Governance

UN Reaches Consensus on AI. Now Comes the Hard Part

(Vidisha Mishra, Nicole Manger – Tech Policy Press – 11 September 2025) On August 26, the UN General Assembly adopted Resolution A/RES/79/325 (the AI Modalities Resolution) by consensus, without a vote, following months of negotiations. At a time when consensus is elusive and the multilateral system is under constraint, this follow-up to the Global Digital Compact (GDC) marks a timely and substantive step towards global AI governance. This challenge transcends borders and demands collective action. The resolution creates two new institutional mechanisms that stand out for their ambition and design. The “Independent International Scientific Panel on AI” will comprise forty experts serving in their personal capacity, with appointments balanced by gender and geography. Members will serve three-year terms, disclose conflicts of interest, and elect co-chairs from both developed and developing countries. The Panel is tasked with producing an annual, evidence-based assessment synthesizing existing research, accompanied by thematic briefs as needed. These reports will be presented not only to the General Assembly but also to the Global Dialogue on AI Governance, anchoring political deliberations in independent science. The “Global Dialogue on AI Governance” is the second mechanism. It is significant as the first truly global and inclusive platform, bringing more than 100 countries – particularly from the Global South and LDCs – onto equal footing in shaping AI governance. These countries have largely been excluded from existing processes led by the G7, G20, Council of Europe, EU, AU, and OECD. By design, the Dialogue broadens participation, deepens North–South exchange, and helps close digital divides. – https://www.techpolicy.press/un-reaches-consensus-on-ai-now-comes-the-hard-part/

UN Launches AI Panel and Dialogue, But Questions Linger Over Inclusion and Impact

(Amaj Rahimi-Midani – Tech Policy Press – 11 September 2025) On August 26, after months of difficult negotiations, the United Nations General Assembly achieved a historic milestone: all 193 Member States reached consensus on one of the most consequential questions of our time — how should artificial intelligence be governed. Under the co-facilitation of Costa Rica and Spain, governments agreed to establish an Independent International Scientific Panel on AI and to launch a Global Dialogue on AI Governance. This resolution is more than just another UN document — it marks the first collective attempt to place AI governance within the world’s most universal forum, where every country has a voice. Based on recommendations made nearly a year earlier by the UN High-Level Advisory Body on AI, which called for creating a scientific panel and a global policy dialogue, alongside proposals for a global AI fund to reduce inequalities between developed and developing nations, the process was designed to engage a broader international process, involving a range of stakeholders beyond powerful states or technology companies. – https://www.techpolicy.press/un-launches-ai-panel-and-dialogue-but-questions-linger-over-inclusion-and-impact/

Egypt launches AI readiness report with EU support

(DigWatch – 12 September 2025) Egypt has released its first AI Readiness Assessment Report, developed by the Ministry of Communications and Information Technology with UNESCO Cairo and supported by the EU funding. The report reviews Egypt’s legal, policy, institutional and technical environment, highlighting the strengths and gaps in the country’s digital transformation journey. It emphasises ensuring that AI development is human-centred and responsibly governed. – https://dig.watch/updates/egypt-launches-ai-readiness-report-with-eu-support – https://south.euneighbours.eu/news/eu-and-egypt-launch-ai-readiness-report-to-foster-responsible-and-inclusive-digital-transformation/

New AI and digital ministry in Kazakhstan to drive national transformation

(DigWatch – 12  September 2025) Kazakhstan is stepping up its digital ambitions with President Kassym-Jomart Tokayev announcing the creation of a Ministry of Artificial Intelligence and Digital Development. Speaking during his annual state-of-the-nation address, Tokayev said the move is part of the country’s goal to become a fully digital nation within three years. The new ministry will lead efforts to modernise the economy through AI, big data, and platform technologies, guided by a forthcoming national strategy called Digital Kazakhstan. Tokayev stressed that digital transformation is not optional but essential for Kazakhstan, which he described as being at the crossroads of global technological shifts. – https://dig.watch/updates/new-ai-and-digital-ministry-in-kazakhstan-to-drive-national-transformation – https://astanatimes.com/2025/09/kazakhstan-to-establish-ministry-of-artificial-intelligence-and-digital-development/

Japan plans first national AI strategy

(DigWatch – 12 September 2025) Japan is preparing its first national AI basic plan to boost AI adoption in public institutions and beyond. The draft sets out four core policies to balance innovation with risk management, with final Cabinet approval expected later this year. The plan targets low AI usage rates in the country, around 20% for individuals and 50% for corporations. Policies include accelerating AI adoption, strengthening development capacity, leading in AI governance, and fostering continuous social transformation toward an AI-integrated society. – https://dig.watch/updates/japan-plans-first-national-ai-strategy – https://asianews.network/japanese-governments-ai-basic-plan-to-promote-use-of-ai-in-public-institutions/

Lessons from Nepal on the High Cost of Controlling Online Expression

(Samik Kharel – Tech Policy Press – 12 September 2025) In a matter of days, Nepal has gone from a fragile democracy to a digital battleground where the future of the nation is being shaped not in parliament, but on Discord servers. The government’s blanket ban on 26 major social media platforms, including Facebook, Instagram, WhatsApp, YouTube, and Discord, set off a chain of events that have now left the country without a government. At least 51 people are dead, hundreds are injured, and government buildings in Kathmandu lie in ruins after protests led by Gen Z escalated into mass unrest. What began as a fight for digital freedom quickly snowballed into a political reckoning. Amid the chaos, a new center of power emerged from an unlikely place: Discord. – https://www.techpolicy.press/lessons-from-nepal-on-the-high-cost-of-controlling-online-expression/

Amidst Boom, Data Center Lobby Expands Its Influence, Spending, and Tactics

(Justin Hendrix – Tech Policy Press – 12 September 2025) Demand for computing power is fueling a massive surge in investment in data centers worldwide. McKinsey, a consultancy, estimates investment in data centers will reach $6.7 trillion by 2030. Recent industry estimates suggest that more than $1 trillion will be invested in the United States over the next five years. The scale of capital expenditures is evidenced by projects such as the massive Stargate initiative announced by President Donald Trump at the White House alongside executives from OpenAI, Oracle, and Softbank the day after his inauguration, and subsequent multibillion-dollar announcements by tech firms including Microsoft, Amazon, Meta, and Google in recent months. As the boom expands, public scrutiny of industry plans and promises is increasing across the country. Concerns over environmental impacts, demands on energy infrastructure, and other economic and social justice considerations are increasingly the subject of contentious debate at the state and local levels. Technology firms and data center developers face a thicket of zoning, utility, and taxation questions, some of which are the subject of new regulations and legislation. In the US, one industry group is increasingly at the center of these policy debates, an investigation led by Agência Pública and the Centro Latino-americano de Investigação Jornalística (CLIP), with Tech Policy Press, has found. As the stakes in the fight over data centers increase, the Data Center Coalition is spending more money on behalf of its corporate members to lobby governments and public entities. And, it is engaging in new tactics to shift perceptions of the industry among key constituencies, including legislators and the public. – https://www.techpolicy.press/amidst-boom-data-center-lobby-expands-its-influence-spending-and-tactics/

Albania’s prime minister wants to appoint an AI to his ministry

(The Register – 12 September 2025) Albania’s prime minister has proposed appointing an artificial intelligence as a minister. Speaking at his socialist party’s annual conference yesterday, prime minister Edi Rama named his planned cabinet, and suggested the nation’s government services chatbot “Djella” for a promotion to the ministry. According to a session description for the 2025 European Political Community Summit, which Albania hosted in May 2025, Djella started life in 2024 as a virtual assistant that Albanian citizens could use to answer questions about government services. Version 2.0, launched in early 2025, was “an intelligent avatar integrating natural language understanding with vocal and visual interaction, providing real-time information for public services and online applications on e-Albania.” – https://www.theregister.com/2025/09/12/albania_ai_minister/

FTC Opens Inquiry Into AI Chatbots and Their Impact on Children

(Ben Lennett – Tech Policy Press – 11 September 2025) The US Federal Trade Commission (FTC) has launched an inquiry into consumer-facing AI chatbots, with a particular focus on how these technologies affect children and teenagers. On Thursday, the agency issued orders to seven companies – Alphabet, Character Technologies, Instagram, Meta Platforms, OpenAI, Snap, and X.AI. The orders enable the FTC to conduct broad studies without pursuing a specific law enforcement action. According to the FTC announcement, the inquiry aims to understand how companies 1) “evaluate the safety of their chatbots when acting as companions,” 2) “limit the products’ use by and potential negative effects on children and teens,” and 3) “apprise users and parents of the risks associated with the products.” – https://www.techpolicy.press/ftc-opens-inquiry-into-ai-chatbots-and-their-impact-on-children/

EU considers social media restrictions for minors

(DigWatch – 11 September 2025) European Commission President Ursula von der Leyen announced that the EU is considering tighter restrictions on children’s access to social media platforms. During her annual State of the Union address, von der Leyen said the Commission is closely monitoring Australia’s approach, where individuals under 16 are banned from using platforms like TikTok, Instagram, and Snapchat. – https://dig.watch/updates/eu-considers-social-media-restrictions-for-minors – https://www.euronews.com/next/2025/09/10/von-der-leyen-looking-into-eu-social-media-ban-for-minors

A dynamic dialogue with Southeast Asia to put the OECD AI Principles into action

(Naoko Ueda, Karine Perset – OECD.AI – 10 September 2025) A policy roundtable in Tokyo and a workshop in Bangkok deepened the dialogue between Southeast Asia and the OECD, fostering collaboration on AI governance across countries, sectors, and policy communities. – https://oecd.ai/en/wonk/a-dynamic-dialogue-with-southeast-asia-to-put-the-oecd-ai-principles-into-action

Report: Hybridisation of Online Threats

(Tech Against Terrorism – 9 September 2025) At this year’s TS Forum, Tech Against Terrorism, in partnership with the Trust & Safety Forum, hosted a series of workshops on the Hybridisation of Online Threats. Across three interactive sessions, we brought together subject matter experts, policymakers, NGOs, and platforms to address how distinct harms – terrorist and violent extremist content (TVEC), child sexual abuse material (CSAM), image-based sexual abuse (IBSA), disinformation, and the misuse of AI – are increasingly intersecting online. Participants agreed that this hybridisation demands an equally hybrid, rights-respecting, and survivor-informed response. This post-event white paper details findings and recommendations to help drive further research and policymaking in this crucial space. – https://techagainstterrorism.org/news/report-hybridisation-of-online-threats

US Court’s Remedies Not Enough Given Google Can Continue to Pay for Search Defaults

(Sumit Sharma – Tech Policy Press – 12 September 2025) Following his 2024 liability decision that Google has a monopoly in general search and general search text advertising and that it illegally maintains its monopoly, US District Court Judge Amit Mehta issued his remedies decision last week. The remedies decision is a win for Google. The remedies ordered will not pry open competition, because Google can continue to pay to be the default service on iPhones, carriers, OEMs and browsers — for both general search and its AI services. Judge Mehta’s decision explicitly acknowledges as much on page 108 of the remedies decision: Even with newfound flexibility, distributors still are likely to select Google as its primary, if not only, default GSE. – https://www.techpolicy.press/us-courts-remedies-not-enough-given-google-can-continue-to-pay-for-search-defaults/

Decision in US vs. Google Gets it Wrong on Generative AI

(Kate Brennan – Tech Policy Press – 11 September 2025) Last week, federal court Judge Amit Mehta issued a ruling in the US v. Google remedy case rejecting many of the Department of Justice’s bold, structural remedies to unseat Google’s search monopoly position. As I wrote in Tech Policy Press earlier this year, generative AI was the elephant in the room in this case given Google’s exceptional position to leverage its search monopoly to achieve dominance in the generative AI market. In fact, the DOJ’s proposed remedies were acutely aware of this, initially proposing that Google must, among other bold structural remedies, divest from any interests in rival AI companies. Despite how much attention generative AI is given in the decision—Judge Mehta wrote that “The emergence of GenAI changed the course of this case”—the decision fails to seriously contend with the tremendous advantages Google holds in the generative AI market due to its search monopoly. It also fails to draw the necessary connections between Google’s search dominance and its control of AI across the entire AI stack. – https://www.techpolicy.press/decision-in-us-vs-google-gets-it-wrong-on-generative-ai/

Google decision demonstrates need to overhaul competition policy for AI era

(Tom Wheeler, Bill Baer – Brookings – 9 September 2025) The remedies in the Department of Justice’s case against Google reflected the inadequacy of antitrust policies in the age of AI and rapid innovation. The rise of generative AI played a sizable role in the court’s decision, with the judge finding that the technology has made online search more competitive. Yet, Google still dominates the search engine market. To keep pace with technological change, antitrust policies cannot be the only tool used. Forward-looking AI competition policy is needed. - https://www.brookings.edu/articles/google-decision-demonstrates-need-to-overhaul-competition-policy-for-ai-era/

Why humans matter most in the age of AI: Jacob Taylor on collaboration, vibe teaming, and the rise of collective intelligence

(Jacob Taylor, Junjie Ren – Brookings – 9 September 2025) Amid breathless coverage of AI’s power to disrupt, CSD Fellow Jacob Taylor asks a sharper question: How can humans design and use AI to amplify—rather than eclipse—our collective problem-solving? From the rugby scrum to the policy arena, Taylor reflects on teamwork, explores the promise of collective intelligence, and lays out the principles of vibe teaming. At the core: Human embodiment, expertise, creativity, and judgment remain indispensable in the age of AI. – https://www.brookings.edu/articles/jacob-taylor-on-ai-collective-intelligence-vibe-teaming/

Why Liability and Insurance Won’t Save AI: Lessons From Cyber Insurance

(Daniel Schwarcz, Josephine Wolff – Lawfare – 9 September 2025) In 2024, California’s state legislature considered a bill, SB 1047, or the Safe and Secure Innovation for Frontier Artificial Intelligence Models Act, that would have, among other provisions, imposed liability on some AI companies for harm caused by their machine learning models. Ultimately, California Gov. Gavin Newsom vetoed the bill, but its proponents contend that imposing legal liability is the best way to force AI companies to make their machine learning models safer and more secure. That’s not an uncommon viewpoint these days—that the most effective path to promoting AI safety is to make sure that the people developing it are held responsible for any harms that their systems cause. But the history of data breaches, cybersecurity, and cyber insurance offers some cautionary lessons about just how useful liability is likely to be for making AI safer given the key role that liability insurers will inevitably play in pricing and covering this risk. – https://www.lawfaremedia.org/article/why-liability-and-insurance-won-t-save-ai–lessons-from-cyber-insurance

The AI Deregulation Agenda Has Helped Create an AI Bubble and May Hasten a Crash

(Amber Sinha  – Tech Policy Press – 9 September 2025) A little over a month ago, Euractiv journalist Thomas Moller-Nielsen warned that the global trade war instigated by United States President Donald Trump takes attention away from a much graver economic threat — the Trump administration’s deregulation agenda. His analysis drew from a recent speech by US Federal Reserve Governor Michael Barr that pointed out that instances of significant deregulation preceded “all three of the most infamous financial meltdowns over the past century, namely the Great Depression in the 1930s, the ‘Savings & Loan’ crisis in the late 1980s and early 1990s, and the 2009 global financial crisis.” The same arguments may be applicable to digital deregulation, which may exacerbate risks to the market from unregulated artificial intelligence products and services. – https://www.techpolicy.press/the-ai-deregulation-agenda-has-helped-create-an-ai-bubble-and-may-hasten-a-crash/

AI adoption drops at large US companies for the first time since 2023

(DigWatch – 9 September 2025) Despite the hype surrounding AI, new data suggests corporate adoption of AI is slowing. A biweekly survey by the US Census Bureau found AI use among firms with over 250 employees dropped from nearly 14 percent in mid-June to under 12 percent in August, marking the largest decline since the survey began in November 2023. – https://dig.watch/updates/ai-adoption-drops-at-large-us-companies-for-the-first-time-since-2023 – https://futurism.com/ai-hype-automation-decline

AI threatens the future of entry level jobs

(DigWatch – 9 September 2025) The rise of AI puts traditional entry-level roles under pressure, raising concerns that career ladders may no longer function as they once did. Industry leaders, including Anthropic CEO Dario Amodei, warn that AI could replace half of all entry-level jobs as machines operate nonstop. A venture capital firm, SignalFire, found that hiring for graduates with under one year of experience at major tech firms fell by 50% between 2019 and 2024. The decline has been consistent across business functions, from sales and marketing to engineering and operations. –   https://dig.watch/updates/ai-threatens-the-future-of-entry-level-jobs – https://www.cnbc.com/2025/09/07/ai-entry-level-jobs-hiring-careers.html

Anthropic’s Settlement Shows the U.S. Can’t Afford AI Copyright Lawsuits

(Stewart Baker – Lawfare – 8 September 2025) Anthropic just paid $1.5 billion to settle a copyright case that it largely won in district court. Future litigants are likely to hold out for much more. A uniquely punitive provision of copyright law will allow plaintiffs who may not have suffered any damage to seek awards in the trillions. (Indeed, observers estimated that Anthropic dodged $1 trillion in liability by settling.)  The avalanche of litigation, already forty lawsuits and counting, doesn’t just put the artificial intelligence (AI) industry at risk of spending their investors’ money on settlements instead of advances in AI. It raises the prospect that the full bill won’t be known for a decade, as different juries and different courts reach varying conclusions. A decade of massive awards and deep uncertainty poses a major threat to the U.S. industry. The Trump administration saw the risk even before the  Anthropic settlement, but its AI action plan offered no solution. That’s a mistake; the litigation could easily keep the U.S. from winning its race with China to truly transformational AI. The litigation stems from AI’s insatiable hunger for training data. To meet that need, AI companies ingested digital copies of practically every published work on the planet, without getting the permission of the copyright holders. That was probably the only practical option they had. There was no way to track down and negotiate licenses with millions of publishers and authors. And the AI companies had a reasonable but untested argument that making copies for AI training was a “fair use” of the works. Publishers and authors disagreed; they began filing lawsuits, many of them class actions, against AI companies. – https://www.lawfaremedia.org/article/anthropic-s-settlement-shows-the-u.s.-can-t-afford-ai-copyright-lawsuits

Washington and Brussels Vie for Control Over Big Tech

(Mark Scott – Tech Policy Press – 8 September 2025) When US lawmakers gathered in Washington last Wednesday for a hearing about European online safety rules, you could be forgiven for thinking the almost 5-hour hearing would focus on the policy complexities of the European Union’s Digital Services Act and the United Kingdom’s Online Safety Act. Instead, the House of Representatives’ Judiciary Committee hearing became a physical representation of how the culture wars, which began in the United States, but have subsequently spread across most democratic countries, have engulfed global attempts at policing the digital world. – https://www.techpolicy.press/washington-and-brussels-vie-for-control-over-big-tech/

Empowered Workers Are a Bulwark Against Illegal Monopoly

(Stephen McMurtry – Tech Policy Press – 8 September 2025) Last week, US District Court Judge Amit Mehta issued his decision on remedies in the Department of Justice’s antitrust case against Google’s search monopoly. The ruling was widely viewed as a victory for Google, as the judge rejected most of the DOJ’s proposed remedies as “overreach” and accepted Google’s claim that it now competes in a broader information retrieval market being rapidly upended by the rise of LLM-powered chatbots and search engines such as OpenAI’s ChatGPT, Anthorpic’s Claude, Perplexity’s search engine, and even Google’s own Gemini chatbot. Google still faces a ban on exclusive distribution agreements and some data sharing requirements—a far cry from the DOJ’s push for divestment of Chrome and Android and a blanket prohibition on distribution agreements with partners such as Apple. – https://www.techpolicy.press/empowered-workers-are-a-bulwark-against-illegal-monopoly/

European Commission proposes mutual data flow agreement with Brazil

(DigWatch – 8 September 2025) The European Commission has initiated the adoption of a data protection adequacy decision with Brazil, recognising that the country offers a level of data protection comparable to the EU’s General Data Protection Regulation (GDPR). The agreement will enable seamless data transfers between the EU and Brazil across sectors, including business, government, and research. This mutual decision marks one of the broadest scopes of data adequacy granted by the EU and is expected to boost economic ties between the two regions, which together serve 670 million consumers, Tech Sovereignty, Security, and Democracy Executive Vice President Henna Virkkunen said. – https://dig.watch/updates/european-commission-proposes-mutual-data-flow-agreement-with-brazil – https://www.eureporter.co/world/brazil/2025/09/08/commission-proposes-to-create-area-of-free-and-safe-data-flows-between-the-eu-and-brazil/

ITU warns global Internet access by 2030 could cost nearly USD 2.8 trillion

(DigWatch – 7 September 2025) Universal Internet connectivity by 2030 could cost up to $2.8 trillion, according to the International Telecommunication Union (ITU) and Saudi Arabia’s Communications, Space, and Technology (CST) Commission. The blueprint urges global cooperation to connect the one-third of humanity still offline. The largest share, up to $1.7 trillion, would be allocated to expanding broadband through fibre, wireless, and satellite networks. Nearly $1 trillion is needed for affordability measures, alongside $152 billion for digital skills programmes. – https://dig.watch/updates/itu-warns-global-internet-access-by-2030-could-cost-nearly-usd-2-8-trillion – https://www.itu.int/en/mediacentre/Pages/PR-2025-09-01-Connecting-Humanity.aspx

Sovereign AI for assistive and public technologies: How policymakers can reinforce critical digital capacities

(Yonah Welker – OECD.AI – 5 September 2025) Along with many AI roboticists and other EU tech experts, I recently joined an open study group organised by the Commission’s DG CNECT on AI and robotics that focuses on policy and the deployment of critical digital capacities. This is a reflection of the geopolitical context and intensifying global technological AI race, echoing the EU and Commission’s efforts to address digital resilience and technological sovereignty. It concerns data, AI value, and supply chains, such as Eurostack, the AI Continent Action Plan, AI Factories, EU InvestAI, and the EU Chips Act. Additionally, it includes action plans from the United States and China, spanning from competition in high-performance computing and foundation models to AI infrastructure, semiconductors, and export controls. Our input helped expand this perspective to reflect on how the Commission can better approach public and assistive systems, healthcare, education, vulnerable and emerging humanitarian contexts, including robust and energy-resilient models that can operate offline, on-device or in embodied systems, supported by sensors, 3D models, spatial data, and specialised chips and materials. In fact, the taxonomies of assistive AI technologies continue to expand. The OECD’s repository of AI supporting labour with disabilities at workplaces includes over 140 systems. Cities like Barcelona deploy NaviLens, a computer vision-based system, providing real-time guidance for visually impaired travellers or Singapore’s SiLViA instantly translates spoken and written words into sign language. Furthermore, the World Health Organisation projects that the number of people needing assistive technology will grow from 2.5 to 3.5 billion—approximately 36% of the population—by 2050, yet only 10% have access to such technologies. – https://oecd.ai/en/wonk/sovereign-ai-for-assistive-and-public-technologies-critical-digital-capacities

Perplexing Agents: Redefining the Boundaries of Online Agency

(Siddharth Yadav, Ambika Sondhi – Observer Research Foundation – 5 September 2025) Disagreements and legal actions regarding copyright infringement and intellectual property theft have proliferated since the dawn of large language models (LLMs) in 2022. While LLMs offer unprecedented capabilities to streamline and automate core economic sectors from coding to public service delivery, rapid advancements in AI capabilities have challenged regulators to keep pace with the emergent capabilities of frontier models. The scramble to regulate LLMs is set to become increasingly complicated with AI agents on the algorithmic horizon. Agents are AI systems that offer additional layers of capabilities on top of LLMs, such as executing user commands in virtual environments or edge devices, accessing and retrieving user-requested information from websites, and building and deploying software by automating multi-stage decision-making and execution. While LLMs use a mix of user-generated and synthetic data to provide textual and audio-visual output to user queries, agents are programmed to take actions in the digital space. Traditionally, algorithmic tools such as ‘crawlers’ are deployed to collect data from websites en masse. While agents may be similar to crawlers in the function they perform, they differ in purpose. Crawlers are fundamentally designed to collect information for building databases. Conversely, AI agents are designed to retrieve specific data based on user inputs and can automate and execute entire workflows. While such novel capabilities may be paving the way for a paradigm shift in users’ experience on the internet, they also raise questions regarding the future of an internet increasingly populated by computational entities. A key set of questions has arisen due to a legal battle between the AI company Perplexity and web security solutions provider Cloudflare. In August 2025, Cloudflare accused Perplexity of allowing its AI agents to scrape data from websites that explicitly prohibit data scraping by crawlers. In response, Perplexity argued that agents are different from crawlers in that agents should be seen as extensions of the users themselves. Consequently, an agent accessing a website following a user query should be treated as a user, not a crawler. The difference between AI agents and crawlers may seem pedantic, but such definitional ambiguity can evolve into a fundamental obstacle for AI development and adoption — issues that have become strategic goals for nations across the world. – https://www.orfonline.org/research/perplexing-agents-redefining-the-boundaries-of-online-agency

Legislation

Through 300+ Bills, US Lawmakers Juggle Data Center Priorities

(Tim Bernard – Tech Policy Press – 12 September 2025) At last month’s National Conference of State Legislatures Legislative Summit, participants in a session entitled “AI and the Grid: A Love-Hate Relationship” discussed the pressures on energy generation and transmission that are emerging from the AI boom. While many states have introduced substantial incentives to bring in data centers, now that they are commencing operations, some of the problems they create have become clearer and legislators are reckoning with externalities like increased energy bills for other businesses and households. In a Tech Policy Press review of over 300 proposed and passed bills related to data centers, while some unidirectional proposed legislation was evident, such as introducing or broadening tax incentives or proposing new regulations, the overall picture is of a balancing act as state legislatures engage with the tradeoffs between investment and jobs on one side and a host of problems for infrastructure, neighboring communities, and the environment on the other. – https://www.techpolicy.press/through-300-bills-us-lawmakers-juggle-data-center-priorities/

(DigWatch – 12 September 2025) The Swiss government has proposed a new regulation that would require digital service providers with more than 5,000 users to collect government-issued identification, retain subscriber data for six months, and, in some cases, disable encryption. The proposal, which does not require parliamentary approval, has triggered alarm among privacy advocates and technology companies worldwide. – https://dig.watch/updates/switzerland-weighs-new-digital-security-measures –

Ted Cruz AI bill could let firms bribe Trump to avoid safety laws, critics warn

(Ars Technica – 11 September 2025) Critics are slamming Sen. Ted Cruz’s (R-Texas) new AI policy framework, which they claim would give the White House unprecedented authority to allow Big Tech companies to make “sweetheart” deals with the Trump administration to void laws designed to protect the public from reckless AI experiments. Under the framework, Cruz calls for a “light-touch” regulatory approach to “advance American leadership” in AI and ensure that “American values” are at the heart of the world’s leading technology—not Chinese values. – https://arstechnica.com/tech-policy/2025/09/ted-cruz-bill-would-let-big-tech-go-wild-with-ai-experiments-for-10-years/

California Assembly Passes Landmark Bill Regulating AI Companion Chatbots

(AI Insider – 11 September 2025) The California State Assembly has passed SB 243, legislation introduced by Senators Steve Padilla and Josh Becker to regulate AI companion chatbots and protect minors and vulnerable users. The bill, which passed with bipartisan support, now heads to the state Senate for a final vote. – https://theaiinsider.tech/2025/09/11/california-assembly-passes-landmark-bill-regulating-ai-companion-chatbots/

Experts scrutinized Ofcom’s Online Safety Act governance. They’re concerned

(The Register – 11 September 2025) Industry experts expressed both concern and sympathy for Ofcom, the Brit regulator that is overseeing the Online Safety Act, as questions mount over the effectiveness of the controversial legislation. The UK’s Communications and Digital Committee heard from academic and industry leaders this week about various aspects of how the Online Safety Act (OSA) is being enforced, many of whom criticized the regulator’s recent comments. Particular attention was given to Ofcom’s previous claim that the OSA could potentially have prevented situations similar to the Southport riots which broke out in 2024, had it been in force at the time. – https://www.theregister.com/2025/09/11/concern_and_sympathy_as_experts/

The Next Cyber Breach Will Not Wait: Why Congress Must Reauthorize CISA 2015

(Simin Kargar – Just Security – 12 September 2025) On Sept. 30, 2025, the Cybersecurity Information Sharing Act of 2015 (CISA 2015) will expire. For a decade, the statute has been the legal backbone of U.S. cyber defense, born from the high-profile breaches that shook U.S. companies a decade ago and forced Congress to act. By creating a liability shield, CISA 2015 finally gave companies the reassurance to share threat data without fear of lawsuits. Without it, the United States risks sliding back into the pre-2015 world of legal uncertainty, ad hoc disclosures, and systemic blind spots, just as adversaries like China double down on espionage, ransomware, and supply-chain compromise. Last week, on Sept. 3, the House Committee on Homeland Security unanimously advanced the Widespread Information Management for the Welfare of Infrastructure and Government Act (WIMWIG Act), introduced by Rep. Andrew R. Garbarino (R-NY), to reauthorize CISA 2015. The move rightly underscores the statute’s central role in the United States’ collective security. Its renewal is now headed to the House floor and then the Senate––an imperative step at a moment of escalating digital threats. Yet reauthorization remains uncertain, clouded by debates over the law’s institutional legacy––especially the performance of the Cybersecurity and Infrastructure Security Agency (CISA) and its controversial role in issues like countering disinformation. But conflating the statute with the agency would strip away the foundation that underpins the United States cyber defense. History shows that without a firm legal framework, the United States enters the next phase of cyber competition disarmed. – https://www.justsecurity.org/120195/the-next-cyber-breach-will-not-wait-why-congress-must-reauthorize-cisa-2015/

CISA ready to accept any extension for key cyber info-sharing law, official says

(David DiMolfetta – NextGov – 12 September 2025) A top Cybersecurity and Infrastructure Security Agency official said the agency is prepared to accept any extension Congress authorizes for a fundamental cybersecurity threat intelligence-sharing law, which is set to expire Sept. 30 unless renewed by lawmakers. “We’ll take whatever the Congress decides to authorize us, wherever they see fit within their purview, to authorize and to give us our authorities to be able to use,” Nick Andersen, CISA’s executive assistant director for cybersecurity, told reporters Thursday on the sidelines of the Billington Cyber Summit. The Cybersecurity Information Sharing Act of 2015 lets private sector providers freely transmit cyber threat information to government partners with key liability protections in place, shielding firms from lawsuits and regulatory penalties when sharing threat data with the government. – https://www.nextgov.com/cybersecurity/2025/09/cisa-ready-accept-any-extension-key-cyber-info-sharing-law-official-says/408070/?oref=ng-homepage-river

Cybersecurity protections for US companies at risk as key law nears expiration

(DigWatch – 11 September 2025) As cyber threats grow, a vital legal safeguard encouraging US companies to share threat intelligence is on the verge of expiring. The US Cybersecurity Information Sharing Act of 2015 (CISA 2015), which grants liability protection to firms that voluntarily share cyber threat data with peers and the federal government, is set to lapse at the end of the month unless Congress acts swiftly. – https://dig.watch/updates/cybersecurity-protections-for-us-companies-at-risk-as-key-law-nears-expiration – https://news.bloomberglaw.com/privacy-and-data-security/trump-cuts-imperil-private-sector-cybersecurity-cooperation

Geostrategies

2025 State of the Union: tech sovereignty amid geopolitical pressure

(DigWatch – 10 September 2025) The European Commission President, Ursula von der Leyen, delivered her 2025 State of the Union address to the European Parliament in Strasbourg. The speech set out priorities for the coming year, and was framed by growing geopolitical tensions and the push for a more self-reliant Europe. Von der Leyen highlighted that global dynamics have shifted. ‘Battlelines for a new world order based on power are being drawn right now’, she said. In this context, Europe must take a more assertive role in defending its own security and advancing the technologies that will underpin its economic future. The President characterised this moment as a turning point for European independence. Digital policy appeared less prominently than expected in the address. Von der Leyen often referred to ‘technology sovereignty’, to encompass not only digital technologies, but also other types of technologies necessary for the green transition and to achieve energetic autonomy. – https://dig.watch/updates/2025-state-of-the-union-tech-sovereignty-amid-geopolitical-pressure – https://commission.europa.eu/strategy-and-policy/state-union/state-union-2025_en

State of the Union 2025: What’s Ahead for Tech in Europe

(Ramsha Jahangir – Tech Policy Press – 10 September 2025) Today, European Commission President Ursula von der Leyen delivered her 2025 State of the Union address to the European Parliament in Strasbourg, outlining the Commission’s priorities for the year ahead amid rising geopolitical tensions and the drive to establish a more independent Europe. “Battle lines for a new world order based on power are being drawn right now. So, yes, Europe must fight,” von der Leyen declared, emphasizing what she said is the urgent need for the continent to take responsibility for its own defense and security as well as technologies that will fuel European economies. This, she said, must be Europe’s “Independence Moment.”. The European Commission president’s message aligns with recent calls from leaders such as Italian Prime Minister Giorgia Meloni and French President Emmanuel Macron, who have urged Europe to assert itself more forcefully or risk losing influence on the world stage. While von der Leyen expressed strong ambitions for European sovereignty, her speech offered nothing concrete or new on technology. – https://www.techpolicy.press/state-of-the-union-2025-whats-ahead-for-tech-in-europe/

Digital Sovereignty Requires More Than Just European Tech

(Eliot Bendinelli – Tech Policy Press – 10 September 2025) When European leaders make exclusive announcements on X, do they consider what would happen if Elon Musk decided to delete them? Nothing prevents him from doing so. As the US continues to pressure allies and enemies alike to protect its tech industry, the response should be a resounding yes. From social media to cloud services, AI compute power and microchips, governments around the world are waking up to how dependent they are on American tech. As a response, the principle of digital sovereignty is rapidly gaining momentum. Europe, in particular, on the back of years of increasingly ambitious tech legislation, is ripe for discussion and home to a flurry of proposals. A prominent one is the Eurostack, a European values-driven project. While this analysis is accurate, the strategies identified to enact the EU’s digital sovereignty often seem focused on imitating the success of the US tech sector. – https://www.techpolicy.press/digital-sovereignty-requires-more-than-european-tech/

AI export rules tighten as the US opens global opportunities

(DigWatch – 10 September 2025) President Trump has signed an Executive Order to promote American leadership in AI exports, marking a significant policy shift. The move creates new global opportunities for US businesses but also introduces stricter compliance responsibilities. The order establishes the American AI Exports Program, overseen by the Department of Commerce, to develop and deploy ‘full-stack’ AI export packages. – https://dig.watch/updates/ai-export-rules-tighten-as-the-us-opens-global-opportunities – https://www.jdsupra.com/legalnews/u-s-executive-order-propels-american-ai-3952418

Japan and Taiwan deepen semiconductor collaboration beyond business

(DigWatch – 10 September 2025) Experts from Japan and Taiwan have highlighted closer cooperation in semiconductors, stressing that ties extend beyond business into education and research. At a forum in Taipei, SEMI’s Terry Tsao said Taiwan’s chip design and fabrication expertise complements Japan’s materials and manufacturing equipment strengths. He noted that TSMC’s Kumamoto expansion creates opportunities for talent development in partnership with local schools. – https://dig.watch/updates/japan-and-taiwan-deepen-semiconductor-collaboration-beyond-business – https://japantoday.com/category/tech/japan-taiwan-highlight-semiconductor-talent-research-collaboration

Malta: Lessons in Agility, Collaboration, and Vision

(Quantum Insider – 9 September 2025) Malta served as the first stop of the Quantum World Tour, an initiative launched by the ITU and The Quantum Insider to showcase how nations translate quantum strategy into action. Malta positions itself as a testbed and connector, leveraging its small size and geography to advance projects like PRISM, Europe’s first nationwide QKD network. Startups like Merqury Cybersecurity demonstrate how research translates into real-world impact, addressing urgent risks such as “harvest now, decrypt later” threats. Workforce leaders stress the need for reskilling, stronger institutional structures, and collaboration across government, academia, and industry to sustain growth. – https://thequantuminsider.com/2025/09/09/malta-lessons-in-agility-collaboration-and-vision/

National cyber director: U.S. strategy needs to shift cyber risk from Americans to its adversaries

(Tim Starks – Cyberscoop – 9 September 2025) The United States needs a “new, coordinated strategy” to counter its cyber adversaries and “shift the burden of risk in cyberspace from Americans to them,” National Cyber Director Sean Cairncross said Tuesday. “Collectively, we’ve made great progress in identifying, responding to and remediating threats, but we still lack strategic coherence and direction,” he said at the Billington Cybersecurity Summit. “A lot has been done, but it has not been sufficient. We’ve admired the problem for too long, and now it’s time to do something about it.” – https://cyberscoop.com/us-cybersecurity-strategy-sean-cairncross-shift-risk-china-trump-biden-cisa/

New project expands AI access for African languages

(DigWatch – 9 August 2025) Africa is working to close the AI language gap, as most global tools remain trained on English, Chinese, and European languages. The African Next Voices project has created the continent’s largest dataset of spoken African languages, covering 18 tongues across Kenya, Nigeria, and South Africa. Supported by a $2.2m Gates Foundation grant, the dataset includes 9,000 hours of speech in farming, health, and education settings. – https://dig.watch/updates/new-project-expands-ai-access-for-african-languages – https://www.bbc.com/news/articles/crkzgkkpx0lo

A Strategic Bet to Advance America’s Quantum Leadership

(Constanza M. Vidal Bustamante – Just Security – 8 September 2025) The Trump administration has rightly cast securing U.S. quantum leadership as a national security imperative. By harnessing the quirks of subatomic physics, quantum technologies promise vast economic gains in sectors spanning pharmaceuticals, transportation, and communications. They also carry profound security stakes, enabling new battlefield advantages and cyber threats that could rebalance global power. But with tightening federal budgets, advancing America’s quantum ambitions will require smart and creative bets with outsized returns, in the words of Office of Science and Technology Policy Director Michael Kratsios. Quantum sensing—the most mature quantum technology today—offers just such an opportunity. Targeted federal initiatives could accelerate the deployment of quantum sensors in the next three years. This would unlock substantial defense, security, and economic benefits, while simultaneously strengthening U.S. domestic supply chains and global quantum leadership. Yet timely and decisive action is essential. As Kratsios warned, “There is nothing predestined about technological progress.” – https://www.justsecurity.org/120024/advance-us-quantum-leadership/

Trilateral quantum talks highlight innovation and security priorities

(DigWatch – 8 September 2025) The United States, Japan, and South Korea held two Trilateral Quantum Cooperation meetings this week in Seoul and Tokyo. Officials and experts from government and industry gathered to discuss securing quantum ecosystems against cyber, physical, and intellectual property threats. The US State Department stressed that joint efforts will ensure breakthroughs in quantum computing benefit citizens while safeguarding innovation. Officials said cooperation is essential as quantum technologies could reshape industries, global power balances, and economic prosperity. – https://dig.watch/updates/trilateral-quantum-talks-highlight-innovation-and-security-priorities – https://www.upi.com/Top_News/World-News/2025/09/05/9481757100890

Nigeria sets sights on top 50 AI-ready nations

(DigWatch – 8 September 2025) Nigeria has pledged to become one of the top 50 AI-ready nations, according to presidential adviser Hadiza Usman. Speaking in Abuja at a colloquium on AI policy, she said the country needs strong leadership, investment, and partnerships to meet its goals. She stressed that policies must address Nigeria’s unique challenges and not simply replicate foreign models. The government will offer collaboration opportunities with local institutions and international partners. – https://dig.watch/updates/nigeria-sets-sights-on-top-50-ai-ready-nations – https://gazettengr.com/nigeria-aims-to-be-among-top-50-ai-ready-countries-fg/

China Refocuses Hong Kong Labs on Quantum and National Priorities

(Quantum Insider – 6 September 2025) Hong Kong’s research labs have been restructured to align with Beijing’s national technology goals, with quantum science emerging as a central priority, according to the South China Morning Post. The overhaul closed underperforming labs, rebranded others, and launched new State Key Laboratories, including two focused on quantum information and optical quantum materials. The changes reflect China’s broader strategy to direct scientific resources toward fields like quantum, AI, and brain science, positioning Hong Kong’s universities as instruments in the country’s competition with the U.S. – https://thequantuminsider.com/2025/09/06/china-refocuses-hong-kong-labs-on-quantum-and-national-priorities/

Defence, Warfare

Shield AI and HII Partner to Accelerate Modular Cross-Domain Mission Autonomy Solutions

(AI Insider – 11 September 2025) HII, America’s largest military shipbuilder, and Shield AI announced a partnership at DSEI 2025 to advance modular, cross-domain mission autonomy by integrating Shield AI’s Hivemind software with HII’s Odyssey suite. The collaboration will focus first on unmanned surface vehicles (USVs) and later expand to unmanned undersea vehicles (UUVs), leveraging HII’s maritime expertise and Shield AI’s autonomy proven on air platforms such as the MQ-20 Avenger, V-BAT, and F-16. By pairing autonomy software with decades of shipbuilding and integration experience, the companies aim to deliver scalable, resilient solutions enabling U.S. and allied forces to operate across air, land, surface, and undersea domains with greater speed and flexibility. – https://theaiinsider.tech/2025/09/11/shield-ai-and-hii-partner-to-accelerate-modular-cross-domain-mission-autonomy-solutions/

Milrem Robotics, Overwatch Aerospace, MSI-Defence Systems, and Pearson Engineering Join Forces to Deliver UK-Specific THeMIS Combat UGV

(AI Insider – 10 September 2025) Milrem Robotics and UK partners Overwatch Aerospace, MSI-Defence Systems, and Pearson Engineering have launched a UK-specific version of the THeMIS Combat Unmanned Ground Vehicle (UGV) at DSEI 2025. The platform combines Milrem’s THeMIS UGV with MSI’s TERRAHAWK remote weapon station and Overwatch’s APEX unmanned aerial system, while Pearson provides integration expertise to tailor the vehicle to UK operational needs. Development, testing, and production will take place in the UK, with the program now entering trials to deliver a deployable, scalable robotic combat solution for the British Army. – https://theaiinsider.tech/2025/09/10/milrem-robotics-overwatch-aerospace-msi-defence-systems-and-pearson-engineering-join-forces-to-deliver-uk-specific-themis-combat-ugv/

Modern military training demands northern Australian digital upgrade

(John Coyne – The Strategist – 8 September 2025) The wars of the future will be won by those who train together in live, networked and constructive environments that reflect the full complexity of modern conflict. Australia’s training infrastructure has not fully kept pace with this reality, leaving gaps in how the Australian Defence Force prepares for high-end warfighting. To close this gap, Australia must make decisive investments in digital infrastructure across its exercise areas, particularly in the Northern Territory. Modern militaries now require more than vast land or airspace to conduct realistic training; they need digitally networked environments that seamlessly integrate live forces with virtual and constructive elements across geographies. Such live-virtual-constructive (LVC) training allows commanders to rehearse operations with combat-like fidelity, linking sensors, shooters and decision-makers in real time. –  https://www.aspistrategist.org.au/modern-military-training-demands-northern-australian-digital-upgrade/

20kW high-energy laser systems that engage with aerial threats delivered to US Army

(Interesting Engineering – 7 September 2025) A Virginia-based company has delivered two high-energy laser system units to the U.S. Army. Developed by AeroVironment, the 20kW LOCUST Laser Weapon System (LWS) identifies, tracks, and engages a wide variety of targets with hard-kill high energy laser. The company delivered the first two mobile counter-unmanned aircraft system (C-UAS) prototype Laser Weapon Systems (LWS) to the U.S. Army Rapid Capabilities and Critical Technologies Office (RCCTO). The Army Multi-Purpose High Energy Laser (AMP-HEL) prototype systems feature AV’s 20kW-class LOCUST LWS integrated on the General Motors Defense Infantry Squad Vehicle (ISV) platform. – https://interestingengineering.com/military/high-energy-laser-systems-engage-threat

US F-15E Strike Eagle jets get laser-guided precision rockets to eliminate attack drones

(Interesting Engineering – 7 September 2025) A new type of weapon system has been integrated into U.S. F-15E fighter jets. The warplanes recently conducted operational flight testing of AGR-20F live fire at targets over land and water. The F-15E Strike Eagle equipped with AGR-20F Advanced Precision Kill Weapon System II laser-guided rockets deliver drone targeting capability. The Air Force revealed that the goal of adding this capability to the Strike Eagle’s arsenal was to bring it into the counter-unmanned aerial system fight in a cost-effective way alongside the F-16 Fighting Falcon. Within a week of the demonstrations, Strike Eagles flew with the new weapon in a geographic combatant command’s area of responsibility. – https://interestingengineering.com/military/f15-with-advanced-weapon-eliminate-drone

Radicalization, Extremism, Terrorism

The Online Radicalization of Youth Remains a Growing Problem Worldwide

(The Soufan Center – 9 September 2025) The online radicalization of youth worldwide is a growing problem that policymakers and government officials continue to grapple with. Youth radicalization is not only a problem in the United States and Europe but also in Asia, particularly in South Korea, India, the Philippines, and Singapore. Social media platforms like TikTok, X, and Facebook enable violent extremists to recruit youths more expediently than in-person; algorithms channel those youths to more emotionally charged content; and online gaming enables both isolation and community building; each of these has a radicalizing effect. A public health approach where practitioners work to “inoculate” youths against extremism by providing knowledge, alternative narratives, and community is beginning to bear fruit, but funding cuts could challenge the initiatives. – https://thesoufancenter.org/intelbrief-2025-september-9/

Security

NATO and Seoul expand cybersecurity dialogue and defence ties

(DigWatch – 12 September 2025) South Korea and NATO have pledged closer cooperation on cybersecurity following high-level talks in Seoul this week, according to Yonhap News Agency. The discussions, led by Ambassador for International Cyber Affairs Lee Tae Woo and NATO Assistant Secretary General Jean-Charles Ellermann-Kingombe, focused on countering cyber threats and assessing risks in the Indo-Pacific and Euro-Atlantic regions. – https://dig.watch/updates/nato-and-seoul-expand-cybersecurity-dialogue-and-defence-ties – https://www.aa.com.tr/en/asia-pacific/south-korea-nato-hold-talks-to-enhance-cybersecurity-cooperation/3685423#

UK launches CAF 4.0 for cybersecurity

(DigWatch – 12 September 2025) The UK’s National Cyber Security Centre has released version 4.0 of its Cyber Assessment Framework to help organisations protect essential services from rising cyber threats. An updated CAF that provides a structured approach for assessing and improving cybersecurity and resilience across critical sectors. – https://dig.watch/updates/uk-launches-caf-4-0-for-cybersecurity – https://www.ncsc.gov.uk/blog-post/caf-v4-0-released-in-response-to-growing-threat

Vietnam confirms cyberattack on National Credit Information Center

(DigWatch – 12 September 2025) Vietnam’s National Credit Information Centre (CIC), a key financial data hub under the State Bank of Vietnam, confirmed a cybersecurity attack, according to the Vietnam Cyber Emergency Response Centre (VNCERT). Initial investigations suggest the attack was a deliberate attempt by cybercriminals to steal personal data. VNCERT reported signs of unauthorized data access and potential leaks of sensitive information. The Department of Cybersecurity and High-Tech Crime Prevention has tasked VNCERT with leading the incident response and coordinating with major cybersecurity firms, including Viettel, VNPT, and NCS. – https://dig.watch/updates/vietnam-confirms-cyberattack-on-national-credit-information-center – https://vietnamnews.vn/economy/1725079/national-credit-info-centre-hacked-personal-data-at-risk.html

AI-powered penetration tool, an attacker’s dream, downloaded 10K times in 2 months

(The Register – 11 September 2025) Villager, a new penetration-testing tool linked to a suspicious China-based company and described by researchers as “Cobalt Strike’s AI successor,” has been downloaded about 10,000 times since its release in July. The package, published on Python Package Index, operates as a Model Context Protocol (MCP) client and integrates multiple security tools. It includes Kali Linux, which legitimate defenders use to automate penetration testing, and it contains hundreds of tools that can also be used to launch cyber attacks at scale. Villager also contains DeepSeek AI models to automate testing workflows, plus a ton of other AI tools like a database of 4,201 AI system prompts to generate exploits and other mechanisms to make it difficult to detect. – https://www.theregister.com/2025/09/11/cobalt_strikes_ai_successor_downloaded/

Leverage the CLOUD Act To Protect Encryption

(Bob Goodlatte, Greg Nojeim – Tech Policy Press – 11 September 2025) When you meet in person with a neighbor or a friend, you would be deeply troubled if the government demanded that you spoke loudly enough so its agents could eavesdrop on everything you both say. Why should that be any different in your online communications as well? Encryption ensures that emails, texts, and other communications are kept private between the sender and the receiver. In an age of cybersecurity threats, encryption protects sensitive personal medical and financial information, and the trade secrets that underlie the success of American business. This technology, among the most important data security tools companies offer, is important for more than just privacy. Encryption protects journalists from malevolent governments. It shields women and their children from stalkers and abusers. Such reasons are partly why late last year the cybersecurity agencies of four of the Five Eyes countries urged network managers to ensure that communications traffic is encrypted end-to-end. These agencies made this recommendation in the wake of the “Salt Typhoon” attacks, in which hackers believed to be associated with the Chinese government infiltrated the networks of major US telecom operators. The UK stood apart, refusing to join cybersecurity agencies in the US, Canada, Australia and New Zealand in this pro-encryption declaration. – https://www.techpolicy.press/leverage-the-cloud-act-to-protect-encryption/

Cyberattacks against schools driven by a rise in student hackers, ICO warns

(Alexander Martin – The Record – 11 September 2025) The U.K.’s Information Commissioner’s Office (ICO) warned on Thursday that student hackers motivated by dares are driving an increasing number of cyberattacks and data breaches affecting schools. It advised parents to “to have regular conversations with their children about what they get up to online” and warned that children hacking into their school’s computer systems may be setting themselves up for lives of cybercrime. The privacy regulator said it identified “a worrying pattern” in the 215 insider threat breach reports from the education sector between January 2022 and August 2024, with 57% of incidents caused by students who were likely motivated by “dares, notoriety, financial gain, revenge and rivalries.” – https://therecord.media/cyberattacks-against-schools-driven-by-student-hackers

When the Vibes Are Off: The Security Risks of AI-Generated Code

(Carolin Kemper – Lawfare – 10 September 2025) In 2023, Andrej Karpathy, co-founder of OpenAI, prophesized that “[t]he hottest new programming language is English,” envisioning a future in which artificial intelligence (AI) would generate code from natural language instructions. Now, he is heralding the era of “vibe coding”—where software is developed entirely by prompting AI agents. So now, you can “fully give in to the vibes (…) and forget that the code even exists,” as Karpathy put it. Vibe coding is a novel approach to developing software: Based on prompts, large language models (LLMs) take over the entire software development process, rather than simply supporting it. This innovation is possible because of agentic AI: systems of multiple AI agents that collaborate to solve complex, multi-step tasks. “Coding agents”—such as CodeGPT, Cursor, or Claude Opus 4—assist software developers in churning out ready-to-run code. They design the architecture of a program, generate code, find appropriate open-source components, test software, and fix errors. They can also provide explanations about the code to help their users understand the logic of the software or the purpose of different functions and variables. Since developers can rely on coding agents, with no need for programming knowledge or experience, this technology makes programming accessible to everyone. Vibe coding is the ideal solution if you have “a vision that you can’t execute but AI can,” as one MIT Media Lab researcher put it. At the same time, the ease of vibe coding introduces severe risks: This approach implies accepting AI-generated code without thorough code review or manual testing. All quality assurance occurs through additional prompts rather than traditional verification methods, such as (static) code analysis or dynamic testing. As a result, vibe coders—according to the very definition of vibe coding—do “not need to understand how or why the code works, and often will have to accept that a certain number of bugs and glitches will be present.” – https://www.lawfaremedia.org/article/when-the-vibe-are-off–the-security-risks-of-ai-generated-code

KillSec Ransomware is Attacking Healthcare Institutions in Brazil

(Security Affairs – 10 September 2025) The KillSec Ransomware group has threatened to leak sensitive data unless negotiations are initiated promptly. According to threat intelligence reporting by Resecurity, the root cause of the incident – data exfiltration from insecure AWS S3 bucket. Considering the investigation performed by cybersecurity experts, the window of exposure can be estimated at ‘several months.’ Probably, this is the first notable supply chain incident affecting the healthcare industry in Brazil. Notably, it is not the first time the ransomware group has targeted Brazil. Some time ago, the actors leaked personal and business data containing CNPJ/CPF identifiers, transaction amounts, banking information, and other data from government resources in Brazil. At that time, the group did not clarify the full scope of the breach or its possible source. KillSec Ransomware was known for both confirmed incidents and fakes or speculations. – https://securityaffairs.com/182063/cyber-crime/killsec-ransomware-is-attacking-healthcare-institutions-in-brazil.html

Threat Actor Accidentally Exposes AI-Powered Operations

(Alessandro Mascellino – Infosecurity Magazine – 9 September 2025) A threat actor has unintentionally revealed their methods and day-to-day activities after installing Huntress security software on their own operating machine. The unusual incident gave analysts a remarkable inside look into how attackers use artificial intelligence (AI), research tools and automation to refine their workflows. – https://www.infosecurity-magazine.com/news/threat-actor-exposes-operations/

Salty2FA Phishing Kit Unveils New Level of Sophistication

(Alessandro Mascellino – Infosecurity Magazine – 9 September 2025) A phishing campaign leveraging the Salty2FA kit has been uncovered by cybersecurity researchers, revealing advanced techniques that highlight the growing professionalism of cybercrime operations. The kit demonstrates a high degree of technical innovation, with layered defenses designed to bypass traditional detection. – https://www.infosecurity-magazine.com/news/salty2fa-phishing-kit/

Axios User Agent Helps Automate Phishing on “Unprecedented Scale”

(Phil Muncaster – Infosecurity Magazine – 9 September 2025) Security experts have warned of a huge uptick in automated phishing activity abusing the Axios user agent and Microsoft’s Direct Send feature. ReliaQuest claimed in a new report today that it observed a 241% increase in phishing activity using Axios between June and August 2025. Axios accounted for nearly a quarter (24%) of all malicious user-agent activity analyzed in the period, making it 10 times more common than any other agents tracked by ReliaQuest. The threat intelligence vendor said Axios-powered attacks had a 58% success rate versus just 9% for incidents without the user agent. – https://www.infosecurity-magazine.com/news/axios-user-agent-automate-phishing/

LunaLock Ransomware threatens victims by feeding stolen data to AI models

(Pierluigi Paganini – Security Affairs – 9 September 2025) A new ransomware group, named LunaLock, appeared in the threat landscape with a unique cyber extortion technique, threatening to turn stolen art into AI training data. Recently, the LunaLock group targeted the website Artists&Clients and stole digital art. The group demanded $50K to the victims, threatening leaks and the use of the stolen data to train large language models (LLMs). “We have breached the website Artists&Clients to steal and encrypt all its data. If you are a user of this website, you are urged to contact the owners and insist that they pay our ransom. If the ransom is not paid, we will release all data publicly on this Tor site, including source code and personal data of users. Additionally, we will submit all artwork to AI companies to be added to training datasets.” reads the announcement published by the ransomware group on its Tor data leak site. – https://securityaffairs.com/182014/malware/lunalock-ransomware-threatens-victims-by-feeding-stolen-data-to-ai-models.html

CISA pushes final cyber incident reporting rule to May 2026

(Tim Starks – Cyberscoop – 8 September 2025) The Cybersecurity and Infrastructure Agency is delaying finalization of a rule until May of next year that will require critical infrastructure owners and operators to swiftly report major cyber incidents to the federal government, according to a recent regulatory notice. Under the Cyber Incident Reporting for Critical Infrastructure Act (CIRCIA) of 2022, CISA was supposed to produce a final rule enacting the law by October of this year. But last week, the Office of Management and Budget’s Office of Information and Regulatory Affairs published an update that moved the final rule’s arrival to May 2026. A CISA official told CyberScoop that the move would give the agency time to consider streamlining and reducing the burden on industry of a previously proposed version of the rule, citing public comments in response to that version, as well as harmonizing the law with other agencies’ cyber regulations. – https://cyberscoop.com/cisa-pushes-final-cyber-incident-reporting-rule-to-may-2026/

MostereRAT Targets Windows Users With Stealth Tactics

(Alessandro Mascellino – Infosecurity Magazine – 8 September 2025) A phishing campaign delivering a new strain of malware, MostereRAT, has been uncovered by cybersecurity researchers. The Remote Access Trojan (RAT) targets Microsoft Windows systems and gives attackers complete control over compromised machines. According to FortiGuard Labs, which discovered the threat, what sets this campaign apart is its layered use of advanced evasion techniques. The malware is written in Easy Programming Language (EPL), a Chinese-based coding language rarely used in cyberattacks, and relies on multiple stages to hide malicious behavior. It can disable security tools, block antivirus traffic and establish secure communications with its command-and-control (C2) server using mutual TLS (mTLS). – https://www.infosecurity-magazine.com/news/rat-targets-windows-users-stealth/

Remote Access Abuse Biggest Pre-Ransomware Indicator

(James Coker – Infosecurity Magazine – 8 September 2025) Abuses of remote access software and services are the most common ‘pre-ransomware’ indicators, according to new research from Cisco Talos. Adversaries frequently leverage legitimate remote services such as RDP, PsExec and PowerShell, the researchers observed. Additionally, remote access software such as AnyDesk, Atera and Microsoft Quick Assist were often exploited. Cisco identified these tactics, techniques and procedures (TTPs) as part of efforts by cybercriminals to gain enterprise-level domain administrator access in compromised systems. Pre-ransomware refers to the stage in an attack where adversaries conduct activities such as privilege escalation, credential harvesting and remote access deployment without yet executing full-scale encryption.   – https://www.infosecurity-magazine.com/news/remote-access-abuse-pre-ransomware/

Venezuela’s President Maduro said his Huawei Mate X6 cannot be hacked by US cyber spies

(Pierluigi Paganini – Security Affairs – 8 September 2025) Last week, Venezuelan President Nicolás Maduro showcased a Huawei Mate X6 smartphone, reportedly gifted by China’s President Xi Jinping, claiming that US cyber spies cannot hack it. Venezuelan President Maduro said that his device is “the best phone in the world”. “Impressive, I find out everything through this, the phone that Xi Jinping gave me. Look, Xi Jinping gave me this, a Huawei, the best phone in the world, the Huawei, and the Americans can’t hack it, neither their spy planes, nor their satellites.” said Venezuela’s president, Nicolás Maduro during a press conference on September 1st, 2025. Is Huawei Mate X6 smartphone really difficult to hack? Why? – https://securityaffairs.com/181984/security/venezuelas-president-maduro-said-his-huawei-mate-x6-cannot-be-hacked-by-us-cyber-spies.html

Insider Threats Surge: What CISOs Must Know to Protect Their Organizations

(Phil Muncaster – Infosecurity Magazine – 8 September 2025) People are often described as one of the biggest security threats to any organization. At first glance, it would be hard to argue with such a sweeping statement. Whether the result of malice or negligence, the  ‘human element’ featured in around 60% of data breaches over the past year, according to Verizon. A recent spate of attacks targeting corporate Salesforce instances highlights the evolving nature of the social engineering threat – and just what’s at stake. The challenge for CISOs is that insider risk is not just about negligence. Those intent on wrongdoing are usually harder to spot and exact a much heavier toll on their employer. To coincide with International Insider Threat Awareness Month, we take a look at what CISOs can do to push back the tide. – https://www.infosecurity-magazine.com/news-features/insider-threats-what-cisos-must/

Czech cyber agency NUKIB flags Chinese espionage risks to critical infrastructure

(Pierluigi Paganini – Security Affairs – 8 September 2025) The Czech Republic’s National Cyber and Information Security Agency (NUKIB) warns of growing risks from Chinese-linked technologies in critical sectors like energy, healthcare, transport, and government. The agency warns of risks from Chinese-made devices (phones, cars, cameras, LLMs). “The penetration of these technologies and devices into critical industries (such as transport, energy, healthcare, public administration and others) is growing and will continue to grow in the future. Current critical infrastructure systems are increasingly dependent on storing and processing data in cloud storage and on network connectivity that allows remote operation and updates.” reads the statement published by NUKIB. “In practice, this means that suppliers of technological solutions have the ability to fundamentally influence the operation of critical infrastructure and/or access important data, and trust in the reliability of the supplier is therefore absolutely crucial. “. Czech agency warns of data transfers and remote asset control from China-linked threat actors. The entities under the Cyber Security Act must address the threat. – https://securityaffairs.com/181976/intelligence/czech-cyber-agency-nukib-flags-chinese-espionage-risks-to-critical-infrastructure.html

NSA, CISA and others urge for unified approach to strengthen cybersecurity resilience

(DigWatch – 8 September 2025) The National Security Agency (NSA) has joined the Cybersecurity and Infrastructure Security Agency (CISA) and other partners to release a new Cybersecurity Information Sheet (CSI) titled ‘A Shared Vision of Software Bill of Materials’ (SBOM) for Cybersecurity. Aimed at promoting the adoption of SBOM practices, the report highlights their role in improving transparency and addressing risks within the software supply chain. – https://dig.watch/updates/nsa-cisa-and-others-urge-for-unified-approach-to-strengthen-cybersecurity-resilience

Cyberfraud epicentre: Myanmar scam centres are a global cyber and humanitarian threat

(Nathan Ruser – ASPI The Strategist – 8 September 2025) Myanmar has rapidly become the global epicentre of a burgeoning cyberfraud industry, a crisis deeply interwoven with the dynamics of the post-coup conflict that has overtaken the country since 2021. This industry now poses an unprecedented global cybersecurity threat. And it isn’t just a cyber or economic problem, but a humanitarian catastrophe, reliant on human trafficking and forced labour to enable malicious scamming. This shows how conflict and transnational crime syndicates intersect to build an empire of fraud. A new ASPI report, Scamland Myanmar, reveals the explosive growth of dedicated scam compounds along the Thailand-Myanmar border since 2021, increasing from 11 dedicated compounds to 30, and being constructed at an average of 13.5 acres (around 54,000 square metres) per month for the past four years. – https://www.aspistrategist.org.au/cyberfraud-epicentre-myanmar-scam-centres-are-a-global-cyber-and-humanitarian-threat/

Frontiers

China develops world’s first brain-like AI

(DigWatch – 12 September 2025) Chinese scientists have unveiled SpikingBrain1.0, the world’s first large-scale AI language model to replicate the human brain. The model reduces energy use and runs independently of Nvidia chips, departing from conventional AI architectures. Developed by the Chinese Academy of Sciences, SpikingBrain1.0 uses spiking neural networks to activate only the required neurons for each task, rather than processing all information simultaneously. – https://dig.watch/updates/china-creates-brain-inspired-ai-model – https://www.azernews.az/region/247289.html

IonQ Announces IonQ Federal to Meet the Increasing Demand for Quantum Advantage Across U.S. And Allied Governments

(Quantum Insider – 11 September 2025) IonQ announced the creation of IonQ Federal to integrate its quantum computing and networking technologies for the U.S. government and allied agencies. The new entity consolidates IonQ’s government-focused work, building on over $100 million in contracts with agencies such as the Air Force Research Lab, DARPA, Oak Ridge National Labs, and ARLIS. Robert Cardillo, former Director of the National Geospatial-Intelligence Agency and current IonQ board member, has been appointed Executive Chairman of IonQ Federal to lead strategic operations and partnerships. – https://thequantuminsider.com/2025/09/11/ionq-announces-ionq-federal-to-meet-the-increasing-demand-for-quantum-advantage-across-u-s-and-allied-governments/

Teton.ai Raises $20M to Reinvent Elderly Care

(AI Insider – 11 September 2025) Teton.ai raised $20M in a Series A led by Plural with backing from Bertelsmann Investments, Antler Elevate, Nebular, and PSV Tech, to expand its predictive AI healthcare platform. Its proprietary AI and computer vision shift care from reactive to predictive, improving patient outcomes, caregiver efficiency, and operator performance, with reported ARR growth of 13x and partnerships including Nvidia. The funding will support U.S. nationwide launch, European expansion, and scaling of its engineering team to advance predictive intelligence in senior care and hospitals. – https://theaiinsider.tech/2025/09/11/teton-ai-raises-20m-to-reinvent-elderly-care/

Astrus Secures $8M to Accelerate AI-Driven Microchip Design

(AI Insider – 11 September 2025) Astrus, a Toronto- and Waterloo-based AI startup, raised $8M led by Khosla Ventures to build the world’s first physics-aware foundation model for chip design. Its reinforcement learning system automates the analog layout process — traditionally months of manual engineering — by generating thousands of high-quality layouts in seconds. Founded by Brad Moon and Zeyi Wang, with a team trained under AI pioneers, Astrus will use the funding to expand research, scale compute infrastructure, and deliver tools to leading semiconductor companies. – https://theaiinsider.tech/2025/09/11/astrus-secures-8m-to-accelerate-ai-driven-microchip-design/

Otovo USA Launches with $4+ Mill in Funding to Create an AI-Driven Home Energy “Power Partner” for American Homeowners

(AI Insider – 11 September 2025) William J. “John” Berger, founder of Sunnova, launched Otovo USA with $4M in initial funding led by EIC Rose Rock, introducing an integrated home energy service combining retail, service, and trading. Partnering with European leader Otovo, the company debuts in Texas with plans to expand nationwide, offering solar, batteries, EV chargers, generators, and AI-driven optimization through its Endurance™ platform. Otovo USA’s membership model provides homeowners with simplified procurement, proactive maintenance, and reliable power management, aiming to transform the U.S. residential energy market. – https://theaiinsider.tech/2025/09/11/otovo-usa-launches-with-4-mill-in-funding-to-create-an-ai-driven-home-energy-power-partner-for-american-homeowners/

Realbotix Enters Distribution Partnership With Grupo Kuo

(AI Insider – 11 September 2025) Realbotix Corp. signed an exclusive strategic partnership with Grupo Kuo to distribute its AI-powered humanoid robots in Spain and Portugal, effective August 31, 2025. Grupo Kuo will deploy Realbotix robots in top hotels and shopping malls, enhancing both guest services and advanced security through autonomous monitoring, concierge functions, and real-time incident response. The alliance positions Realbotix to accelerate adoption in Southern Europe, while Grupo Kuo leverages its 1,500-strong workforce and centralized control center in Madrid to integrate robotics into hospitality and retail security services. – https://theaiinsider.tech/2025/09/11/realbotix-enters-distribution-partnership-with-grupo-kuo/

China unveils ‘world’s first’ brain-like AI that runs 100 times faster than rivals

(Interesting Engineering – 10 September 2025) Researchers at the Chinese Academy of Sciences’ Institute of Automation in Beijing have introduced a new artificial intelligence system called SpikingBrain 1.0. Described by the team as a “brain-like” large language model, it is designed to use less energy and operate on homegrown Chinese hardware rather than chips from industry leader Nvidia. “Mainstream Transformer-based large language models (LLMs) face significant efficiency bottlenecks: training computation scales quadratically with sequence length, and inference memory grows linearly,” said the researchers in a non-peer-reviewed technical paper. – https://interestingengineering.com/innovation/china-unveil-brain-like-ai

SandboxAQ Releases AQCat25 Dataset, Accelerating Next-Generation Catalysis and Materials Discovery with AI

(AI Insider – 10 September 2025) SandboxAQ released AQCat25, a large-scale AI dataset with 11 million data points to accelerate catalyst discovery and industrial chemistry applications. The dataset incorporates spin polarization and covers 40,000 catalyst-intermediate systems, enabling accurate modeling and extending machine learning approaches to new industrially relevant problems. Generated using over 400,000 GPU-hours on NVIDIA DGX Cloud, AQCat25 is publicly available on Hugging Face for academic and industrial use in areas such as sustainable fuels, hydrogen, and fertilizer production. – https://theaiinsider.tech/2025/09/10/sandboxaq-releases-aqcat25-dataset-accelerating-next-generation-catalysis-and-materials-discovery-with-ai/

Kanematsu Invests in Intuition Robotics to Collaborate on Japanese Expansion of AI Companion Robot, ‘ElliQ’

(AI Insider – 10 September 2025) Kanematsu Corporation has partnered with Intuition Robotics to co-develop and launch a Japanese market version of ElliQ, the AI companion robot for older adults, marking the company’s first international expansion outside the U.S. Kanematsu has also invested in Intuition Robotics, bringing the company’s total equity funding to $85 million, and plans to leverage its 20,000+ partner network to introduce ElliQ to Japan in 2026. ElliQ, already used by thousands of U.S. seniors, provides health management, preventive care, reminders, and social interaction, with studies showing it reduces loneliness (94%), improves wellness (97%), and enhances quality of life (90%). – https://theaiinsider.tech/2025/09/10/kanematsu-invests-in-intuition-robotics-to-collaborate-on-japanese-expansion-of-ai-companion-robot-elliq/

Resemble AI Launches Generative AI Voice-Based Deepfake Simulation Platform, Available to the Public Sector Through Carahsoft

(AI Insider – 10 September 2025) Resemble AI and Carahsoft Technology Corp. have launched a generative AI-based simulation platform to help public sector organizations defend against voice phishing, identity fraud, and deepfake-driven social engineering. The platform uses hyper-realistic, real-time simulations—powered by Resemble’s proprietary voice models and LLM integration—to train employees through phone, WhatsApp, and email attack scenarios, assigning risk scores and highlighting organizational blind spots. Available through Carahsoft’s government contracts, the system has shown up to a 90% reduction in successful attacks, providing agencies with scalable training, compliance reporting, and protection against an estimated $2.6 billion in global deepfake fraud. – https://theaiinsider.tech/2025/09/10/resemble-ai-launches-generative-ai-voice-based-deepfake-simulation-platform-available-to-the-public-sector-through-carahsoft/

Harvard Researchers Develop New AI Tool that Pinpoints Genes, Drug Combos to Restore Health in Diseased Cells

(AI Insider – 10 September 2025) Harvard Medical School researchers have developed PDGrapher, an AI tool that identifies drug targets capable of reversing disease states in cells, potentially transforming drug discovery. Tested on 19 datasets across 11 cancer types, PDGrapher outperformed other AI models—ranking correct targets up to 35% higher and delivering results up to 25 times faster—while surfacing both known and novel treatment candidates. Backed by federal grants and industry partnerships, the model is already being applied to cancer, Parkinson’s, Alzheimer’s, and rare neurodegenerative disorders, with long-term potential for personalized combination therapies. – https://theaiinsider.tech/2025/09/10/harvard-researchers-develop-new-ai-tool-that-pinpoints-genes-drug-combos-to-restore-health-in-diseased-cells/

AI in weather forecasting takes centre stage in Abu Dhabi

(DigWatch – 10 September 2025) Abu Dhabi hosted a Weather Summit that explored how AI could transform forecasting and support operations, such as cloud seeding. Experts emphasised that AI enhances analysis but must complement, rather than replace, human judgement. Discussions focused on Earth-system forecasting using satellite datasets, IoT devices, and geospatial systems. Quality, interoperability, and equitable access to weather services were highlighted as pressing priorities. – https://dig.watch/updates/ai-in-weather-forecasting-takes-centre-stage-in-abu-dhabi – https://gulfnews.com/uae/weather/artificial-intelligence-takes-the-lead-in-weather-forecasting-cloud-seeding-1.500263700

RobCo Launches in the US With New San Francisco HQ to Meet Automation Demand

(AI Insider – 9 September 2025) RobCo, the robot-as-a-service (RaaS) automation company, expanded its U.S. presence with a new San Francisco HQ, two Austin assembly sites, and the acquisition of Rapid Robotics’ assets, team, and customers. The expansion comes amid U.S. labor shortages and reshoring efforts, with RobCo research showing 61% of U.S. manufacturers accelerating automation plans and 95% planning new deployments within three years. RobCo’s modular robots, AI-driven no-code software, and pay-as-you-go RaaS model make automation faster, more affordable, and accessible, already serving brands like BMW, Wildpack Beverage, and Linamar with backing from Sequoia, Lightspeed, and Kindred Capital. – https://theaiinsider.tech/2025/09/09/robco-launches-in-the-us-with-new-san-francisco-hq-to-meet-automation-demand/

SCOREalytics Raises $3M+ Seed Funding Amid Surge in AI-Powered Legal Intelligence Adoption by Legal Teams and Global Law Firms

(AI Insider – 9 September 2025) SCOREalytics raised over $3M in seed funding led by Moneta Ventures with participation from Ollin Ventures, aiming to scale its AI-powered legal intelligence platform and expand adoption among legal and compliance teams. The platform uses large language models and proprietary data pipelines to help enterprises and law firms detect litigation threats, regulatory changes, and compliance risks in real time. Following rapid growth with 25 new Fortune 500 clients and new product launches, SCOREalytics plans to expand its team, accelerate product development, and build the “legal command center of the future.” – https://theaiinsider.tech/2025/09/09/scorealytics-raises-3m-seed-funding-amid-surge-in-ai-powered-legal-intelligence-adoption-by-legal-teams-and-global-law-firms/

DroneDeploy Hits Break-Even, Raises $15M to Fuel Construction Progress AI Product

(AI Insider – 9 September 2025) DroneDeploy reached break-even, giving it full control over its growth strategy, and secured $15 million in strategic funding from existing investors to accelerate AI and robotics product development. The company is doubling down on new AI offerings like Progress AI and Safety AI, already used to automate site analysis and reduce costs on large-scale construction projects. Funding will also expand DroneDeploy’s robotics capabilities across quadruped, aerial, and future humanoid systems, cementing its role at the intersection of AI, robotics, and industrial automation. –  https://theaiinsider.tech/2025/09/09/dronedeploy-hits-break-even-raises-15m-to-fuel-construction-progress-ai-product/

Blattner and Built Robotics Announce Contract for Solar Construction Robots

(AI Insider – 9 September 2025) Blattner, a Quanta Services company and leader in renewable energy construction, signed a three-year deal with Built Robotics to deploy dozens of AI-powered robots on U.S. solar projects. The robots will handle tasks like pile driving, surveying, material handling, drilling, and trenching, aiming to boost efficiency while significantly improving worker safety. Built’s machines use an 8-Layer Safety System with a perfect safety record to date, and the partnership builds on prior collaborations that demonstrated both quality and accelerated project timelines. – https://theaiinsider.tech/2025/09/09/blattner-and-built-robotics-announce-contract-for-solar-construction-robots/

Autonomous Driving Startup QCraft to Open European HQ, Forms Partnership With Qualcomm

(AI Insider – 9 September 2025) QCraft announced at IAA Mobility 2025 that it will partner with Qualcomm to build next-generation assisted driving systems on the Snapdragon Ride platform, aiming to cut production costs and speed deployment for global automakers. The company also unveiled plans for a European headquarters in Germany to oversee R&D integration, certification, and business development, reinforcing its strategy for international expansion. QCraft’s Navigate on Autopilot technology is expected to roll out to about one million vehicles, with mass production beginning in 2026 across Europe, the U.S., Japan, and South Korea, positioning the company as a flexible partner supporting Qualcomm, NVIDIA, and Horizon Robotics chipsets. – https://theaiinsider.tech/2025/09/09/autonomous-driving-startup-qcraft-to-open-european-hq-forms-partnership-with-qualcomm/

Nebius Announces $17.4 Billion Data Center Deal with Microsoft

(AI Insider – 9 September 2025) Nebius Group (NASDAQ: NBIS) signed a multi-year agreement to supply Microsoft with dedicated AI infrastructure capacity from its new Vineland, New Jersey data center starting later this year. CEO Arkady Volozh said the contract is the first of several expected long-term deals with major AI labs and tech companies, adding that it strengthens Nebius’s core AI cloud business. The company will finance related capital expenditures through cash flow from the deal and debt secured against the contract, while exploring additional financing options to accelerate growth. – https://theaiinsider.tech/2025/09/09/nebius-announces-17-4-billion-data-center-deal-with-microsoft/

Superconducting qubits power Stanford’s quantum router advance

(DigWatch – 9 September 2025) Quantum computers could become more efficient with a new quantum router that directs data more quickly within machines. Researchers at Stanford have built the component, which could eventually form the backbone of quantum random access memory (QRAM). The router utilises superconducting qubits, controlled by electromagnetic pulses, to transmit information to quantum addresses. Unlike classical routers, it can encode addresses in superposition, allowing data to be stored in two places simultaneously. – https://dig.watch/updates/superconducting-qubits-power-stanfords-quantum-router-advance

AI-tissue collaboration could transform drug trials and precision medicine

(DigWatch – 9 September 2025) Researchers combine human tissue models with explainable AI to analyse patient data and identify treatments that work best for specific patients. First applied to inflammatory bowel disease, the approach could improve clinical trial success rates and accelerate drug discovery. REPROCELL, IBM, and the STFC Hartree Centre have developed Pharmacology-AI, a platform uniting tissue models with machine learning. Delivered through the Hartree National Centre for Digital Innovation, it reduces costs, enhances trial design, and enables more targeted therapies. – https://dig.watch/updates/ai-tissue-collaboration-could-transform-drug-trials-and-precision-medicine – https://www.drugtargetreview.com/article/185362/ai-meets-human-tissue-to-fast-track-precision-medicine-development/

AI Mode in Google Search adds multilingual support to Hindi and four more languages

(DigWatch – 9 September 2025) Google has announced an expansion of AI Mode in Search to five new languages, including Hindi, Indonesian, Japanese, Korean and Brazilian Portuguese. The feature was first introduced in English in March and aims to compete with AI-powered search platforms such as ChatGPT Search and Perplexity AI. – https://dig.watch/updates/ai-mode-in-google-search-adds-multilingual-support-to-hindi-and-four-more-languages

Why accessibility might be AI’s biggest breakthrough

(Ben Edwards – Ars Technica – 9 September 2025) While tech companies market AI as a productivity tool for everyone, a UK government study reveals an unexpected result: Neurodiverse employees may be benefiting far more from chatbots than their neurotypical colleagues. The UK’s Department for Business and Trade recently released evaluation results from its Microsoft 365 Copilot trial showing that while overall satisfaction was 72 percent, neurodiverse employees reported statistically higher satisfaction (at a 90 percent confidence level) and were more likely to recommend the tool (at a 95 percent confidence level) than other respondents. “It’s leveled the playing field,” one participant with ADHD told researchers during follow-up interviews. One user with dyslexia said that the tool “empowered” them to perform tasks with confidence they previously lacked, particularly in report writing. Another dyslexic participant drew direct comparisons to existing accessibility software, noting that Copilot “does a hell of a lot more” than traditional assistive technology while being “embedded in your applications” rather than requiring separate programs. – https://arstechnica.com/information-technology/2025/09/study-finds-neurodiverse-workers-more-satisfied-with-ai-assistants/

What Are The Top Universities For Students Interested in AI? Study Shows Stanford, Princeton and MIT Among Leaders Driving Global AI Research

(AI Insider – 8 September 2025) U.S. universities — including Stanford, Princeton, MIT, and Johns Hopkins — dominated the list of top U.S. universities for students who want to study AI. A new study by Studocu analyzed global academic output to identify the universities and countries driving the most influential AI research. The United States led with over 232,000 AI-related publications in the past four years, followed by China with 217,000 and the United Kingdom with more than 109,000. – https://theaiinsider.tech/2025/09/08/what-are-the-top-universities-for-students-interested-in-ai-study-shows-stanford-princeton-and-mit-among-leaders-driving-global-ai-research/

XiFin Accelerates AI-Driven RCM Future with New Growth Funding and Addition of Visionary Leader Jeff Margolis to Its Board

(AI Insider – 8 September 2025) XiFin secured new growth capital led by Goldman Sachs External Investing Group and Avista Healthcare Partners, following 19% year-over-year growth driven by adoption of its AI-powered RCM platform, XiFin® Empower. The funding will accelerate innovation in revenue cycle management, unifying automation, data analytics, and AI to improve financial performance, streamline workflows, and enhance patient and provider experiences. Healthcare IT veteran Jeff Margolis, founder of TriZetto and Senior Advisor to Blackstone, has joined XiFin’s Board of Directors, bringing over 35 years of leadership in healthcare technology and AI innovation. – https://theaiinsider.tech/2025/09/08/xifin-accelerates-ai-driven-rcm-future-with-new-growth-funding-and-addition-of-visionary-leader-jeff-margolis-to-its-board/

Polestar Analytics Raises $12.5M to Advance AI Capabilities & 1Platform for Converged Data Ecosystem

(AI Insider – 8 September 2025) Polestar Analytics has raised $12.5M in growth capital to accelerate R&D in artificial intelligence and strengthen its proprietary 1Platform for enterprise-scale analytics. The company appointed veteran telecom and tech executive Michel Combes as Chair of the Board, bringing decades of global leadership experience at firms including SoftBank, Sprint, Altice, Alcatel-Lucent, and Vodafone Europe. With fresh investment and seasoned leadership, Polestar Analytics aims to scale globally, enhance its AI capabilities, and deliver advanced data convergence and analytics solutions to enterprises worldwide. – https://theaiinsider.tech/2025/09/08/polestar-analytics-raises-12-5m-to-advance-ai-capabilities-1platform-for-converged-data-ecosystem/

X Square Robot Raises USD $140M in Series A+ Round Led By Alibaba Cloud

(AI Insider – 8 September 2035) X Square Robot secured nearly RMB 1 billion in Series A+ funding co-led by Alibaba Cloud and CAS Investment, marking Alibaba Cloud’s first investment in embodied AI and signaling rising strategic interest in the sector. The startup, founded in late 2023, has developed breakthroughs including the WALL-A Vision–Language–Action model with zero-shot task generalization, an embodied chain-of-thought framework, and the open-sourced WALL-OSS foundation model. Funds will accelerate training of large embodied AI models, advance hardware like the Quanta X2 humanoid robot, and expand real-world deployments across industrial and service applications. – https://theaiinsider.tech/2025/09/08/x-square-robot-raises-usd-140m-in-series-a-round-led-by-alibaba-cloud/

Hypershell Launches Outdoor Exoskeleton

(AI Insider – 8 September 2025) Hypershell launched the X Ultra, its next-generation outdoor exoskeleton and IFA Innovation Award 2025 winner, designed to enhance human performance across hiking, running, cycling, and climbing, available globally starting today. Powered by the AI MotionEngine Ultra and new M-One Ultra Motors, the exoskeleton reduces exertion by up to 39% in cycling and 20% in walking, while offering 42,000 steps per battery and a range of 60 km with dual batteries. Featuring five intelligent modes, watchOS integration, and validation from SGS, the lightweight 1.8 kg device aims to democratize exoskeleton technology by delivering lab-verified performance in a consumer-ready form factor. – https://theaiinsider.tech/2025/09/08/hypershell-launches-outdoor-exoskeleton/

Perplexity For Government Launched

(AI Insider – 8 September 2025) Perplexity launched Perplexity for Government, offering federal users free access to its most advanced AI models with “secure-by-default” protections that prevent government data from being used to train systems. The company also introduced Enterprise Pro for Government, a tailored version of its enterprise platform priced at $0.25 per agency for the first 15 months, with integration of both Perplexity and third-party models. The initiative aligns with the Trump administration’s AI Action Plan and GSA’s OneGov strategy, as Perplexity positions itself as a trusted AI provider for U.S. agencies amid rising pressure for secure, frontier AI adoption. – https://theaiinsider.tech/2025/09/08/perplexity-for-government-launched/

Living Optics Showcases Real-Time Hyperspectral Imaging on NVIDIA Jetson

(AI Insider – 8 September 2025) Living Optics unveiled a real-time blood perfusion demo on NVIDIA’s Jetson platform, showcasing how its hyperspectral imaging (HSI) can distinguish oxygenated from deoxygenated blood at the edge for surgical, triage, and monitoring use. The company’s compact, software-defined platform brings lab-grade spectral vision to healthcare, robotics, food inspection, and industrial applications, while meeting the power and size constraints of real-world edge systems. Backed by $25 million in Series A funding, the Oxford spinout is expanding globally, positioning its spectral intelligence as a transformative tool for machine vision across industries from healthcare to environmental monitoring. – https://theaiinsider.tech/2025/09/08/living-optics-showcases-real-time-hyperspectral-imaging-on-nvidia-jetson/

U-M to Lead $19.4 Million DOE-Funded AI Project for Hypersonic Flight

(AI Insider – 8 September 2025) The University of Michigan will lead a $19.4 million Department of Energy–funded center to use artificial intelligence for solving complex physics problems, with applications in hypersonic flight. The Center for Prediction, Reasoning and Intelligence for Multiphysics Exploration (C-PRIME) will develop AI agents that build trustworthy physics models, run simulations on supercomputers, and support the design of rotating detonation combustors. The project involves 13 Michigan investigators, partners from Princeton University, and collaborations with Sandia, Los Alamos, and Lawrence Livermore national laboratories. – https://theaiinsider.tech/2025/09/08/u-m-to-lead-19-4-million-doe-funded-ai-project-for-hypersonic-flight/

Quantum Leaders: Quantum is Moving From Lab to The Marketplace

(Quantum Insider – 8 September 2025) Quantum computing is moving from research into commercial use, with experts emphasizing it will complement rather than replace classical high-performance computing. Panelists from Fujitsu, Oxford Instruments, Quantinuum, and D-Wave highlighted early applications including error-corrected qubits, cryogenic infrastructure, verifiable randomness for cybersecurity, and annealing for optimization. The discussion stressed that adoption depends on hybrid integration, accessibility, and education, with the International Year of Quantum Science and Technology aiming to raise awareness and reduce skepticism. – https://thequantuminsider.com/2025/09/08/quantum-leaders-quantum-is-moving-from-lab-to-the-marketplace/

Superradiance Discovery Extends Quantum Entanglement Range 17-Fold

(Quantum Insider – 7 September 2025) Researchers from the University of Namur, Harvard, Michigan Technological University, and Sparrow Quantum have shown that near-zero refractive index materials enable long-range quantum superradiance, advancing prospects for scalable quantum technologies. Their theoretical model demonstrates a photonic chip that can extend entanglement between emitters up to 17 times farther than in a vacuum, using nitrogen vacancy diamonds. The work suggests practical applications in quantum computing, secure telecommunications, and advanced sensing, while highlighting the need to move from simulations to experimental realizations. – https://thequantuminsider.com/2025/09/07/superradiance-discovery-extends-quantum-entanglement-range-17-fold/

Ronovo Surgical Closes Series D Financing

(AI Insider – 6 September 2025) Ronovo Surgical closed a $67 million Series D financing round led by Johnson & Johnson’s JJDC, with participation from Lilly Asia Ventures, INCE Capital, and Granite Asia, bringing its 2025 fundraising total to over $100 million. The company signed a collaboration with Johnson & Johnson Medical (Shanghai) to expand access to Carina™, its flagship modular robotic surgery platform, in select China markets alongside Johnson & Johnson MedTech technologies. Carina™ became the first modular surgical robotic system approved by China’s NMPA across four specialties and is now advancing global expansion with regulatory submissions in Europe and South America. – https://theaiinsider.tech/2025/09/06/ronovo-surgical-closes-series-d-financing/

Why Do Language Models Hallucinate? OpenAI Scientists Say LLMs Rewarded For Being Too Cocky

(AI Insider – 6 September 2025) An OpenAI team of scientists report that language models hallucinate because their training and evaluation processes reward confident guesses over admitting uncertainty. Hallucinations are predictable statistical errors that arise during pretraining and persist because benchmarks penalize responses that express doubt. Fixing the problem requires changing mainstream evaluations to credit uncertainty, aligning incentives toward more trustworthy AI systems. – https://theaiinsider.tech/2025/09/06/why-do-language-models-hallucinate-openai-scientists-say-llms-rewarded-for-being-too-cocky/