Governance, Legislation, and Geostrategies
The United Nations calls for urgent regulation of military AI
(Digital Watch Observatory – 23 May 2025) The UN and global experts have emphasised the urgent need for comprehensive regulation of AI in military applications. UN Secretary has called for ‘global guardrails’ to govern the use of autonomous weapons, warning that rapid technological development has outpaced current policies. – https://dig.watch/updates/the-united-nations-calls-for-urgent-regulation-of-military-ai
Australia’s AI ambitions hinge on collaboration with China
(East Asia Forum – 23 May 2025) Australia faces a challenging path forward for AI development amid the US–China tech rivalry. Long-standing alliances with the US sit awkwardly with extensive economic ties with China, with the US putting pressure on Australia to untangle research collaborations with China. This pressure could compromise academic freedom and disrupt valuable research partnerships that are integral to Australia’s AI development. Australia must leverage artificial intelligence advancements through international partnerships, without sacrificing its technological independence. – https://eastasiaforum.org/2025/05/23/australias-ai-ambitions-hinge-on-collaboration-with-china/
Hong Kong approves stablecoin licensing law
(Digital Watch Observatory – 23 May 2025) Hong Kong’s legislature has approved a bill introducing a licensing framework for fiat-referenced stablecoin issuers. The move provides legal clarity and aims to enhance the city’s position as a global digital asset hub. Any issuer of stablecoins in Hong Kong or of HKD-backed stablecoins abroad must obtain a licence from the Hong Kong Monetary Authority. The law outlines standards for reserve asset management, redemption, and risk controls to protect investors and wider public. – https://dig.watch/updates/hong-kong-approves-stablecoin-licensing-law
The AI Workforce: What LinkedIn data reveals about ‘AI Talent’ trends in OECD.AI’s live data
(Rosie Hood – OECD.AI – 22 May 2025) LinkedIn’s Data for Impact programme’s collaboration with the OECD.AI Policy Observatory brings together the latest LinkedIn data on AI workforce trends to provide policymakers with information to understand the AI ecosystem better and make evidence-based decisions. LinkedIn’s more than one billion members represent an economic graph that is a granular representation of the global economy updated in real time. While administrative systems can lag in updating taxonomies on AI occupations and skills, LinkedIn’s data identifies the latest trends in AI and the labour market. LinkedIn’s latest methodology considers a member as AI engineering talent if they have explicitly added at least two AI Engineering Skills (see list below) – such as model training, AI agents, or large language models – to their profile and/or they are or have been employed in an AI job like an AI engineer. By capturing occupational and skills components, this refined definition represents the wider AI workforce more clearly and incorporates ongoing efforts to upskill and utilise AI across multiple sectors. – https://oecd.ai/en/wonk/ai-workforce-what-linkedin-data-reveals-about-ai-talent-trends-in-oecd-ai-live-data
Judges Shouldn’t Rely on AI for the Ordinary Meaning of Text
(Justin Curl, Peter Henderson, Kart Kandula, Faiz Surani – Lawfare – 22 May 2025) Judges are debating how large language models (LLMs) should fit into judicial work. One popular idea is to consult LLMs for the “ordinary meaning” of text, a key issue in statutory interpretation. At first glance, this may seem promising: These models, trained on massive amounts of human language, should reflect everyday usage. Several critiques, however, have already highlighted technological flaws with this conception, including inaccuracies introduced by nonrepresentative training data and privacy risks posed by use of third-party artificial intelligence (AI) services. But valid as these concerns are, they obscure a more fundamental issue: LLMs embody the discretionary choices of their creators, not the median of human language. Their responses are controlled by private entities, each with their own economic interests and ideological commitments. LLMs are powerful tools that will likely have many applications in the judiciary—for instance, retrieving historical materials relevant to determining meaning, or assisting with legal drafting. But they are not themselves neutral arbiters of ordinary meaning, and judges should not treat them as such. Relying on models in this way risks delegating core interpretive functions to actors with incentives to manipulate judicial outcomes. This transfer of influence to unaccountable private interests represents not merely a technical challenge requiring judicial caution, but a structural incompatibility with the judicial role. – https://www.lawfaremedia.org/article/judges-shouldn-t-rely-on-ai-for-the-ordinary-meaning-of-text
BRICS conclude Understanding on Data Economy Governance
(Digital Watch Observatory – 22 May 2025) Trade Ministers from BRICS countries convened in Brasília to exchange views on trade and investment challenges. They issued a Joint Declaration, expressing strong support to the multilateral trading system centred on the WTO, and emphasising the need to strengthen it against protectionist measures and fragmentation. The Ministers approved the BRICS Declaration on WTO Reform and Strengthening of the Multilateral Trading System (Annex I of the Joint Declaration), expressing serious concerns about rising unilateral tariff and non-tariff measures inconsistent with WTO rules. A significant outcome was the approval of the BRICS Data Economy Governance Understanding (Annex II). This document acknowledges the key role of data for economic growth, innovation, and public governance, while also noting the challenges raised by lack of statistical data and regulatory fragmentation, hindering developing countries’ integration into the global digital economy. – https://dig.watch/updates/brics-concludes-understanding-on-data-economy-governance
HBKU Launches Qatar’s First Quantum Computing Laboratory Under QC2 Initiative
(Quantum Insider – 22 May 2025) Hamad Bin Khalifa University launched Qatar’s first quantum computing lab under the Qatar Center for Quantum Computing, supported by a $10 million grant from the Ministry of Defence. The lab will focus on applied research in quantum algorithms, error correction, cryptography, and hardware benchmarking to advance quantum information science. Led by Dr. Saif Al-Kuwari, the lab is intended to develop local expertise and support national applications in sectors such as energy, cybersecurity, and transportation. – https://thequantuminsider.com/2025/05/22/hbku-launches-qatars-first-quantum-computing-laboratory-under-qc2-initiative/
Taiwan targets Facebook scam ads with new penalties
(Digital Watch Observatory – 22 May 2025) Taiwan’s Ministry of Digital Affairs plans to impose penalties on Meta for failing to enforce real-name verification on Facebook ads, according to Minister Huang Yen-nan. The move follows a recent meeting with law enforcement and growing concerns over scam-related losses. A report from CommonWealth Magazine found Taiwanese victims lose NT$400 million (US$13 million) daily to scams, with 70% of losses tied to Facebook. Facebook has been the top scam-linked platform for two years, with over 60% of users reporting exposure to fraudulent content. – https://dig.watch/updates/taiwan-targets-facebook-scam-ads-with-new-penalties
Pakistan sets up digital asset authority
(Digital Watch Observatory – 22 May 2025) Pakistan has formed a new body to regulate its growing digital asset market and embrace blockchain-based financial innovation. The Pakistan Digital Assets Authority (PDAA), backed by the Ministry of Finance, will license and oversee exchanges, custodians, wallets, stablecoins, and decentralised finance platforms. Federal finance minister Muhammad Aurangzeb said the goal is not only to catch up but to lead the sector globally. PDAA will also tokenise national assets and government debt, and monetise excess electricity through regulated Bitcoin mining. – https://dig.watch/updates/pakistan-sets-up-digital-asset-authority
United Nations’ ITU Launches ‘Quantum for Good’ to Align Innovation with Global Impact
(Quantum Insider – 21 May 2025) The International Telecommunication Union (ITU) will launch Quantum for Good this July as a new track at the AI for Good Global Summit, designed to align quantum innovation with global development priorities. The initiative focuses on equity, access, and impact, working to scale real-world quantum use cases and build inclusive standards, education pipelines, and international partnerships. Speakers from IBM, Algorithmiq, QuIC, and the UN will lead sessions on industry innovation, diplomacy, ethics, and standards, framing quantum’s role in shaping inclusive global futures. Quantum for Good marks the beginning of a long-term global effort, and ITU invites stakeholders across sectors to participate, collaborate, and help ensure quantum benefits all. – https://thequantuminsider.com/2025/05/21/united-nations-itu-launches-quantum-for-good-to-align-innovation-with-global-impact/
Revitalizing Corporate Governance for the Quantum Age
(Michael Siebecker – Just Security – 21 May 2025) As quantum computing and AI advance, the legal framework governing corporate fiduciary duties requires significant recalibration. The prevailing fiduciary standards of care and oversight — rooted in doctrines developed for a radically different informational and technological context — largely permit corporate directors to escape liability for errors in judgment (even when significant damage or harm results) absent “gross negligence” in the decision-making process or an “utter failure” to monitor corporate operations. Today, company directors now have access to highly sophisticated analytical tools. AI systems, such as IBM’s Financial Crimes Insight for Alert Triage or Oracle’s newly announced Automated Scenario Calibration Cloud Service, can detect financial anomalies and flag regulatory risks across an enterprise in real time. Quantum computing, as it becomes commercially viable, will allow boards to simulate high-dimensional strategic scenarios that would overwhelm traditional models. These technologies materially alter the scope of what information is reasonably available to boards. When boards fail to engage with widely accessible tools capable of identifying and modeling operational risks and strategies in real time, the resulting failure is no longer one of ignorance, but of blind neglect. Fiduciary corporate governance principles must evolve accordingly to keep pace with the coming quantum age. – https://www.justsecurity.org/113334/revitalizing-corporate-governance-quantum-age/
ENISA unveils cyber stress testing handbook to strengthen critical infrastructure resilience under NIS2
(Digital Watch Observatory – 21 May 2025) The European Union Agency for Cybersecurity (ENISA) has released a Handbook for Cyber Stress Testing to support national and sectoral authorities in assessing the cybersecurity and resilience of critical infrastructure, in line with the NIS2 Directive. The guidance is intended for use at the national, regional, and EU levels and complements regulatory frameworks such as the Digital Operational Resilience Act (DORA) and the Critical Entities Resilience (CER) directive. – https://dig.watch/updates/enisa-unveils-cyber-stress-testing-handbook-to-strengthen-critical-infrastructure-resilience-under-nis2
Business Software Group Urges Congress to Reauthorize Quantum Initiative
(Quantum Insider – 21 May 2025) The Business Software Alliance is urging Congress to reauthorize and expand the National Quantum Initiative Act to keep the U.S. competitive in the global race to deploy quantum technologies. In a letter to lawmakers, BSA warned that other countries are moving beyond research toward commercialization and national security uses, calling for a federal Special Coordinator for Quantum Leadership. The group emphasized a “whole-of-nation” strategy to integrate quantum tech into the economy, train a skilled workforce, and build resilient supply chains to support industry adoption. – https://thequantuminsider.com/2025/05/21/business-software-group-urges-congress-to-reauthorize-quantum-initiative/
Invest Qatar Partners With Quantinuum to Advance Region’s Quantum Computing Ecosystem
(Quantum Insider – 21 May 2025) Invest Qatar has signed a strategic agreement with Quantinuum to expand Qatar’s quantum computing ecosystem and accelerate local adoption of advanced quantum technologies. The partnership will provide Quantinuum with access to key stakeholders, research institutions, and national initiatives, while promoting education, training, and R&D collaboration in Qatar. Quantinuum will support Qatar’s tech development through seminars, workshops, joint research, and internships, as part of broader U.S.–Qatar ties and regional innovation goals. – https://thequantuminsider.com/2025/05/21/invest-qatar-partners-with-quantinuum-to-advance-regions-quantum-computing-ecosystem/
Supreme Court pushes India to regulate crypto
(Digital Watch Observatory – 21 May 2025) India’s Supreme Court has urged the government to regulate cryptocurrencies, citing a gap between taxing digital assets and failing to govern them properly. The court raised concerns about the economic risks posed by unregulated crypto activity, particularly Bitcoin. – https://dig.watch/updates/supreme-court-pushes-india-to-regulate-crypto
Dutch government passes law to criminalize cyber-espionage
(Daryna Antoniuk – The Record – 20 May 2025) The Dutch government has approved a law criminalizing a broader range of espionage activities, including digital espionage, in an effort to protect national security, critical infrastructure and sensitive technologies. The new legislation, passed over the weekend, extends existing espionage laws that already make it a criminal offense to share state secrets. Under the updated law, leaking sensitive information that is not classified as a state secret or engaging in activities on behalf of a foreign government that harm Dutch interests can also result in criminal charges. – https://therecord.media/netherlands-law-criminalizes-cyber-espionage
On the Path to AI Sovereignty, AI Agency Offers a Shortcut
(Aubra Anthony – Lawfare – 20 May 2025) Artificial intelligence’s (AI’s) explosive evolution from a niche field of academic study to a ubiquitous and loudly hyped headline has created a pressing sense of urgency as countries vie for superiority. No doubt intensified by the Trump administration’s aggressive “America First” posture, many countries have fundamentally reevaluated how tightly coupled they should allow their nation’s AI infrastructure to be to the whims of the U.S. or other AI superpowers. In response, many are turning to the strategic goal of sovereign AI—an approach to AI that prioritizes geopolitical independence and autonomy over a nation’s AI stack, which includes the data, compute, energy, and talent necessary to run AI. AI sovereignty, in its strictest sense, implies full ownership of a great deal of costly infrastructure—a feat that the world’s AI superpowers largely don’t currently achieve thanks to a complex global supply chain. – https://www.lawfaremedia.org/article/on-the-path-to-ai-sovereignty–ai-agency-offers-a-shortcut
The EU’s AI Power Play: Between Deregulation and Innovation
(Raluca Csernatoni – Carnegie Europe – 20 May 2025) From the outset, the European Union (EU) has positioned itself as a trailblazer in AI governance with the world’s first comprehensive legal framework for AI systems in use, the AI Act. The EU’s approach to governing artificial intelligence (AI) has been characterized by a strong precautionary and ethics-driven philosophy. This ambitious regulation reflects the EU’s long-standing approach of prioritizing high ethical standards and fundamental rights in tech and digital policies—a strategy of fostering both excellence and trust in human-centric AI models. Yet, framed as essential to keep pace with U.S. and Chinese AI giants, the EU has recently taken a deregulatory turn that risks trading away democratic safeguards, without addressing systemic challenges to AI innovation.The EU now stands at a crossroads: it can forge ahead with bold, home-grown AI innovation underpinned by robust regulation, or it can loosen its ethical guardrails, only to find itself stripped of both technological autonomy and regulatory sway. While Brussels’s recent deregulatory turn is framed as a much needed competitiveness boost, the real obstacles to Europe’s digital renaissance lie elsewhere: persistent underfunding, siloed markets, and reliance on non-EU infrastructures. – https://carnegieendowment.org/research/2025/05/the-eus-ai-power-play-between-deregulation-and-innovation?lang=en¢er=europe
America first, ethics second: The implications of Trump’s AI Executive Order
(Niusha Shafiabady, Andrew O’Neil – The Interpreter – 20 May 2025) One development that has tended to pass under the radar in Donald Trump’s “America First” agenda is the decision to revoke regulations on artificial intelligence (AI). This raises serious concerns about ethical standards and the potential misuse of cutting-edge AI technology. It also confirms that the United States is no longer willing to exercise leadership in containing the more egregious impacts of this revolutionary technology. AI development transcends borders, affecting international collaboration, ethical standards, and the balance of power in technology. The choices made in one country can have downstream effects for others, influencing how AI is harnessed for good—or misused—worldwide. Not only does AI promise rapid advances, AI also poses significant challenges to labour force composition, with some large corporations shedding large sections of their workforce based on a cost-benefit analysis of leveraging AI to undertake tasks where humans previously led. For an AI superpower like the US, any move to strip away controls over AI is especially significant. – https://www.lowyinstitute.org/the-interpreter/america-first-ethics-second-implications-trump-s-ai-executive-order
South Korea’s Digital Regulation Proposal Sparks U.S. Pushback
(Andrew Yeo – Lawfare – 20 May 2025) Although the Trump administration’s tariff proposal has sucked much of the wind from U.S.-South Korea trade discussions, a different battle—over South Korea’s proposed online platform regulation—has been brewing for more than a year. The current legislation, known as the Monopoly Regulation and Fair Trade Act (MRFTA), has drawn the ire of several major U.S. tech companies, including Google, Apple, and Meta. These companies have criticized the act as discriminatory and urged the U.S. Chamber of Commerce, industry lobby groups, Congress, and the U.S. trade representative to push back against it. The potential fight among U.S. big tech companies, South Korean regulators, and the Trump administration bears greater attention given the implications for ongoing tariff negotiations, digital trade policy, and the Trump administration’s approach to the evolving global online platform regulatory landscape. – https://www.lawfaremedia.org/article/south-korea-s-digital-regulation-proposal-sparks-u.s.-pushback
AI threatens one in four jobs – but transformation, not replacement, is the real risk
(UN News – 20 May 2025) One in four jobs worldwide is potentially exposed to what’s known as generative artificial intelligence – or Generative AI – according to a new joint study from the UN labour agency (ILO) and Poland’s National Research Institute. The study finds that transforming job descriptions, not widespread job loss, is the more likely result. Generative AI (GenAI) refers to systems that can create content such as text, images, code or data summaries in response to user prompts. As such tools become more widely used, they are expected to reshape the tasks employees perform each day. The new ILO-NASK index draws from nearly 30,000 real-world job descriptions using worker surveys, expert reviews and AI models to identify occupations most susceptible to AI driven change. “Few jobs consist of tasks that are fully automatable with current AI technology,” the authors write. “Transformation of jobs is the most likely impact of GenAI.” – https://news.un.org/en/story/2025/05/1163486
How governments are driving AI adoption for economic growth
(Andre Nakazawa, Michael Pisa – OECD.AI – 20 may 2025) Around the world, policymakers are recognising artificial intelligence’s immense potential to reshape economies, transform industries, and strengthen global competitiveness. While generative AI chatbots and image generators have captured the public’s attention, these tools are just a fraction of the sweeping economic transformation AI is poised to catalyse through increasing labour productivity, improving public services, and accelerating scientific progress. However, realising these benefits demands not only continued technological advancement but also widespread adoption and adaptation across all sectors of the economy. History suggests that countries that gain the most from innovation are not necessarily those that invent technologies first, but rather those that deploy them most effectively. – https://oecd.ai/en/wonk/how-governments-are-driving-ai-adoption-for-economic-growth
Half of Japan’s municipalities adopt or plan generative AI use
(Digital Watch Observatory – 20 May 2025) Nearly 30% of local governments in Japan had introduced generative AI into their operations by December 2023, according to a government survey. When including those planning or seriously considering its use, that figure rises to 50.8%, marking a 10-point increase from the previous year. – https://dig.watch/updates/half-of-japans-municipalities-adopt-or-plan-generative-ai-use
Lords reject UK AI copyright bill again
(Digital Watch Observatory – 20 May 2025) The UK government has suffered a second defeat in the House of Lords over its Data (Use and Access) Bill, as peers once again backed a copyright-focused amendment aimed at protecting artists from AI content scraping. Baroness Kidron, a filmmaker and digital rights advocate, led the charge, accusing ministers of listening to the ‘sweet whisperings of Silicon Valley’ and allowing tech firms to ‘redefine theft’ by exploiting copyrighted material without permission. – https://dig.watch/updates/lords-reject-uk-ai-copyright-bill-again
Panama City mayor hints at Bitcoin reserve
(Digital Watch Observatory – 20 May 2025) Panama City Mayor Mayer Mizrachi has hinted at creating a Bitcoin reserve at the city level. His remarks followed a meeting with El Salvador’s Bitcoin leaders Max Keiser and Stacy Herbert. In a short post on X, he wrote ‘Bitcoin Reserve’, sparking speculation ahead of his appearance at Bitcoin 2025 in Las Vegas. – https://dig.watch/updates/panama-city-mayor-hints-at-bitcoin-reserve
ICYMI: President Trump Signs TAKE IT DOWN Act into Law
(The White House – 19 May 2025) President Donald J. Trump signed the TAKE IT DOWN Act into law — a key initiative of First Lady Melania Trump and a landmark step in the fight to protect victims of digital exploitation. – https://www.whitehouse.gov/articles/2025/05/icymi-president-trump-signs-take-it-down-act-into-law/
AI use in financial services could add to bias risks, GAO warns
(Alexandra Kelley – NextGov – 19 May 2025) The use of artificial intelligence in financial institutes’ business operations can pose data privacy and bias risks, which demand new risk management guidance from regulatory bodies, according to a new Government Accountability Office report released Monday. In documenting the benefits AI software can lend financial institutions in their operations — such as improved customer service, investment decisions and detection of threats or fraud — GAO authors found that the same AI tools can “amplify” present risks: harm to fair lending practices, threats to privacy, conflicts of interest and false information. – https://www.nextgov.com/artificial-intelligence/2025/05/ai-use-financial-services-could-add-bias-risks-gao-warns/405432/?oref=ng-homepage-river
Closing the Digital Divide in AI
(Stanford HAI – 19 May 2025) Scholars find that large language models suffer a digital divide: The ChatGPTs and Geminis of the world work well for the 1.52 billion people who speak English, but they underperform for the world’s 97 million Vietnamese speakers, and even worse for the 1.5 million people who speak the Uto-Aztecan language Nahuatl. The main culprit is data: These non-English languages lack the needed quantity and quality of data to build and train effective models. That means most major LLMs are predominantly trained using English (or other high-resource languages) data or poor-quality local language data and not attuned to the rest of the world’s contexts and cultures. The impact? Not just inconvenience, but systematic exclusion. Entire cultures and communities are being left out of the AI revolution, risk being harmed by AI-generated misinformation and bias, and lose crucial economic and educational opportunities that English speakers gain through effective technology. (…) Stanford School of Engineering Assistant Professor Sanmi Koyejo, senior author of a new policy white paper on this topic, discusses the risks of this divide and, importantly, what developers can do to close it. – https://hai.stanford.edu/news/closing-the-digital-divide-in-ai
Why ITU’s legacy still shapes our digital world
(Digital Watch Observatory – 19 May 2025) On 17 May 1865, 20 European countries came together to create the International Telecommunication Union (ITU), a response to the tedious and inefficient telegraph system that required messages to be rewritten at every border. This practical move—born not from idealism but necessity—paved the way for a global communications framework that continues to underpin today’s digital world. – https://dig.watch/updates/why-itus-legacy-still-shapes-our-digital-world
On AI Policy, Congress Shouldn’t Cut States Off at the Knees
(Katie Fry Hester, Gary Marcus – Lawfare – 19 May 2025) Artificial intelligence holds immense promise—from accelerating disease detection to streamlining services—but it also presents serious risks, including deepfake deception, misinformation, job displacement, exploitation of vulnerable workers and consumers, and threats to critical infrastructure. As AI rapidly transforms our economy, workplaces, and civic life, the American public is calling for meaningful oversight. According to the Artificial Intelligence Policy Institute, 82 percent of voters support the creation of a federal agency to regulate AI. A Pew Research Center survey found that 52 percent of Americans are more concerned than excited about AI’s potential, and 67 percent doubt that government oversight will be sufficient or timely. – https://www.lawfaremedia.org/article/on-ai-policy–congress-shouldn-t-cut-states-off-at-the-knees
UAE to host world’s biggest AI site outside the US
(Digital Watch Observatory – 19 May 2025) The United Arab Emirates will build the largest artificial intelligence infrastructure outside the United States, following a high-level meeting between UAE President Sheikh Mohamed bin Zayed Al Nahyan and President Trump in Abu Dhabi. – https://dig.watch/updates/uae-to-host-worlds-biggest-ai-site-outside-the-us
France, Germany, and the Netherlands Launch $33M Trilateral Quantum Initiative
(Quantum Insider – 19 May 2025) France, Germany, and the Netherlands launched a trilateral quantum innovation call, selecting high-impact projects with over €30 million (approximately $33.8 million) in combined funding, including €11 million (approximately $12.4 million) for French partners through France 2030. The initiative attracted 120+ applications in quantum computing, communication, and sensing, and selected projects involving leading institutions like IQM, Quandela, Alice&Bob, and Sorbonne University. The funded projects are centered around scaling quantum computing, securing quantum networks, and advancing quantum sensing, to reinforce Europe’s leadership through cross-border collaboration and commercialization. – https://thequantuminsider.com/2025/05/19/france-germany-and-the-netherlands-launch-33m-trilateral-quantum-initiative/
Congress Urged to Double Quantum Funding to Compete With China
(Quantum Insider – 19 May 2025) Lawmakers and experts are urging Congress to double federal quantum technology funding in FY2026 to remain competitive with China. The CSIS Commission on U.S. Quantum Leadership recommends the increase, noting it would amount to just 0.5% of the discretionary budget. Concerns were raised that stagnant funding and federal workforce cuts could deter private investment and weaken U.S. leadership in quantum innovation. – https://thequantuminsider.com/2025/05/19/congress-urged-to-double-quantum-funding-to-compete-with-china/
Former Director of UK’s Intelligence and Cybersecurity Agency Joins Oxford Quantum Circuits
(Quantum Insider – 19 May 2025) OQC appointed Sir Jeremy Fleming, former Director of GCHQ, to its Board, highlighting the strategic role of quantum computing in UK national security, defence, and economic growth. Sir Jeremy brings decades of leadership in intelligence and cybersecurity, including founding the NCSC and co-leading the National Cyber Force, and emphasizes quantum’s centrality to protecting the UK’s future digital infrastructure. OQC’s architecture and its recent UK Government tender win for a Quantum Error Correction testbed cements its position within secure quantum infrastructure. – https://thequantuminsider.com/2025/05/19/former-director-of-uks-intelligence-and-cybersecurity-agency-joins-oxford-quantum-circuits/
UK workers struggle to keep up with AI
(Digital Watch Observatory – 19 May 2025) AI is reshaping the UK workplace, but many employees feel unprepared to keep pace, according to a major new study by Henley Business School. While 56% of full-time professionals expressed optimism about AI’s potential, 61% admitted they were overwhelmed by how quickly the technology is evolving. – https://dig.watch/updates/uk-workers-struggle-to-keep-up-with-ai
UK to enforce strict crypto transaction reporting
(Digital Watch Observatory – 19 May 2025) Crypto firms operating in the United Kingdom will be required to report detailed customer transaction data from 1 January 2026. The move is part of the government’s wider plan to improve tax transparency in the crypto sector by aligning with international reporting standards. – https://dig.watch/updates/uk-to-enforce-strict-crypto-transaction-reporting
The case for a resounding framework for AI-biotech convergence
(Lakshmy Ramakrishnan – Observer Research Foundation – 18 May 2025) While the convergence of AI with biotechnology holds immense potential for innovative developments in science and medicine it also carries the scope for misuse. Governance of AI in popular discourse includes the risk of AI-biotechnology (AI-bio) tools being utilised for malicious purposes or the production of bioweapons. Detailed understandings need to be acquired to determine the feasibility of the development of bioweapons with AI-bio tools and to formulate safeguards against such threats. – https://www.orfonline.org/research/the-case-for-a-resounding-framework-for-ai-biotech-convergence
Japan enacts new Active Cyberdefense Law allowing for offensive cyber operations
(Alexander Martin – The Record – 16 May 2025) Japan enacted a new law that would permit the country’s authorities to preemptively engage with adversaries through offensive cyber operations to ensure threats are suppressed before they cause significant damage. The new law, which was first mooted in 2022, is intended to help Japan strengthen its cyber defense “to a level equal to major Western powers” and marks a break from the country’s traditional approach to cyber defense, which had tracked closely to its Article 9 constitutional commitment to pacifism. – https://therecord.media/japan-enacts-new-law-allowing-offensive-cyber-operations
Wuhan’s AI Development. China’s Alternative Springboard to Artificial General Intelligence (AGI)
(William Hannas, Huey-Meei Chang, and Daniel Chou – Center for Security and Emerging Technology – May 2025) Wuhan, China’s inland metropolis, is paving the way for a nationwide rollout of “embodied” artificial intelligence meant to fast-track scientific discovery, optimize production, streamline commerce, and facilitate state supervision of social activities. Grounded in real-world data, the AI grows smarter, offering a pathway to artificial “general” intelligence that will reinforce state ideology and boost economic goals. This report documents the genesis of Wuhan’s AGI initiative and its multifaceted deployment. – https://cset.georgetown.edu/publication/wuhans-ai-development/
Security
Operation ENDGAME strikes again: the ransomware kill chain broken at its source
(Europol – 23 May 2025) Cybercriminals around the world have suffered a major disruption after law enforcement and judicial authorities, coordinated by Europol and Eurojust, dismantled key infrastructure behind the malware used to launch ransomware attacks. From 19 to 22 May, authorities took down some 300 servers worldwide, neutralised 650 domains, and issued international arrest warrants against 20 targets, dealing a direct blow to the ransomware kill chain. – https://www.europol.europa.eu/media-press/newsroom/news/operation-endgame-strikes-again-ransomware-kill-chain-broken-its-source
270 arrested in global dark web crackdown targeting online drug and criminal networks
(Europol – 22 May 2025) A global law enforcement operation coordinated by Europol has struck a major blow to the criminal underground, with 270 arrests of dark web vendors and buyers across ten countries. Known as Operation RapTor, this international sweep has dismantled networks trafficking in drugs, weapons, and counterfeit goods, sending a clear signal to criminals hiding behind the illusion of anonymity. The suspects were identified through coordinated investigations based on intelligence from the takedowns of the dark web marketplaces Nemesis, Tor2Door, Bohemia and Kingdom Markets. Many had conducted thousands of sales on illicit marketplaces, using encryption tools and cryptocurrencies to cover their tracks — but law enforcement closed in. This international action follows Operation SpecTor in 2023, which led to 288 arrests. Together, these operations show the increasing ability of law enforcement to penetrate the dark web’s cloak of secrecy. – https://www.europol.europa.eu/media-press/newsroom/news/270-arrested-in-global-dark-web-crackdown-targeting-online-drug-and-criminal-networks
Experts urge stronger safeguards as jailbroken chatbots leak illegal data
(Digital Watch Observatory – 22 May 2025) Hacked AI-powered chatbots pose serious security risks by revealing illicit knowledge the models absorbed during training, according to researchers at Ben Gurion University. Their study highlights how ‘jailbroken’ large language models (LLMs) can be manipulated to produce dangerous instructions, such as how to hack networks, manufacture drugs, or carry out other illegal activities. – https://dig.watch/updates/experts-urge-stronger-safeguards-as-jailbroken-chatbots-leak-illegal-data
Asia Produces More APT Actors, as Focus Expands Globally
(Robert Lemos – Dark Reading – 21 May 2025) Advanced persistent threat (APT) actors aligned with China and North Korea now account for the majority of sophisticated attacks detected during the past two quarters, as tension in the Asia-Pacific region heats up and China increasingly flexes its muscle. While the most significant growth in attacks from the two Asian nations have focused on European targets, the Southeast Asian region has seen increases in state-backed attacks as well, with government and educational organizations among the favorite targets, according to the biannual “APT Activity Report” published by cybersecurity firm ESET. The expansion in global activity comes as more regional players are developing their own capabilities in cyber offense, with India, Taiwan, and the Philippines all seeing more attacks but developing their own capabilities as well. – https://www.darkreading.com/cyber-risk/asia-apt-actors-focus-expands-globally
New Google program targeting children with AI chatbot may violate FTC privacy rules
(Suzanne Smalley – The Record – 21 May 2025) Google’s recent launch of a program targeting its AI-powered Gemini chatbot at children under age 13 has prompted outrage from privacy and children’s rights advocates who on Wednesday told the Federal Trade Commission (FTC) the effort appears to violate the agency’s privacy rule protecting children online. The tech giant recently sent emails to parents signed up for Family Link — which allows them to create Gmail accounts for their children and offers other special services — advising them that their children can engage with the chatbot on homework or in conversation to “create stories, songs, and poetry.” – https://therecord.media/new-google-children-program-gemini
Nearly 70,000 impacted by Coinbase breach involving $20 million ransom demand
(Jonathan Greig – The Record – 21 May 2025) Cryptocurrency platform Coinbase said 69,461 people had information leaked during a data breach that began in December 2024. In documents filed with regulators in Maine on Tuesday, Coinbase said the information leaked included photos of passports, government IDs, names, dates of birth, the last four digits of Social Security numbers, bank account numbers and account information including balances and transaction history. – https://therecord.media/nearly-70000-impacted-coinbase-breach
Europol and Microsoft disrupt world’s largest infostealer Lumma
(Europol – 21 May 2025) Europol’s European Cybercrime Centre has worked with Microsoft to disrupt Lumma Stealer (“Lumma”), the world’s most significant infostealer threat. This joint operation targeted the sophisticated ecosystem that allowed criminals to exploit stolen information on a massive scale. Europol coordinated with law enforcement in Europe to ensure action was taken, leveraging intelligence provided by Microsoft. Between 16 March and 16 May 2025, Microsoft identified over 394 000 Windows computers globally infected by the Lumma malware. In a coordinated follow-up operation this week, Microsoft’s Digital Crimes Unit (DCU), Europol, and international partners have disrupted Lumma’s technical infrastructure, cutting off communications between the malicious tool and victims. In addition, over 1 300 domains seized by or transferred to Microsoft, including 300 domains actioned by law enforcement with the support of Europol, will be redirected to Microsoft sinkholes. The Head of Europol’s European Cybercrime Centre, Edvardas Šileris, said: “This operation is a clear example of how public-private partnerships are transforming the fight against cybercrime. By combining Europol’s coordination capabilities with Microsoft’s technical insights, a vast criminal infrastructure has been disrupted. Cybercriminals thrive on fragmentation – but together, we are stronger.” – https://www.europol.europa.eu/media-press/newsroom/news/europol-and-microsoft-disrupt-world%E2%80%99s-largest-infostealer-lumma
Anticipating AI’s Impact on the Cyber Offense-Defense Balance
(Andrew Lohn – Center for Security and Emerging Technology – May 2025) Artificial intelligence (AI) is beginning to change cybersecurity. This report takes a comprehensive look across cybersecurity to anticipate whether those changes will help cyber defense or offense. Rather than a single answer, there are many ways that AI will help both cyber attackers and defenders. The report finds that there are also several actions that defenders can take to tilt the odds to their favor. – https://cset.georgetown.edu/publication/anticipating-ais-impact-on-the-cyber-offense-defense-balance/
How extremists are manipulating AI chatbots
(James Paterson – The Interpreter – 21 May 2025) In an environment where people are feeling increasingly dislocated and disconnected, Artificial Intelligence (AI) chatbots have proven themselves a compelling form of social connection. Chatbots can identify our needs and our biases and, in turn, feed our desires. The more an algorithm tells us what we want to hear, as it is designed to do, the more we return to it. As younger individuals turn towards chatbots for any number of needs, including as a therapist or lover, it is entirely possible for users to develop an addiction to their AI companions, a potentially harmful addiction. Of significant concern is that this desire for companionship can be exploited by extremist actors. The open-source large language models that underpin AI chatbots can increasingly be manipulated and fine-tuned with ideological datasets to create chatbots that conform to specific worldviews, as has been demonstrated by certain far-right actors. – https://www.lowyinstitute.org/the-interpreter/how-extremists-are-manipulating-ai-chatbots
AI Risks Surge as Quantum Threats Lurk, Warns 2025 Thales Data Threat Report
(AI Insider – 21 May 2025) The 2025 Thales Data Threat Report finds that rapid generative AI adoption is outpacing security safeguards, with data integrity and trust emerging as top concerns. Nearly 70% of respondents cited the fast-moving AI ecosystem as their greatest security risk, but only 14% rated AI-specific tools as effective. While not yet imminent, quantum computing threats are advancing, with a 5,000-qubit system used to decrypt a 50-bit RSA key and post-quantum cryptography preparations now underway. – https://theaiinsider.tech/2025/05/21/ai-risks-surge-as-quantum-threats-lurk-warns-2025-thales-data-threat-report/
Frontiers
Chinese humanoids demonstrate aggressive combat skills ahead of world-first robot boxing
(Interesting Engineering – 22 May 2025) Humanoid robots in China recently demonstrated aggressive combat skills during an event at a school in Hangzhou City. The cutting-edge humanoid robots by Unitree Robotics displayed the skills before the world’s first robot boxing match, which will take place on May 25. The company brought its robots to the school to offer a live demonstration of their capabilities and a hands-on science and technology lesson for students. Robots also performed a 40-minute jogging session at the school’s basketball court, showcasing the machine’s endurance and drawing enthusiastic cheers from students. – https://interestingengineering.com/innovation/china-humanoid-robot-perform-boxing
Germany’s shape-shifting smart facade uses compressed air and AI to cut emissions
(Interesting Engineering – 22 May 2025) Scientists in Germany have unveiled a groundbreaking smart façade system that changes shape in response to weather conditions to help optimize the energy efficiency and indoor comfort of buildings, thus offering a new path toward adaptive building technologies. Named FlectoLine, the 898-square-foot (83.5-square-meter) system underwent two years of real-world testing and was recently awarded a special prize at the inaugural “Award for Bio-Inspired Innovations Baden-Württemberg.”. The innovative technology, developed through the international research project Flectuation, is the product of more than a decade of work led by the Institute of Building Structures and Structural Design (ITKE), the Institute for Textile and Fiber Technologies (ITFT) at the University of Stuttgart, and the PBG Plant Biomechanics Group at the University of Freiburg. Currently installed on a greenhouse at the Botanical Garden in Freiburg, the demonstrator consists of 101 flexible flaps, shading elements made from fiber-reinforced plastic laminates, that respond dynamically to environmental conditions using compressed air and artificial intelligence (AI). – https://interestingengineering.com/innovation/germanys-shape-shifting-smart-facade
Nuclear reactors, semiconductors to get smarter with next-gen US plasma tech
(Interesting Engineering – 22 May 2025) The US Department of Energy’s Princeton Plasma Physics Laboratory (PPPL) has developed a new simulation method that can significantly enhance both fusion research and the manufacturing of computer chips. “Researchers developed a faster, more stable way to simulate the swirling electric fields inside industrial plasmas — the kind used to make microchips and coat materials,” said PPPL in a press release. – https://interestingengineering.com/energy/us-plasma-simulation-could-drive-nuclear-fusion
Google brings sign language translation to AI
(Digital Watch Observatory – 22 May 2025) Google has introduced Gemma 3n, an advanced AI model that can operate directly on mobile devices, laptops, and tablets without relying on the cloud. The company also revealed MedGemma, its most powerful open AI model for analysing medical images and text. – https://dig.watch/updates/google-brings-sign-language-translation-to-ai
Microsoft Brings Post-Quantum Cryptography to Windows and Linux in Early Access Rollout
(Quantum Insider – 21 May 2025) Microsoft has released early-access post-quantum cryptography (PQC) tools for Windows and Linux, allowing organizations to test quantum-resistant security ahead of emerging threats. The update introduces NIST-standardized algorithms ML-KEM and ML-DSA for key exchanges and digital signatures, with support integrated into Windows Insider builds and SymCrypt-OpenSSL 1.9.0 for Linux. Microsoft encourages a hybrid implementation combining classical and quantum-safe encryption methods, citing potential performance trade-offs, compatibility concerns, and the need for crypto-agile systems as PQC standards evolve. – https://thequantuminsider.com/2025/05/21/microsoft-brings-post-quantum-cryptography-to-windows-and-linux-in-early-access-rollout/
US’ first fully digital twin nuclear reactor hits 99% accuracy in energy breakthrough
(Digital Watch Observatory – 21 May 2025) Purdue University Reactor Number One (PUR-1) is at the forefront of a nuclear energy revolution with its unique digital capabilities. This could lead to the development of a new generation of safer, more efficient, and more cost-effective nuclear reactors, ultimately accelerating the transition to carbon-free electricity. PUR-1 stands alone as the US’s first fully digitally controlled and operated reactor. This means its entire “nervous system,” the instrumentation and control system, relies on computer screens, keyboards, and ethernet cables. – https://interestingengineering.com/energy/digital-twin-nuclear-reactor-hits-99-accuracy
Egune AI Secures $3.5M to Advance National AI Infrastructure in Mongolia
(AI Insider – 20 May 2025) Egune AI, Mongolia’s leading foundation model company, has raised $3.5 million in funding to accelerate the development of its national AI infrastructure. The round was backed by Golomt Bank and will support the company’s expansion across key sectors including education, healthcare, and legal services. – https://theaiinsider.tech/2025/05/20/egune-ai-secures-3-5m-to-advance-national-ai-infrastructure-in-mongolia/
Met Office and Microsoft debut AI-ready forecasting system
(Digital Watch Observatory – 20 May 2025) The UK’s Met Office has launched a new supercomputer designed to significantly improve weather and climate forecasting accuracy. Operated via Microsoft’s Azure cloud platform, it is the world’s first cloud-based supercomputer dedicated solely to weather and climate science. – https://dig.watch/updates/met-office-and-microsoft-debut-ai-ready-forecasting-system
Taiwan boosts AI power with Nvidia supercomputer
(Digital Watch Observatory – 20 May 2025) Nvidia and Foxconn have announced plans to build a major AI factory supercomputer in Taiwan, backed by the Taiwanese government. The facility, powered by 10,000 of Nvidia’s Blackwell GPUs, will be delivered through Foxconn’s Big Innovation Company and will support researchers, start-ups, and industry. – https://dig.watch/updates/taiwan-boosts-ai-power-with-nvidia-supercomputer
UAE’s EDGE Group unveils AI Accelerator to power defence and tech
(Digital Watch Observatory – 20 May 2025) EDGE Group, a global leader in advanced technology and defence, has launched the Group AI Accelerator, a new Centre of Excellence (COE) focused on accelerating AI-driven innovation across its portfolio and facilities. The initiative is part of EDGE’s broader strategy to support the UAE’s ambitions of becoming a high-tech global hub. – https://dig.watch/updates/uaes-edge-group-unveils-ai-accelerator-to-power-defence-and-tech
Duke University Develops System for Robots to Better Navigate Rough Terrain
(AI Insider – 20 May 2025) A new sensor fusion system developed at Duke University enables quadruped robots to navigate complex terrain using a combination of vision, vibration, and touch. WildFusion integrates data from cameras, LiDAR, microphones, tactile sensors, and inertial measurement units to create a continuous 3D reconstruction of surroundings, even in visually obstructed environments. Backed by DARPA and the U.S. Army Research Laboratory, the system is designed for applications in forests, disaster zones, and remote terrain, with future enhancements to include thermal and humidity sensors. – https://theaiinsider.tech/2025/05/20/duke-university-develops-system-for-robots-to-better-navigate-rough-terrain/
Taiwan’s NCHC to Launch NVIDIA-Powered Supercomputer Supporting Quantum Research
(Quantum Insider – 20 May 2025) Taiwan’s National Center for High-Performance Computing is deploying a new AI supercomputer powered by NVIDIA hardware with over 8x the performance of the previous Taiwania 2 system. The system will accelerate research in quantum computing, climate modeling, and sovereign AI, supporting projects like Taiwan AI RAP and TAIDE for localized language model development and healthcare/education applications. Quantum research at NCHC uses NVIDIA’s CUDA-Q and cuQuantum libraries, enabling hybrid workflows, large-scale quantum simulations, and tools such as the Quantum Molecular Generator and cuTN-QSVM. NCHC plans to integrate DGX Quantum systems, strengthening Taiwan’s hybrid computing architecture and enabling broader access to researchers in academia, government, and startups through application-based system usage. – https://thequantuminsider.com/2025/05/20/taiwans-nchc-to-launch-nvidia-powered-supercomputer-supporting-quantum-research/
OpenAI, G42 plan world’s largest AI data facility
(Digital Watch Observatory – 20 May 2025) OpenAI is reportedly set to become the anchor tenant in a 5-gigawatt data centre project in Abu Dhabi, part of what could become one of the largest AI infrastructure builds globally, according to Bloomberg. The facility, spanning approximately 10 square miles, is being developed by UAE-based tech firm G42 as part of OpenAI’s broader Stargate initiative, a joint venture announced with SoftBank and Oracle to establish high-capacity AI data centres worldwide. – https://dig.watch/updates/openai-g42-plan-worlds-largest-ai-data-facility
China Telecom Launches Hybrid Quantum-Safe Encryption System, Completes 1,000-Kilometer Secure Call
(Quantum Insider – 20 May 2025) China Telecom Quantum Group has launched what it claims is the world’s first distributed cryptography system resistant to quantum computer attacks, integrating both quantum key distribution (QKD) and post-quantum cryptography (PQC). The system successfully enabled a 1,000-kilometer quantum-encrypted phone call between Beijing and Hefei and is being rolled out across 16 cities, forming a nationwide backbone of quantum-secure communication. Hefei now hosts the world’s most extensive metropolitan quantum communication network, and new platforms like Quantum Secret and Quantum Cloud Seal are already in use by hundreds of government agencies and state-owned enterprises. – https://thequantuminsider.com/2025/05/20/china-telecom-launches-hybrid-quantum-safe-encryption-system-completes-1000-kilometer-secure-call/
IQM Quantum Computers to Supply Finland With a Superconducting 300-Qubit Quantum Computer
(Quantum Insider – 20 May 2025) IQM Quantum Computers has been selected by VTT as the innovation partner to deliver a 300-qubit superconducting quantum computer, positioning Finland as a global leader in quantum technology. The partnership will deliver a 150-qubit system by mid-2026 and a 300-qubit system by late 2027, both integrated into Finland’s high-performance computing infrastructure with cloud access for researchers and businesses. The project includes joint hardware development, time-sharing for universities and companies, and a focus on advancing Finland’s expertise in quantum-aided material modelling and fabrication technologies. – https://thequantuminsider.com/2025/05/20/iqm-quantum-computers-to-supply-finland-with-a-superconducting-300-qubit-quantum-computer/
Sorbonne University And Qubit Pharmaceuticals Unveil Powerful Quantum AI Model For Medical Discovery
(Quantum Insider – 20 May 2025) Qubit Pharmaceuticals and Sorbonne University have unveiled FeNNix-Bio1, a quantum AI foundation model that simulates molecular behavior with unprecedented accuracy and efficiency, accelerating drug discovery and reducing lab experimentation costs. Unlike previous models, FeNNix-Bio1 can model dynamic molecular interactions, such as covalent drug binding and protein flexibility, enabling in silico design of complex therapies for diseases like cancer. The scalable model extends beyond pharmaceuticals to broader chemistry applications, including industrial enzymes, green chemistry, and quantum data-driven molecular simulation, marking a step toward automated molecular design. – https://thequantuminsider.com/2025/05/20/sorbonne-university-and-qubit-pharmaceuticals-unveil-powerful-quantum-ai-model-for-medical-discovery/
What’s The Difference Between AI Agents And Agentic AI? New Study Separates Signal From Noise in the AI Agent Boom
(AI Insider – 19 May 2025) A new study from Cornell University and the University of the Peloponnese offers a comprehensive framework distinguishing AI Agents from Agentic AI, highlighting structural and operational differences that carry major implications for how intelligent systems are built and applied. AI Agents are described as reactive, tool-using programs optimized for single tasks, while Agentic AI refers to coordinated multi-agent systems capable of breaking down complex goals and adapting dynamically. The study also outlines limitations of Agentic AI, including coordination complexity, error propagation, and emergent behavior, and calls for future work on memory systems, orchestration protocols, and use-case alignment. – https://theaiinsider.tech/2025/05/19/whats-the-difference-between-ai-agents-and-agentic-ai-new-study-separates-signal-from-noise-in-the-ai-agent-boom/
World’s first AI nurse? Nurabot joins Taiwan hospitals to battle healthcare crisis
(Interesting Engineering – 19 May 2025) When nurses burn out, robots clock in — and Taiwan’s hospitals are showing the world how it’s done. Amid a grim World Health Organization warning of a 4.5 million nurse shortage by 2030 due to burnout, Foxconn and NVIDIA are wiring up Taiwan’s hospitals with AI-powered robots. Meet Nurabot — a collaborative nursing robot built to combat burnout by taking over some of the most taxing tasks in clinical care. – https://interestingengineering.com/health/foxconn-nvidia-drive-ai-robotics-innovation
US team develops world’s fastest quantum switch to supercharge AI by 1 million times
(Interesting Engineering – 19 May 2025) The speed of modern processors has long hit a ceiling. While software in fields like AI and supercomputing advances rapidly, hardware struggles to keep pace. Traditional silicon-based transistors can only switch so fast, limited by the physical materials and electrical resistance on which they rely. But a new breakthrough may change that entirely by ditching electrons for light. In a major leap toward ultrafast computing, a team of researchers from the University of Arizona and international collaborators have developed a light-powered transistor using graphene. – https://interestingengineering.com/innovation/us-develops-petahertz-transistor-ai-breakthrough
AI helps CERN physicists charm Higgs boson into revealing rare decay
(Interesting Engineering – 19 May 2025) Researchers at CERN have used artificial intelligence (AI) to explore one of the Higgs boson’s most elusive behaviors, shedding light on its subtle interaction with charm quarks and bringing science closer to understanding how the so-called God particle gives matter its mass. Discovered in 2012 at the Large Hadron Collider (LHC), the Higgs boson, the fundamental force-carrying particle of the Higgs field, plays a crucial role in the Standard Model of particle physics by interacting with particles like quarks to bestow mass. – https://interestingengineering.com/science/ai-helps-cern-physicists-charm-higgs-boson-into-revealing-rare-decay
China launches first AI satellites in orbital supercomputer network
(Digital Watch Observatory – 19 May 2025) China has launched the first 12 satellites in a planned network of 2,800 that will function as an orbiting supercomputer, according to Space News. Developed by ADA Space in partnership with Zhijiang Laboratory and Neijang High-Tech Zone, the satellites can process their own data instead of relying on Earth-based stations, thanks to onboard AI models. – https://dig.watch/updates/china-launches-first-ai-satellites-in-orbital-supercomputer-network
Researchers believe AI transparency is within reach by 2027
(Digital Watch Observatory – 19 May 2025) Top AI researchers admit they still do not fully understand how generative AI models work. Unlike traditional software that follows predefined logic, gen AI models learn to generate responses independently, creating a challenge for developers trying to interpret their decision-making processes. Dario Amodei, co-founder of Anthropic, described this lack of understanding as unprecedented in tech history. Mechanistic interpretability — a growing academic field — aims to reverse engineer how gen AI models arrive at outputs. – https://dig.watch/updates/researchers-believe-ai-transparency-is-within-reach-by-2027
Humanoid robots clash in tug of war, pull cart, open doors to build resilience
(Interesting Engineering – 19 May 2025) Researchers have developed a control system that helps make humanoid robots better handle tough, real-world tasks. Named FALCON, and developed by a team at Carnegie Mellon University (CMU), it helps robots walk steadily and use their arms to push, pull, and lift objects, even when external forces push back. FALCON uses two separate AI agents—one to manage stable walking and the other to control arm movements with force-aware precision. Trained together in simulation, these agents help robots perform complex tasks like pulling carts or opening doors. According to researchers at CMU’s Learning and Control for Agile Robotics Lab (LeCAR), FALCON works across different robot designs without special tuning, making real-world deployment easier and more reliable. – https://interestingengineering.com/innovation/video-humanoid-robots-clash-in-tug-of-war
Defense, Intelligence, and Warfare
AI ‘unchained’: NGA’s Maven tool ‘significantly’ decreasing time to targeting, agency chief says
(Theresa Hitchens – Breaking Defense – 22 May 2025) As the National Geospatial-Intelligence Agency (NGA) rolls out its AI-powered imagery analysis software system Maven to more and more users, it also is seeing a sizable reduction in targeting times in exercises, according to the agency’s director. “NGA Maven has decreased targeting workflow timelines by a substantial amount, with one of our fighting element’s targeting cells seeing intelligence operation timelines drop from hours to minutes — from sensing to target engagement — during a recent exercise,” Vice Adm. Frank Whitworth told the annual GEOINT Symposium in St. Louis, Mo. on Wednesday. – https://breakingdefense.com/2025/05/ai-unchained-ngas-maven-tool-significantly-decreasing-time-to-targeting-agency-chief-says/
An 18th-century war power resurfaces in cyber policy talks
(David DIMolfetta – Defense One – 22 May 2025) Might a centuries-old war power be the key to U.S. cyber retaliation? In recent closed-door discussions, Trump administration and industry officials have discussed whether modern-day letters of marque—once used to deputize privately owned ships to lawfully attack other vessels during wartime—might enable private-sector hacking operations against unfriendly nation-states, according to two people familiar with the matter who spoke on the condition of anonymity to freely discuss the sensitive deliberations. The high-level view among U.S. officials is that this old-world maritime authority is unlikely to be directly used in cyberspace, but a more modern, tailored version might arise as the administration seeks ways to even the fight against Chinese-backed groups, one of the people said. “The general consensus from [U.S. government] officials on the topic is that we aren’t going to apply a 200-year-old [privateering] authority to the cyber domain,” said the person. “However, there is a standing question and ongoing debate regarding what modern authorities and authorizations are required by various cybersecurity and tech industry organizations to better enable the defense of the United States.”. Letters of marque played a major role during the War of 1812, when the U.S. government issued them to private ship-owners to capture British vessels. And even further back, they were used to convert pirates into privateers, acting on behalf of their sponsoring governments to raid enemy ships. Privateering was broadly outlawed by the U.S. and other signatories to an 1858 treaty, although the Confederacy used them during the Civil War. – https://www.defenseone.com/policy/2025/05/18th-century-war-power-resurfaces-cyber-policy-talks/405541/
For DOD, the future of large language models is smaller
(Patrick Tucker – Defense One – 22 May 2025) The U.S. military is working on ways to get the power of cloud-based, big-data AI in tools that can run on local computers, draw upon more focused data sets, and remain safe from spying eyes, officials from OpenAI, Scale AI, and U.S. European Command told Defense One, part of a special broadcast airing Thursday. When civilians query OpenAI’s ChatGPT, for example, their answers are generated by machines trained on huge amounts of data acquired from third-party sources and millions of user interactions a day. But the government has also hired OpenAI to make need-specific tools built on smaller and more unique datasets, said Sasha Baker, OpenAI’s head of national security policy. – https://www.defenseone.com/technology/2025/05/dod-future-large-language-models-smaller/405539/?oref=d1-featured-river-secondary
US Army upgrades Cold War missiles to hunt modern drones with new propulsion tech
(Interesting Engineering – 22 May 2025) The FIM-92 Stinger, a combat-proven Cold War-era short-range air defense missile, has been a mainstay in the US Army’s arsenal since 1981. Known for its versatility, the shoulder-fired missile operates in both surface-to-air and air-to-air roles, reaching speeds of up to Mach 2. However, its range has seen little improvement over the decades, constrained by its traditional solid-fuel rocket motor limits. In response to evolving battlefield threats, particularly the increase of unmanned aerial systems (UAS) with advanced intelligence, surveillance, and reconnaissance (ISR) capabilities, the US Army has launched an ambitious next-generation missile propulsion initiative, codenamed Red Wasp. – https://interestingengineering.com/military/us-army-upgrades-cold-war-missiles
Porton Man: Humanoid robot to test chemical gear for US Army in real-world conditions
(Interesting Engineering – 22 May 2025) The US Army has awarded a $1.7 million contract to develop an advanced robotic mannequin system for testing protective gear against chemical warfare agents. The Porton Man Robotic Test System, modeled after a UK version, is designed to replicate the size and shape of real soldiers using data from the Anthropometric Survey of Military Personnel. The system aims to enhance gear testing accuracy and is expected to be delivered within 18 to 24 months from the contract date. During National Robotics Week, held from April 5 to 13, US Army researchers showcased cutting-edge innovations in robotics, AI, and autonomy to boost human-machine teaming between robots and soldiers. – https://interestingengineering.com/military/humanoid-robot-to-test-chemical-gear
Allies review progress with NATO cyber defence pledge, identify next steps to increase cyber resilience
(NATO – 21 May 2025) On 20-21 May 2025, NATO Allies and several Partner nations met in Poland for NATO’s annual Cyber Defence Pledge Conference. Held at the Polish Cyber Command in Legionowo, the Conference brought together representatives from NATO member states as well as from Azerbaijan, Georgia, Iraq, Ireland, Japan, the Republic of Korea, Switzerland, Ukraine and the European Commission and the European External Action Service. Commander of the Polish Cyber Command Major General Karol Molenda and NATO’s Assistant Secretary General for Innovation, Hybrid and Cyber, Ambassador Jean-Charles Ellermann-Kingombe co-chaired the event. – https://www.nato.int/cps/en/natohq/news_235531.htm
Golden Dome creates a new missile defense bargain with US partners
(Léonie Allard and Jean-Loup Samaan – Atlantic Council – 21 May 2025) New details are emerging about the White House’s “Golden Dome” initiative, a missile-defense system intended to intercept long-range and hypersonic missiles fired at the United States. Even in its early stages, the initiative looks set to become the focal point of US missile defense policy. On Tuesday, US President Donald Trump announced an initial $25 billion investment out of a total $175 billion to start building Golden Dome with the aim of completing it within his term. US Secretary of Defense Pete Hegseth further laid out that Golden Dome will be a layered system, will draw from existing and future technologies as it is phased in over time, and will have an open architecture, allowing multiple companies to contribute to it. – https://www.atlanticcouncil.org/blogs/new-atlanticist/golden-dome-creates-a-new-missile-defense-bargain-with-us-partners/
US’ tech could track submarines in hostile, GPS-denied environments, boost war dominance
(Interesting Engineering – 21 May 2025) The United States could soon get advanced technology in unmanned airborne detection and tracking of enemy submarines as two American firms have joined hands. General Atomics Aeronautical Systems, and Ultra Maritime are combining the world’s most advanced and prolific unmanned aerial system GA-ASI’s MQ-9B SeaGuardian with the world’s only miniaturized, low power sonobuoys and receivers. The companies revealed that the strategic partnership will make it possible for the U.S. and its allies to use sophisticated techniques to track submarines in hostile, GPS-denied environments. – https://interestingengineering.com/military/us-tech-could-detect-submarines
Trump: Golden Dome to cost $175B, be ready in three years
(Patrick Tucker – Defense One – 20 May 2025) An architecture, a price tag, and a development leader have been set for the ambitious Golden Dome anti-missile defense system, President Trump said Tuesday. “We have officially selected an architecture for this state-of-the-art system that will deploy next-generation technologies across the land, sea, and space, including space-based sensors and interceptors,” Trump said from the White House. “This design for the Golden Dome will integrate with our existing defense capabilities and should be fully operational before the end of my term. So we’ll have it done in about three years.” – https://www.defenseone.com/technology/2025/05/trump-golden-dome-cost-175b-be-ready-three-years/405474/?oref=d1-homepage-top-story
Ukraine’s tiny thermobaric robot tank can burn enemies with 4,532°F flames
(Interesting Engineering – 20 May 2025) The Ministry of Defense of Ukraine (MOD) has officially authorized the KRAMPUS, a domestically developed thermobaric-armed unmanned ground vehicle (UGV). Intended for use in offensive and defensive operations, the UGV can also be armed with reactive thermobaric launchers for use by the Ukrainian Armed Forces. According to the Ukrainian MOD, the KRAMPUS is compact in size and weight and can be carried on the bed of a pickup truck, minibus, or trailer. The UGV is also very quiet thanks to its silent electric motors and a tracked chassis. – https://interestingengineering.com/military/ukraine-thermobaric-touting-ground-robot
Missile Defense Agency takes delivery of first THAAD radar to track hypersonics
(Theresa Hitchens – Breaking Defense – 19 May 2025) Raytheon has delivered the first upgraded radar for the Terminal High Altitude Area Defense (THAAD) capable of tracking hypersonic missiles to the Missile Defense Agency, the company announced today. The new version of AN/TPY-2 has a longer range and can provide targeting coordinates to other missile defense interceptors beyond just the Army’s THAAD batteries, Jon Norman, Raytheon’s vice president for Air and Space Defense Systems Requirements and Capabilities, told Breaking Defense. “What the TPY-2 does now, with the Gallium Nitride front-end in it, is it can see things twice as far, so we can make that command and control decision a lot earlier on which effector to use, whether it’s an SM series or it’s a Patriot, or it’s a THAAD,” he said. – https://breakingdefense.com/2025/05/missile-defense-agency-takes-delivery-of-first-thaad-radar-to-track-hypersonics/
Space Force FORGEing ahead with missile warning ground system
(Breaking Defense – 19 May 2025) The Space Force is moving to finally put in place the long-troubled next-generation ground system for its missile warning constellations — with service officials now predicting the Future Operationally Resilient Ground Evolution (FORGE) command and control (C2) system to be fully functioning in 2028. Space Systems Command (SSC) in recent weeks has issued a flurry of contracts for the key hardware and software “thrusts” that make up the FORGE program, following a restructuring in 2023 that broke the effort into more manageable pieces. “The modular nature of FORGE allows us to constantly improve and adapt to the needs of the operators and bring in added capabilities or sensors,” Lt. Col. Dan Groller, SSC’s FORGE materiel leader, told Breaking Defense on May 16. – https://breakingdefense.com/2025/05/space-force-forgeing-ahead-with-missile-warning-ground-system/
China claims new anti-stealth radars can detect US F-35s, nuclear subs, drones
(Interesting Engineering – 19 May 2025) China has taken a major step in countering the technological edge of US stealth platforms by unveiling next-generation anti-stealth radar systems. These systems were showcased at the 11th World Radio Detection and Ranging Expo (World Radar Expo), which opened on Saturday in Hefei, Anhui Province. The expo highlighted radar technologies that pose significant threats to US stealth fighter jets, unmanned aerial vehicles (UAVs), and even nuclear-powered submarines, platforms long considered the cornerstone of American strategic dominance. – https://interestingengineering.com/military/china-debuts-anti-stealth-radars
China to deploy world’s largest drone mothership with 4,500-mile range, 100-UAV payload
(Interesting Engineering – 19 May 2025) In a significant development that underlines China’s continued investment in unmanned warfare capabilities, the People’s Liberation Army (PLA) is preparing to deploy the Jiu Tian (“High Sky”) drone carrier for its inaugural mission by the end of June. State broadcaster CCTV confirmed the upcoming test flight on Monday, following preliminary reports from Chinese media over the weekend. The Jiu Tian is a high-altitude, long-range unmanned aerial vehicle (UAV) designed as a force multiplier for the PLA’s drone warfare and swarming capabilities. – https://interestingengineering.com/military/china-deploy-world-largest-drone-carrier