then, defenders’ efforts are routinely overwhelmed by the gains made by attackers. Successful defense is possible—but only with substantial resources and discipline. – https://www.lawfaremedia.org/article/are-cyber-defenders-winning

AI fake news surge tests EU Digital Services Act

(DigWatch – 14 July 2025) Europe is facing a growing wave of AI-powered fake news and coordinated bot attacks that overwhelm media, fact-checkers, and online platforms instead of relying on older propaganda methods. According to the European Policy Centre, networks using advanced AI now spread deepfakes, hoaxes, and fake articles faster than they can be debunked, raising concerns over whether EU rules are keeping up. Since late 2024, the so-called ‘Overload’ operation has doubled its activity, sending an average of 2.6 fabricated proposals each day while also deploying thousands of bot accounts and fake videos. – https://dig.watch/updates/ai-fake-news-surge-tests-eu-digital-services-act

Azerbaijan government workers hit by cyberattacks

(DigWatch – 14 July 2025) In the first six months of the year, 95 employees from seven government bodies in Azerbaijan fell victim to cyberattacks after neglecting basic cybersecurity measures and failing to follow established protocols. The incidents highlight growing risks from poor cyber hygiene across public institutions. – https://dig.watch/updates/azerbaijan-government-workers-hit-by-cyberattacks

Italian defence firms hit by suspected Indian state-backed hackers

(DigWatch – 14 July 2025) An advanced persistent threat (APT) group with suspected ties to India has been accused of targeting Italian defence companies in a cyber-espionage campaign. Security researchers found that the hackers used phishing emails and malicious documents to infiltrate networks, stealing sensitive data. The attacks, believed to be state-sponsored, align with growing concerns about nation state cyber operations targeting critical industries. – https://dig.watch/updates/italian-defence-firms-hit-by-suspected-indian-state-backed-hackers

CBI Shuts Down £390K U.K. Tech Support Scam, Arrests Key Operatives in Noida Call Center

(The Hacker News – 14 July 2025) India’s Central Bureau of Investigation (CBI) has announced that it has taken steps to dismantle what it said was a transnational cybercrime syndicate that carried out “sophisticated” tech support scams targeting citizens of Australia and the United Kingdom. The fraudulent scheme is estimated to have led to losses worth more than £390,000 ($525,000) in the United Kingdom alone. The law enforcement effort, which was carried out on July 7, 2025, as part of Operation Chakra V, involved searches at three locations in Noida, one of which was a fully functional fraudulent call center operating from the Noida Special Economic Zone. – https://thehackernews.com/2025/07/cbi-shuts-down-390k-uk-tech-support.html

eSIM Vulnerability in Kigen’s eUICC Cards Exposes Billions of IoT Devices to Malicious Attacks

(Hacker News – 14 July 2025) Cybersecurity researchers have discovered a new hacking technique that exploits weaknesses in the eSIM technology used in modern smartphones, exposing users to severe risks. The issues impact the Kigen eUICC card. According to the Irish company’s website, more than two billion SIMs in IoT devices have been enabled as of December 2020. The findings come from Security Explorations, a research lab of AG Security Research company. Kigen awarded the company a $30,000 bounty for their report. An eSIM, or embedded SIM, is a digital SIM card that’s embedded directly into a device as software installed onto an Embedded Universal Integrated Circuit Card (eUICC) chip. – https://thehackernews.com/2025/07/esim-vulnerability-in-kigens-euicc.html

Beware the Robots: AI-Enabled Sanctions Evasion is Here

(Aaron Arnold – RUSI – 8 July 2025) This past January, the US Federal Bureau of Investigation issued updated guidance about North Korea’s IT labour networks, noting that they ‘…have been observed using artificial intelligence and face-swapping technology during video job interviews to obfuscate their true identities.’ Iran, too, is finding success in augmenting its global spear-phishing and social-engineering campaigns with AI. While banks and other financial institutions are increasingly supplementing their compliance and monitoring capabilities with AI, most are geared towards financial crime and sanctions evasion as it currently stands, not as it is heading. – https://www.rusi.org/explore-our-research/publications/commentary/beware-robots-ai-enabled-sanctions-evasion-here

Defence, Intelligence, and Warfare

BQP Raises a $5 Million Oversubscribed Seed Round Following Pilot Agreement with Air Force Research Lab for Quantum-Accelerated Digital Twin Platform

(Quantum Insider – 17 July 2025) BQP has raised an oversubscribed $4.9 million seed round to advance its quantum-accelerated digital twin platform, BQPhy®, targeting mission-critical applications in aerospace, defense, and semiconductor industries. The raise follows a Cooperative Research and Development Agreement (CRADA) with the U.S. Air Force Research Laboratory (AFRL/RQ) to enhance simulation capabilities using hybrid quantum-classical computing. BQPhy delivers current 10X performance gains with quantum-inspired solvers on CPU/GPU systems, with future quantum-native versions projected to achieve up to 1000X acceleration. – https://thequantuminsider.com/2025/07/17/bqp-raises-a-5-million-oversubscribed-seed-round-following-pilot-agreement-with-air-force-research-lab-for-quantum-accelerated-digital-twin-platform/

Britain brings together cyber and EW. Australia, take note

(Chris Sheahan – The Strategist – 17 July 2025) This September, Britain will rename Strategic Command as the Cyber & Specialist Operations Command (CSOC) following emphasis on the cyber and electromagnetic (CyberEM) domain within the Strategic Defence Review issued on 2 June, marking deep organisational and doctrinal transformation. While the change is high-level, and doesn’t reflect much restructuring, it is a significant expression of prioritisation and intent. This will hopefully see some level of replication within US and Australian military organisations creating more flexible and adaptable forces in light of the complexity of modern operations. The CSOC will house a range of critical defence functions and agencies, including: intelligence, surveillance and reconnaissance capabilities; Defence Intelligence; joint command and control for targeting; the Integrated Global Defence Network; Special Forces; Defence Medical Services; and the Defence Academy. –  https://www.aspistrategist.org.au/britain-brings-together-cyber-and-ew-australia-take-note/

For drone defence, Canberra should choose independent Australian companies

(Victor Abramowicz – The Strategist – 17 July 2025) A crucial decision will occur in the next few months that will shape Australia’s capability against small drones for decades: the selection of the systems integration partner (SIP) for Canberra’s Land 156 drone-defence project. The government should select genuinely domestic companies both for the SIP and the rest of Land 156 to reap a trifecta of capability, sovereign industry and export benefits. Defence’s urgent need for an improved capability to counter small drones is clear. The Australian Defence Force has scant protection against such weapons, and Ukraine’s successful use of them on 1 June in Operation Spiderweb against Russia has shown that a creative adversary can use such drones to strike anywhere. – https://www.aspistrategist.org.au/for-drone-defence-canberra-should-choose-independent-australian-companies/

Defence AI Centre at heart of Korean strategy

(DigWatch – 16 July 2025) South Korea has unveiled a strategy to share extensive military data with defence firms to accelerate AI-powered weapon systems, inspired by US military cloud initiatives. Plans include a national public–private fund to finance innovation and bolster the country’s defence tech prowess. A specialised working group of around 30 experts, including participants from the Defence Acquisition Program Administration, is drafting standards for safety and reliability in AI weapon systems. Their work aims to lay the foundations for the responsible integration of AI into defence hardware. – https://dig.watch/updates/defence-ai-centre-at-heart-of-korean-strategy

Australia’s Navy tests gravity-based navigation tech to counter GPS signal spoofing

(Interesting Engineering – 16 July 2025) Q-CTRL, a leading company in quantum technology, has announced a breakthrough in using quantum sensors for navigation. Their technology showed strong performance in a recent field trial with the Australian Defence Force on the Navy ship MV Sycamore. This success highlights Q-CTRL’s growing role in creating advanced navigation tools for use at sea. “Quantum sensors provide a near-term opportunity to achieve transformational defense capabilities, but previous deployments in the field have struggled to deliver defense-relevant performance,” said Q-CTRL CEO and Founder, Michael J. Biercuk. – https://interestingengineering.com/military/australias-navy-tests-gravity-based-navigation

Military AI and the void of accountability

(DigWatch – 15 July 2025) In her blog post ‘Military AI: Operational dangers and the regulatory void,’ Julia Williams warns that AI is reshaping the battlefield, shifting from human-controlled systems to highly autonomous technologies that make life-and-death decisions. From the United States’ Project Maven to Israel’s AI-powered targeting in Gaza and Ukraine’s semi-autonomous drones, military AI is no longer a futuristic concept but a present reality. While designed to improve precision and reduce risks, these systems carry hidden dangers—opaque ‘black box’ decisions, biases rooted in flawed data, and unpredictable behaviour in high-pressure situations. Operators either distrust AI or over-rely on it, sometimes without understanding how conclusions are reached, creating a new layer of risk in modern warfare. Bias remains a critical challenge. AI can inherit societal prejudices from the data it is trained on, misinterpret patterns through algorithmic flaws, or encourage automation bias, where humans trust AI outputs even when they shouldn’t. – https://dig.watch/updates/military-ai-and-the-void-of-accountability

Unnatural Disasters: The Next Front in Russia’s Hybrid War

(Matt Ince – RUSI – 14 July 2025) Once confined to science fiction, solar geoengineering is now moving into real-world experimentation, raising the risk of misuse by hostile actors. Also known as solar radiation management, this set of novel technologies – such as stratospheric aerosol injection and marine cloud brightening – aims to artificially slow the rise in global temperatures by reducing the amount of sunlight absorbed by the Earth’s surface. To date, their primary purpose has been to tackle the symptoms of accelerating climate change. However, these technologies also pose dual-use risks: alongside unintended environmental consequences, they could be exploited by powers seeking to tilt the geopolitical balance to cause climate-related disruption. As the UK government prioritises investment in solar geoengineering research and development, it has a crucial window to help shape international norms. Without adequate safeguards, these technologies could become tools of geopolitical coercion in the years ahead. – https://www.rusi.org/explore-our-research/publications/commentary/unnatural-disasters-next-front-russias-hybrid-war

Elon Musk’s xAI wins $200M Pentagon deal to bring Grok AI into federal agencies

(Interesting Engineering – 14 July 2025) Elon Musk’s artificial intelligence startup xAI has secured a $200 million ceiling contract from the U.S. Department of Defense. The company also announced its “Grok for Government” suite on Monday, opening the door for every federal agency to access its tools via the General Services Administration (GSA) schedule. The deal places xAI alongside OpenAI, Anthropic, and Google, all of whom landed identical contracts under the Pentagon’s Chief Digital and Artificial Intelligence Office (CDAO). – https://interestingengineering.com/culture/elon-musk-xai-grok-pentagon-contract

Frontiers

Quranium and Abatis Forge Strategic Alliance to Secure Blockchain Infrastructure at the Endpoint Layer

(Quantum Insider – 18 July 2025) Quranium has partnered with Swiss cybersecurity firm Abatis to integrate tamper-proof endpoint protection into its quantum-secure Layer 1 blockchain infrastructure. The partnership addresses a key vulnerability in blockchain ecosystems by embedding Abatis’s ultra-lightweight, prevention-first security technology into wallets, validator nodes, and transaction platforms. The integration ensures real-time, zero-compromise protection without performance loss, enhancing trust and resilience in Quranium’s high-speed, quantum-resistant Web3 environment. – https://thequantuminsider.com/2025/07/18/quranium-and-abatis-forge-strategic-alliance-to-secure-blockchain-infrastructure-at-the-endpoint-layer/

Method Enables Quantum Memory Systems to Retain More Data and Maintain Spin Orientation Developed by Hebrew University Researchers

(Quantum Insider – 18 July 2025) Researchers from Hebrew University and Cornell University have developed a new method to protect atomic spins from environmental noise, potentially advancing quantum sensors, navigation systems, and quantum information technologies. The team used laser light to synchronize atomic spin orientations in cesium vapor, achieving a ninefold increase in spin coherence time without requiring extreme cooling or complex magnetic tuning. The technique works in warm conditions and could enable more practical, robust, and precise quantum devices for use in medical imaging, space exploration, and GPS-independent navigation. – https://thequantuminsider.com/2025/07/18/method-enables-quantum-memory-systems-to-retain-more-data-and-maintain-spin-orientation-developed-by-hebrew-university-researchers/

UK’s most powerful AI supercomputer powers on to run 21 million trillion ops/sec

(Interesting Engineering – 18 July 2025) The UK’s most powerful publicly accessible AI supercomputer, Isambard-AI, has been officially switched on in Bristol. Named after the famous 19th-century engineer Isambard Kingdom Brunel, the £225 million machine marks a significant step in Britain’s ambition to lead in artificial intelligence research. Secretary of State for Science and Technology Peter Kyle activated the system during a ceremony on Thursday, calling it “the raw computational horsepower that will save lives, create jobs, and help us reach net-zero ambitions faster.” – https://interestingengineering.com/innovation/uks-most-powerful-supercomputer-goes-live

UK Launches £2 Billion Supercomputing and AI Plan to Boost Economy and Public Services

(AI Insider – 18 July 2025) The UK government has announced a $2.6 billion initiative to expand national supercomputing and AI infrastructure, aimed at boosting healthcare, energy, public services, and economic growth. Spearheaded by the Department of Science, Innovation and Technology and UK Research and Innovation (UKRI), the roadmap targets a 20-fold increase in AI research capacity and includes a £750 million national supercomputing center in Edinburgh. The plan includes over £59 million in investments to train digital professionals, bridge skills gaps, and promote business-research partnerships, aiming at positioning Britain as a global leader in AI and high-performance computing (HPC). – https://theaiinsider.tech/2025/07/18/uk-launches-2-billion-supercomputing-and-ai-plan-to-boost-economy-and-public-services/

Norwegian Remora Robotics Raises $16 Million for Autonomous Aquaculture Robot

(AI Insider – 18 July 2025) Remora Robotics has raised 164 million NOK to accelerate deployment of its autonomous net-cleaning and inspection robots for the aquaculture industry. The AI-powered robots improve fish welfare and biosecurity by reducing stress, mortality, and risks tied to traditional vessel-based net cleaning methods. Led by Hatch Blue’s Blue Revolution Fund and joined by existing investors like Grieg Kapital, the funding supports scaling production and expanding Remora’s platform, with a new AI-driven monitoring solution to debut at Aqua Nor in August. – https://theaiinsider.tech/2025/07/18/norwegian-remora-robotics-raises-16-million-for-autonomous-aquaculture-robot/

Faraday Future Secures $105M in Financing to Fund the Company’s Aggressive Growth Strategy, Launch of the FX Super One and Advancement of the Company’s Position in the AIEV Market

(AI Insider – 18 July 2025) Faraday Future raised $105 million, including $82 million in new commitments, to fund the launch and production ramp-up of its FX Super One EV, targeting the AI-driven, affordable EV market. The financing, led by global and Middle Eastern investors, offers more favorable terms than previous rounds and will support expansion of FF and FX brands and AI technologies. The company and key executives received Wells Notices from the SEC over alleged 2021 disclosure violations, potentially impacting future engagements with major investors and banks. – https://theaiinsider.tech/2025/07/18/faraday-future-secures-105m-in-financing-to-fund-the-companys-aggressive-growth-strategy-launch-of-the-fx-super-one-and-advancement-of-the-companys-position-in-the-aiev-market/

Eton Solutions Closes $58M Series C to Bring AI-Driven Wealth Management Technology to the World’s Leading Family Offices, PEs and Funds

(AI Insider – 18 July 2025) Eton Solutions closed a $58M Series C round led by Navis Capital Partners to scale its AtlasFive® platform, which manages over $1 trillion in assets for 800+ ultra-wealthy families. The company has quadrupled revenue in three years and developed over 400 AI use cases to streamline workflows across 130,000 entities and 205,000+ investments. ith a growing global footprint, Eton will use the funding to enhance AI capabilities, launch new products for private equity and fund managers, and serve UHNW clients with $25M+ in assets. – https://theaiinsider.tech/2025/07/18/eton-solutions-closes-58m-series-c-to-bring-ai-driven-wealth-management-technology-to-the-worlds-leading-family-offices-pes-and-funds/

FlyGuys Secures $13M Series A-1 Led by Kevin O’Leary’s Wonder Fund to Power Global Expansion and AI-Ready Reality Data Infrastructure

(AI Insider – 18 July 2025) FlyGuys secured $13 million in Series A-1 funding led by Kevin O’Leary’s Wonder Fund North Dakota to expand its AI-focused reality data platform globally. The company connects 16,000+ FAA-certified drone pilots with enterprises needing high-quality real-world data for AI applications, from infrastructure to agriculture. The funding will boost international expansion, enhance its mission management software, and grow partnerships across industries and use cases. – https://theaiinsider.tech/2025/07/18/flyguys-secures-13m-series-a-1-led-by-kevin-olearys-wonder-fund-to-power-global-expansion-and-ai-ready-reality-data-infrastructure/

OpenAI: GPT-5 is coming, “we’ll see” if it creates a shockwave

(Bleeping Computer – 18 July 2025) OpenAI’s next foundational and state-of-the-art model, GPT-5, is still on its way after a delay. OpenAI won’t tell us the release date for now. In a conversation with a user on X, OpenAI’s researcher Xikun Zhand confirmed that GPT-5 is still coming. When asked if GPT-5 will be another shockwave for the AI industry, Zhand responded with “we will see” and a wink emoji, which seems to suggest that it could be a really significant update. – https://www.bleepingcomputer.com/news/artificial-intelligence/openai-gpt-5-is-coming-well-see-if-it-creates-a-shockwave/

Matlantis Announces Major Upgrade to Its Universal Atomistic Simulator for Materials Discovery, Opens Dedicated U.S. Office

(AI Insider – 18 July 2025) Matlantis Inc., the U.S. arm of Japan’s Preferred Networks (PFN), has launched Version 8 of its AI-powered Matlantis™ universal atomistic simulator, enhancing materials research accuracy and speed through advanced machine learning techniques. The upgraded platform uses PFP (Preferred Potential) Version 8, the first universal machine learning interatomic potential (MLIP) trained with r2SCAN datasets, enabling up to double the simulation accuracy compared to previous PBE-trained models while maintaining rapid simulation speeds. Matlantis’ new U.S. office in Cambridge, Massachusetts, aims to accelerate adoption across North America, providing cloud-based AI simulation tools that streamline materials discovery in industries such as batteries, semiconductors, and catalysts. – https://theaiinsider.tech/2025/07/18/matlantis-announces-major-upgrade-to-its-universal-atomistic-simulator-for-materials-discovery-opens-dedicated-u-s-office/

Mistral Launches Voxtral, Its First Open Audio AI Model for Businesses

(AI Insider – 18 July 2025) French AI startup Mistral has entered the speech AI arena with the launch of Voxtral, a new family of open-weight audio models designed to bring advanced speech intelligence to businesses at a fraction of the cost of closed systems. The release marks Mistral’s latest push to challenge proprietary AI incumbents by offering developers greater affordability and control. – https://theaiinsider.tech/2025/07/18/mistral-launches-voxtral-its-first-open-audio-ai-model-for-businesses/

Generative AI now powers 20% of new Steam games

(DigWatch – 18 July 2025) Nearly 20 percent of video games released on Steam in 2025 include generative AI, according to a new report by Totally Human Media. The report, based on data gathered from Steam, states that around 7,818 games currently disclose using generative AI. The figure represents roughly 7 percent of the platform’s entire catalogue. For games launched in 2025, nearly one in five incorporate AI tools or openly disclose doing so. Compared to 2024, this marks a nearly 700 percent increase in generative AI adoption, reflecting a broader industry trend towards automation and machine-generated content. – https://dig.watch/updates/generative-ai-now-powers-20-of-new-steam-games

Elephant Robot Demonstrates Bioinspired 3D Printing Technology

(AI Insider – 17 July 2025) Researchers at EPFL have developed a 3D-printable, programmable lattice structure made from a single foam material that mimics biological tissues, enabling robots to combine soft and rigid parts within a single design. The lattice uses customizable unit cells that can assume over one million configurations, allowing continuous modulation of stiffness and load-bearing properties, demonstrated in a musculoskeletal-inspired elephant robot combining soft trunk motion and rigid joint control. Published in Science Advances, the technology is scalable, lightweight, waterproof, and sensor-compatible, offering potential applications in adaptable, efficient robotics across fluid environments and precision tasks. – https://theaiinsider.tech/2025/07/17/elephant-robot-demonstrates-bioinspired-3d-printing-technology/

Bedrock Robotics Emerges from Stealth with $80M in Funding for Autonomous Construction Technology

(AI Insider – 17 July 2025) Bedrock Robotics emerged from stealth with $80 million in Seed and Series A funding to develop autonomous systems that retrofit existing construction equipment for fully autonomous operation, targeting the $13 trillion global construction market. Founded by former Waymo and Segment leaders, the company focuses on upgrading heavy machinery fleets with reversible, same-day hardware and software installations to address U.S. infrastructure demands and a critical shortage of construction workers. Bedrock’s autonomous systems are deployed on active construction sites in Arizona, Texas, and Arkansas, with plans to scale operations and partnerships toward initial operator-less deployments by 2026, supported by investors including 8VC, Eclipse, NVIDIA’s NVentures, and Two Sigma Ventures. –  https://theaiinsider.tech/2025/07/17/bedrock-robotics-emerges-from-stealth-with-80m-in-funding-for-autonomous-construction-technology/

Chinese Robotics Firm AiMOGA Showcases Deployment of its Car Dealership Humanoid Robot

(AI Insider – 17 July 2025) Chinese robotics firm AiMOGA is expanding globally after showcasing its humanoid robot retail solutions to over 3,000 automotive dealers at its first Partnership Conference in Beijing, held in collaboration with Chery Auto Group. AiMOGA’s humanoid robots, already deployed at OMODA&JAECOO dealerships in Malaysia and Hong Kong, support showroom operations and are credited with improving lead conversion rates and reducing front-desk workloads. Leveraging Chery’s autonomous systems and AiMOGA’s proprietary CheryGPT platform, the company plans broader applications across retail, service hubs, and residential sectors, positioning itself within a projected €50 billion humanoid robotics market by 2035. – https://theaiinsider.tech/2025/07/17/chinese-robotics-firm-aimoga-showcases-deployment-of-its-car-dealership-humanoid-robot/

Rwazi Announces $12M Funding Round to Replace Every Gut Call with an AI Copilot

(AI Insider – 17 July 2025) Rwazi has secured $12 million in Series A funding to advance its AI copilot and simulation engine, offering real-time decision intelligence based on live consumer activity. Unlike AI built on passive data, Rwazi analyzes billions of direct behavioral signals to help enterprises detect market shifts, simulate outcomes, and act with speed and precision. Already used by Fortune 100 companies, the funding will expand Rwazi’s global data infrastructure and enhance its ability to guide strategic decisions across marketing, product, and operations. – https://theaiinsider.tech/2025/07/17/rwazi-announces-12m-funding-round-to-replace-every-gut-call-with-an-ai-copilot/

OpenEvidence Secures $210M Series B at $3.5B Valuation to Expand AI Medical Platform for Physicians

(AI Insider – 17 July 2025) OpenEvidence, the AI-powered medical search engine built exclusively for physicians, has raised $210 million in a Series B funding round, bringing its valuation to $3.5 billion. The round was co-led by Google Ventures and Kleiner Perkins, with continued support from Sequoia Capital and participation from Coatue, Conviction, Thrive, and others. Founded by Daniel Nadler, Ph.D., OpenEvidence has quickly become one of the most widely used medical platforms in the U.S., with more than 40% of clinicians using the tool across over 10,000 hospitals and medical centers. The platform offers instant, evidence-based clinical answers powered by proprietary generative AI trained on peer-reviewed content from trusted partners including the AMA, JAMA, and the New England Journal of Medicine. – https://theaiinsider.tech/2025/07/17/openevidence-secures-210m-series-b-at-3-5b-valuation-to-expand-ai-medical-platform-for-physicians/

Qubitcore, an OIST Spin‑Out, Raises Pre‑Seed Funding to Advance Japan’s Ion‑Trap Quantum Computers

(Quantum Insider – 17 July 2025) Qubitcore Inc. closed its pre-seed funding round led by the OIST Lifetime Ventures Fund and signed an exclusive IP licensing agreement with the Okinawa Institute of Science and Technology (OIST). The company is developing a distributed quantum computing architecture using OIST-developed ion-trap modules with integrated optical cavities and photonic links to enable scalable, fault-tolerant quantum computers. Funding will support hardware prototyping, team expansion, and a multi-phase roadmap that targets a 1,000-qubit-class prototype by 2029 and commercial deployment by 2030. – https://thequantuminsider.com/2025/07/17/qubitcore-an-oist-spin%e2%80%91out-raises-pre%e2%80%91seed-funding-to-advance-japans-ion%e2%80%91trap-quantum-computers/

Oxford Ionics And Iceberg Quantum Partner to Accelerate Fault-Tolerant Quantum Computing

(Quantum Insider – 17 July 2025) Oxford Ionics has partnered with Iceberg Quantum to integrate advanced quantum error correction using qLDPC codes into its trapped-ion hardware, aiming to accelerate the path to fault-tolerant quantum computing. Unlike traditional surface codes that require high hardware overhead, qLDPC codes provide strong error protection with fewer physical qubits, leveraging a sparse, non-local structure ideal for real-time quantum operations. Oxford Ionics’ trapped-ion platform—with record-setting gate fidelities and long-range connectivity—is particularly suited to implementing qLDPC architectures developed by Iceberg Quantum. – https://thequantuminsider.com/2025/07/17/oxford-ionics-and-iceberg-quantum-partner-to-accelerate-fault-tolerant-quantum-computing/

HRL Laboratories Launches Open-Source Solution for Solid-State Spin Qubits

(Quantum Insider – 17 July 2025) HRL Laboratories has released spinQICK, an open-source extension of the Quantum Instrumentation Control Kit (QICK), providing a low-cost, FPGA-based solution for controlling semiconductor spin-qubits. The platform builds on Fermilab’s QICK toolkit and offers firmware and software tailored to spin-qubit research, supporting techniques like single-spin control, two-qubit gates, and electrostatic tune-up. Designed to accelerate research adoption and workforce development, spinQICK enables academic and industry users to deploy standardized spin-qubit control methods using affordable commercial hardware. – https://thequantuminsider.com/2025/07/17/hrl-laboratories-launches-open-source-solution-for-solid-state-spin-qubits/

Microsoft and Atom Computing Partner on Level 2 Quantum System for Nordic Users

(Quantum Insider – 17 July 2025) Denmark’s EIFO and the Novo Nordisk Foundation are investing €80 million to establish QuNorth and acquire Magne, a powerful commercial quantum computer based on logical qubits. The system will be built by Atom Computing and Microsoft, combining neutral atom hardware with advanced quantum software, and is expected to be operational by early 2027. QuNorth will be based in Copenhagen and aims to provide Nordic researchers and industries with priority access to Level 2 quantum computing for applications in chemistry, materials science, and beyond. – https://thequantuminsider.com/2025/07/17/microsoft-and-atom-computing-partner-on-level-2-quantum-system-for-nordic-users/

Quobly and Inria Partner on Integrated, Scalable Silicon Qubit Architecture

(Quantum Insider – 17 July 2025) Quobly and Inria have formed a strategic partnership to co-develop a fully integrated, scalable quantum computing architecture based on silicon qubits and advanced control software. The collaboration focuses on middleware and error correction protocols tailored to silicon-based quantum hardware, aiming to bridge the gap between physical qubits and quantum algorithms. This partnership supports France’s national Q-Loop program and aligns with Quobly’s industrial roadmap, including its recent launch of a quantum emulator deployed on OVHcloud. – https://thequantuminsider.com/2025/07/17/quobly-and-inria-partner-on-integrated-scalable-silicon-qubit-architecture/

Unify Raises $40M Series B to Transform Go-To-Market with AI

(AI Insider – 17 July 2025) Unify has closed a $40 million Series B round led by Battery Ventures, with backing from OpenAI Startup Fund and Thrive Capital, just nine months after raising $12 million. The platform uses AI agents to surface high-intent leads, personalize outreach at scale, and automate sales workflows — powering pipeline growth for clients like Airwallex and Perplexity. With 8x revenue growth in the past year, Unify is positioning itself as a next-gen GTM solution transforming outbound sales into a repeatable, scalable science. – https://theaiinsider.tech/2025/07/17/unify-raises-40m-series-b-to-transform-go-to-market-with-ai/

Boffins detail new algorithms to losslessly boost AI perf by up to 2.8x

(The Register – 17 July 2025) We all know that AI is expensive, but a new set of algorithms developed by researchers at the Weizmann Institute of Science, Intel Labs, and d-Matrix could significantly reduce the cost of serving up your favorite large language model (LLM) with just a few lines of code. Presented at the International Conference on Machine Learning this week and detailed in this paper, the algorithms offer a new spin on speculative decoding that they say can boost token generation rates by as much as 2.8x while also eliminating the need for specialized draft models. Speculative decoding, if you’re not familiar, isn’t a new concept. It works by using a small “draft” model (“drafter” for short) to predict the outputs of larger, slower, but higher quality “target” models. – https://www.theregister.com/2025/07/17/new_algorithms_boost_ai_perf/

Robots can now feel difference between pat and punch with new artificial nerves

(Interesting Engineering – 17 July 2025) KAIST has unveiled a neuromorphic semiconductor-based artificial sensory nervous system that lets robots ignore safe, familiar stimuli and react quickly to dangerous ones, much like living organisms. The technology aims to support intelligent, energy‑efficient responses in applications ranging from ultra‑small robots to robotic prosthetics. Animals and humans conserve energy by disregarding safe or familiar cues while staying alert to important or harmful signals. The steady sound of an air conditioner or the feel of clothing on skin soon fades from notice, but hearing your name or being touched by something sharp triggers rapid focus. – https://interestingengineering.com/innovation/robots-feel-difference-between-pat-and-punch

AI’s Inner Monologue Offers New Clues for Safety—But the Window May Be Closing

(AI Insider – 17 July 2025) A new study from researchers at OpenAI, DeepMind, Anthropic, and others finds that monitoring an AI system’s internal chain-of-thought (CoT) reasoning offers a promising but fragile opportunity for AI safety. CoT monitoring can reveal misalignment or harmful intent in intermediate reasoning steps, especially on complex tasks that require models to “think out loud.”. The researchers warn that future model architectures, training methods, and incentives could reduce or eliminate this visibility, and call for proactive efforts to preserve monitorability. – https://theaiinsider.tech/2025/07/17/ais-inner-monologue-offers-new-clues-for-safety-but-the-window-may-be-closing/

South Korean firm unveils faster AI data centre architecture with CXL-over-Xlink

(DigWatch – 17 July 2025) South Korean company Panmnesia has introduced a new architecture for AI data centres aimed at improving speed and efficiency. Instead of using only PCIe or RDMA-based systems, its CXL-over-Xlink approach combines Compute Express Link (CXL) with fast accelerator links such as UALink and NVLink. The company claims this design can deliver up to 5.3 times faster AI training and reduce inference latency sixfold. By allowing CPUs and GPUs to access large shared memory pools via the CXL fabric, AI workloads are no longer restricted by the fixed memory limits inside each GPU. – https://dig.watch/updates/south-korean-firm-unveils-faster-ai-data-centre-architecture-with-cxl-over-xlink

AI system screens diabetic eye disease with near-perfect accuracy

(DigWatch – 17 July 2025) A new AI programme is showing remarkable accuracy in detecting diabetic retinopathy, a leading cause of preventable blindness. The SMART system, short for Simple Mobile AI Retina Tracker, can scan retinal images using even basic smartphones and has achieved over 99% accuracy. Researchers at the University of Texas Health Sciences Center in the US trained the AI using thousands of retinal images from diverse populations across six continents. The system processes images in under a second and can distinguish diabetic retinopathy from other eye diseases. – https://dig.watch/updates/ai-system-screens-diabetic-eye-disease-with-near-perfect-accuracy

AI tool uses walking patterns to detect early signs of dementia

(DigWatch – 17 July 2025) Fujitsu and Acer Medical are trialling an AI-powered tool to help identify early signs of dementia and Parkinson’s disease by analysing patients’ walking patterns. The system, called aiGait and powered by Fujitsu’s Uvance skeleton recognition technology, converts routine movements into health data. Initial tests are taking place at a daycare centre linked to Taipei Veterans Hospital, using tablets and smartphones to record basic patient movements. The AI compares this footage with known movement patterns associated with neurodegenerative conditions, helping caregivers detect subtle abnormalities. – https://dig.watch/updates/ai-tool-uses-walking-patterns-to-detect-early-signs-of-dementia

AI blood test for breast cancer enters Australian clinics

(DigWatch – 17 July 2025) Scientists at UNSW Sydney are developing Australia’s first commercial lipid-based blood test for breast cancer, powered by AI. The tool is already in clinical use and promises non-invasive detection of early-stage tumours through biomarkers in the bloodstream. Led by Associate Professor Fatemeh Vafaee, the Vafaee Lab is also exploring blood-based detection for cancers of the lung, liver and brain. ‘The current detection methods rely heavily on imaging and tissue biopsies, which often miss the full tumour picture,’ says Vafaee. She explains that tumours consist of many cell types, so biopsy samples may not reflect their complete complexity. AI-driven blood tests offer a less invasive way to detect cancer early — potentially before symptoms or imaging can identify it. – https://dig.watch/updates/ai-blood-test-for-breast-cancer-enters-australian-clinics

H2Ok Innovations Raises $12M Series A for AI-Powered Sensors That Disrupt and Transform CPG Manufacturing

(AI Insider – 17 July 2025) H2Ok Innovations raised $12.42M in Series A funding led by Greycroft to scale its patented inline sensor and AI platform that helps manufacturers optimize production and reduce waste. Trusted by companies like AB InBev, Coca-Cola, and Unilever, the system enables faster Clean-in-Place (CIP) and product changeovers, cutting downtime and utility usage by up to 20%. With deployments across six continents, the startup aims to transform industrial operations by combining real-time monitoring, AI optimization, and sustainable manufacturing practices. – https://theaiinsider.tech/2025/07/17/h2ok-innovations-raises-12m-series-a-for-ai-powered-sensors-that-disrupt-and-transform-cpg-manufacturing/

Callidus Legal AI Closes $10M in Funding to Build the AI Operating System for Litigators

(AI Insider – 17 July 2025) Callidus Legal AI has raised $10 million in an oversubscribed round led by Cervin Ventures, bringing total funding to $13 million to scale its end-to-end AI platform for litigators. The platform automates legal research and drafting using proprietary agentic AI and a 10M+ case law database, reducing week-long tasks to minutes while keeping lawyers in control. With tripled revenue in early 2025 and plans to expand its team and features, Callidus is rapidly emerging as a transformative force in legal tech. – https://theaiinsider.tech/2025/07/17/callidus-legal-ai-closes-10m-in-funding-to-build-the-ai-operating-system-for-litigators/

Starlink says SpaceX targeting 2026 for launch of Starship-ready terabit satellites

(The Register – 16 July 2025) Elon Musk’s space broadband service Starlink has hinted that Elon Musk’s Starship will be ready for commercial flights in 2026. Starlink on Tuesday posted a network update in which it discussed its third-generation satellites, each of which can provide “over a terabit per second of downlink capacity and over 200 Gbps of uplink capacity to customers on the ground.”. The spec of the third-gen sats has been public knowledge for months. The network update adds a useful nugget of info by stating “SpaceX is targeting to begin launching its third-generation satellites in the first half of 2026.” – https://www.theregister.com/2025/07/16/starlink_network_update/

How AI Can Degrade Human Performance in High-Stakes Settings

(Dane A. Morey, Mike Rayo, and David Woods – AI Frontiers – 16 July 2025) Last week, the AI nonprofit METR published an in-depth study on human-AI collaboration that stunned experts. It found that software developers with access to AI tools took 19% longer to complete their tasks, despite believing they had finished 20% faster. The findings shed important light on our ability to predict how AI capabilities interact with human skills. Since 2020, we have been conducting similar studies on human-AI collaboration, but in contexts with much higher stakes than software development. Alarmingly, in these safety-critical settings, we found that access to AI tools can cause humans to perform much, much worse. A 19% slowdown in software development can eat into profits. Reduced performance in safety-critical settings can cost lives. – https://aifrontiersmedia.substack.com/p/how-ai-can-degrade-human-performance

Google and Westinghouse unleash AI to build nuclear reactors faster than ever

(Interesting Engineering – 15 July 2025) In a first-of-its-kind move, Westinghouse Electric Company and Google Cloud have teamed up to leverage artificial intelligence for streamlining nuclear reactor construction. Their AI-powered tools autonomously generate and optimize modular work packages for advanced reactors. The collaboration pairs Westinghouse’s proprietary HiVE™ and bertha™ nuclear AI solutions with Google Cloud technologies such as Vertex AI, Gemini, and BigQuery. – https://interestingengineering.com/culture/westinghouse-google-cloud-ai-nuclear-reactors

US AI supercomputer Nexus will compute faster than 8 billion humans combined

(Interesting Engineering – 15 July 2025) The U.S. research community is set to gain a major AI-powered boost. Georgia Tech and its partners have secured $20 million from the National Science Foundation to build Nexus, one of the nation’s fastest supercomputers, built to accelerate scientific discovery using artificial intelligence. Once completed in spring 2026, Nexus will deliver over 400 quadrillion operations per second, the equivalent of everyone in the world continuously performing 50 million calculations every second. – https://interestingengineering.com/innovation/georgia-tech-nexus-supercomputer-ai-research

German humanoid robot welder to tackle high-risk jobs at Hyundai’s shipyard

(Interesting Engineering – 15 July 2025) A partnership will result in Korean firms HD Hyundai Robotics and HD Hyundai Samho testing advanced robots in shipbuilding, marking a major step in automating one of the world’s toughest industries. Interestingly, the trial will use robots from Germany’s Neura Robotics, not Hyundai’s. Despite owning Boston Dynamics and a leading automation arm, Hyundai is turning to external innovation for this initiative. – https://interestingengineering.com/innovation/humanoid-robot-welder-hyundai-shipyard

Germany creates material ‘that has never existed’ to unlock quantum tech power

(Interesting Engineering – 15 July 2025) Scientists have merged four elements from Group IV of the periodic table to design a new material that could redefine the future of quantum computing, microelectronics, and photonics. The stable semiconductor alloy of carbon (C), silicon (Si), germanium (Ge), and tin (Sn) was developed by researchers at Forschungszentrum Jülich (FZJ), one of the largest interdisciplinary research institutions in Europe, and the Leibniz Institute for Innovative Microelectronics (IHP). – https://interestingengineering.com/innovation/material-to-unlock-quantum-tech-power

World’s first quantum light factory chip built on standard semiconductor tech

(Interesting Engineering – 14 July 2025) In a leap toward practical quantum systems, researchers from Boston University, UC Berkeley, and Northwestern University have built the world’s first integrated electronic–photonic–quantum chip. The study showcases a device that merges quantum light sources with stabilizing electronics on a single platform using a standard 45-nanometer semiconductor process. The chip can produce streams of correlated photon pairs, particles of light crucial for future quantum computing, sensing, and secure communication. – https://interestingengineering.com/innovation/worlds-first-quantum-light-factory-chip

New SSD can self-destruct on command to protect your most sensitive data

(Interesting Engineering – 14 July 2025) “This message will self-destruct in 3… 2… 1” is something you’ve definitely seen in Mission Impossible films over the years. Now, we finally have tech that feels just as futuristic, thanks to a new kind of storage hardware. Taiwanese company TeamGroup has unveiled a new internal SSD that can literally destroy itself at the press of a button, ensuring sensitive data never falls into the wrong hands. – https://interestingengineering.com/innovation/self-destructing-ssd-teamgroup-p250q

Meta reveals plans for giant AI superclusters to outpace OpenAI and Google

(Interesting Engineering – 14 July 2025) Meta Platforms Inc., the technology conglomerate behind Facebook, is going all-in on artificial intelligence infrastructure. CEO Mark Zuckerberg on Monday announced plans to invest “hundreds of billions of dollars” into building some of the world’s largest AI superclusters. In a Facebook post, the tech mogul revealed that Meta’s first supercluster, named Prometheus, is set to go live next year, with several other multi-gigawatt clusters also in development. – https://interestingengineering.com/innovation/meta-ai-superclusters-prometheus-hyperion

Powerful polymer could fuel safer EVs, smartphones, drones and space probes

(Interesting Engineering – 14 July 2025) Scientists in the US have blended a common polymer with a specially charged one, in a breakthrough that could potentially lead to safer, longer-lasting solid-state batteries for smartphones, EVs, drones, and space probes. The FAMU-FSU College of Engineering research team at Florida State University used polyethylene oxide (PEO), a polymer widely used in lithium-ion (Li-ion) batteries for its ionic conductivity and mechanical strength. They then combined it with a custom-designed charged polymer known as p5. – https://interestingengineering.com/innovation/powerful-polymer-could-fuel-safer-evs

NASA tests supersonic muscle in Japan as mini X-59 jet hits 925 mph in Tokyo tunnel

(Interesting Engineering – 14 July 2025) Researchers from NASA and the Japanese Aerospace Exploration Agency (JAXA) have completed a fresh round of supersonic wind-tunnel testing on a miniature version of NASA’s X-59 quiet-supersonic experimental aircraft. The campaign was conducted at JAXA’s 3-by-3-foot facility in Chofu, Tokyo. It assessed how the aircraft’s pressure signature, audible on the ground as a sonic “thump,” will reach people beneath its flight path. – https://interestingengineering.com/innovation/nasa-jaxa-test-19-inch-x-59-model

How AI agents are reshaping the marketing landscape

(DigWatch – 14 July 2025) Marketers have weathered many disruptions, but a bigger shift is emerging—AI agents are starting to make purchasing decisions. As machines begin choosing what to buy, brands must rethink how they build visibility and relevance in this new landscape. AI agents do not shop like humans. They use logic, structured data, and performance signals—not emotion, nostalgia or storytelling. They compare price, reviews, and specs. Brand loyalty and lifestyle marketing may carry less weight when decisions are made algorithmically. According to Salesforce, 24% of people are open to AI shopping on their behalf—rising to 32% among Gen Z. Agents interpret products as data tables. Structured information, such as features and sentiment analysis, guide choices—not impulse or advertising flair – https://dig.watch/updates/how-ai-agents-are-reshaping-the-marketing-landscape

Stanford study flags dangers of using AI as mental health therapists

(DigWatch – 14 July 2025) A new Stanford University study warns that therapy chatbots powered by large language models (LLMs) may pose serious user risks, including reinforcing harmful stigmas and offering unsafe responses. Presented at the upcoming ACM Conference on Fairness, Accountability, and Transparency, the study analysed five popular AI chatbots marketed for therapeutic support, evaluating them against core guidelines for assessing human therapists. The research team conducted two experiments, one to detect bias and stigma, and another to assess how chatbots respond to real-world mental health issues. Findings revealed that bots were more likely to stigmatise people with conditions like schizophrenia and alcohol dependence compared to those with depression. – https://dig.watch/updates/stanford-study-flags-dangers-of-using-ai-as-mental-health-therapists

AI could save billions but healthcare adoption is slow

(DigWatch – 14 July 2025) AI is being hailed as a transformative force in healthcare, with the potential to reduce costs and improve outcomes dramatically. Estimates suggest widespread AI integration could save up to 360 billion dollars annually by accelerating diagnosis and reducing inefficiencies across the system. Although tools like AI scribes, triage assistants, and scheduling systems are gaining ground, clinical adoption remains slow. Only a small percentage of doctors, roughly 12%, currently rely on AI for diagnostic decisions. This cautious rollout reflects deeper concerns about the risks associated with medical AI. – https://dig.watch/updates/ai-could-save-billions-but-healthcare-adoption-is-slow

Canada firm’s 99.9% reliable robot vision training studio completes week-long task in hours

(Interesting Engineering – 13 July 2025) A Canadian company has advanced its web-based design and AI training studio that simplifies 4D vision-guided robotic projects. Called Apera Forge, the system, without any hardware, is a browser-based AI vision design studio. The system’s latest features support advanced robotic cell design, end-of-arm-tooling (EOAT)-mounted camera configurations, and full simulation and training for de-racking applications. The design studio can streamline 4D vision-guided robotics automation projects. – https://interestingengineering.com/innovation/training-studio-completes-robotic-vision-tasks

Metaverse for nuclear plants: Hitachi’s platform to boost efficiency in design, ops

(Interesting Engineering – 13 July 2025) A Japanese company has launched a “Metaverse Platform for Nuclear Power Plants.” Hitachi plans to use the platform to streamline, operations, including nuclear power plants’ safety enhancement and maintenance. The latest approach recreates nuclear power plants in a metaverse using high-precision point cloud data and 3D CAD data. The system that leverages a metaverse and AI technology to enhance productivity in information sharing, schedule coordination, and asset management among stakeholders by utilizing it with partners such as electric utilities and contractors. – https://interestingengineering.com/energy/metaverse-for-nuclear-power-plants-hitachi

Coinbase teams with Perplexity to bring crypto data into AI search

(DigWatch – 13 July 2025) Coinbase has partnered with Perplexity AI to deliver real-time crypto market data through AI-powered tools. The integration will begin with Coinbase’s COIN50 index on Perplexity’s Comet browser, followed by a deeper rollout enabling direct access to token insights and trade ideas through natural language queries. CEO Brian Armstrong described the collaboration as a step towards a more intelligent and decentralised financial system. He said the fusion of AI and crypto could be a ’10x unlock,’ adding that the future lies in integrating crypto wallets into large language models (LLMs). – https://dig.watch/updates/coinbase-teams-with-perplexity-to-bring-crypto-data-into-ai-search

Hybrid jobs: How AI is rewriting work in finance

(Eduardo Levy Yeyati – Brookings – 10  July 2025) Artificial intelligence (AI) is not destroying jobs in finance, it is rewriting them. As models begin to handle underwriting, compliance, and asset allocation, the traditional architecture of financial work is undergoing a fundamental shift. This is not about coders replacing bankers. It is about a sector where knowing how the model works—what it sees and how it reasons—becomes the difference between making and automating decisions. It is also about the decline of traditional credentials and the rise of practical experience and critical judgement as key assets in a narrowing workforce. In what follows, we explore how the rise of generative AI and autonomous systems is reshaping the financial workforce: Which roles are fading, which ones are emerging, and how institutions—and policymakers—can bridge the looming talent divide. – https://www.brookings.edu/articles/hybrid-jobs-how-ai-is-rewriting-work-in-finance/