Governance
New AI export marketplace looks to connect domestic AI firms with foreign buyers
(Alexandra Kelley – NextGov – 22 September 2025) The Commerce Department launched its latest initiative to lay the groundwork for a strong artificial intelligence export environment by unveiling the American AI Exports Program on Tuesday. The program was designed to further expand the U.S. technology industry’s access to global customers. Commerce has christened the effort with a new website, which was created based on feedback from an earlier request for information. – https://www.nextgov.com/artificial-intelligence/2025/10/new-ai-export-marketplace-looks-connect-domestic-ai-firms-foreign-buyers/408995/
Shared Residual Liability for Frontier AI Firms
(Ben Gil Friedman – Lawfare – 22 October 2025) As artificial intelligence (AI) systems become more capable, they stand to dramatically improve our lives—facilitating scientific discoveries, medical breakthroughs, and economic productivity. But capability is a double-edged sword. Despite their promise, advanced AI systems also threaten to do great harm, whether by accident or because of malicious human use. Many of those closest to the technology warn that the risk of an AI-caused catastrophe is nontrivial. In a 2023 survey of over 2,500 AI experts, the median respondent placed the probability that AI causes an extinction-level event at 5 percent, with 10 percent of respondents placing the risk at 25 percent or higher. Dario Amodei, co-founder and CEO of Anthropic—one of the world’s foremost AI companies—believes the risk to be somewhere between 10 percent and 25 percent. Nobel laureate and Turing Award winner Geoffrey Hinton, the “Godfather of AI,” after once venturing a similar estimate, now places the probability at more than 50 percent. Amodei and Hinton are among the many leading scientists and industry players who have publicly urged that “mitigating the risk of extinction from AI should be a global priority” on par with “pandemics and nuclear war” prevention. – https://www.lawfaremedia.org/article/shared-residual-liability-for-frontier-ai-firms
Kenya leads the way in AI skilling across Africa
(DigWatch – 22 October 2025) Kenya’s AI Skilling Initiative (AINSI) is offering valuable insights for African countries aiming to build digital capabilities. With AI projected to create 230 million digital jobs across Africa by 2030, coordinated investment in skills development is vital to unlock this potential. Despite growing ambition, fragmented efforts and uneven progress continue to limit impact. – https://dig.watch/updates/kenya-leads-the-way-in-ai-skilling-across-africa – https://news.microsoft.com/source/emea/features/tapping-into-africas-230-million-ai-powered-jobs-opportunity/
Most EU workers now rely on digital tools and AI
(DigWatch – 22 October 2025) A new EU study finds that 90% of workers rely on digital tools, while nearly a third use AI-powered chatbots in their daily work. The JRC and European Commission surveyed over 70,000 workers across all EU Member States between 2024 and 2025. The findings show that AI is most commonly used for writing and translation tasks, followed by data processing and image generation. Adoption rates are particularly high in Northern and Central Europe, especially in office-based sectors. – https://dig.watch/updates/most-eu-workers-now-rely-on-digital-tools-and-ai – https://joint-research-centre.ec.europa.eu/jrc-news-and-updates/impact-digitalisation-30-eu-workers-use-ai-2025-10-21_en
Dutch watchdog warns AI chatbots threaten election integrity
(DigWatch – 22 October 2025) The Dutch data authority warns AI chatbots are biased and unreliable for voting advice ahead of national elections. An AP investigation found chatbots often steered users to the same two parties, ignoring their actual preferences. In over half of the tests, the bots suggested either Geert Wilders’ far-right Freedom Party (PVV) or the leftwing GroenLinks-PvdA led by Frans Timmermans. Other parties, such as the centre-right CDA, were rarely mentioned even when users’ answers closely matched their platforms. – https://dig.watch/updates/dutch-watchdog-warns-ai-chatbots-threaten-election-integrity
What federal buyers need to succeed with AI-enabled procurement
(Tim Cooke – NextGov – 22 October 2025) For five years, the federal government has been under a legal mandate to train acquisition professionals in using and buying artificial intelligence (AI). During that time, federal spending on AI has risen precipitously. As AI technology accelerates into the future, buying relevant, rapidly evolving capabilities requires foresight and understanding. Eighty percent of chief procurement officers (CPOs) surveyed plan to deploy AI tools for spend analytics, contract management and supplier selection over the next three years. Leaders are expecting that procurement operations will be radically better and faster than they are today. Anticipating this, the 2022 Artificial Intelligence Training for the Acquisition Workforce Act requires the Office of Management and Budget (OMB) to provide AI training for procurement professionals. The legislation mandates regular updates addressing capabilities, risks and ethical implications. The proposed 2025 AI and Critical Technology Workforce Framework Act would expand on this. – https://www.nextgov.com/ideas/2025/10/what-federal-buyers-need-succeed-ai-enabled-procurement/408978/?oref=ng-homepage-river
Meta and Google’s Ad Ban Upends Political Campaigning in Europe
(Sam Jeffers – Tech Policy Press – 22 September 2025) The collective decision by Meta and Google to ban political, social and issue-based ads in the European Union, citing the European Union’s new Transparency and Targeting of Political Advertising (TTPA) regulation, is a seismic shift that will fundamentally reshape political campaigning across the continent. Some will rejoice. Political ads can be annoying, particularly if a politician or argument you don’t like is repeatedly thrust into your social media feeds. After the 2016 US presidential election and Brexit referendum, people, mostly incorrectly, ascribed Facebook ads with magical powers of persuasion. Surely stopping them will also stop bad things like these happening again? Unfortunately, this view is wrong. There are no winners here — not the tech companies, not the EU, not parties, campaigns or charities, not voters, not advocacy and research organisations like ourselves, nor democracy. The primary result of collective ban on political ads is that online political speech in the EU will become even more algorithmically mediated. Political actors will have significantly less control over whether anyone sees their material than they do now. – https://www.techpolicy.press/meta-and-googles-ad-ban-upends-political-campaigning-in-europe/
The Tech Arms Race is Reshaping Our Lives — and Threatening Democracy
(Ilia Siatitsa – Tech Policy Press – 22 October 2025) Earlier this year, Europe saw a milestone: three defense start-ups reached unicorn status, each valued at over €1 billion. These companies, building a range of ‘products’ from autonomous drones to robotic systems — even equipping real cockroaches with micro-backpacks — are part of a growing wave of defense-tech innovation. Europe is now racing to catch up with the United States and China, where defense-tech start-ups have already been attracting significant venture capital and government backing. But this isn’t just about military competition. It’s about a deeper transformation that’s already reshaping our societies. A new doctrine is emerging: technological dominance at any cost. What we’re witnessing is not just an arms race — it’s a race of technological dependency where civil and military infrastructures are fusing, and the battlefield is no longer “over there.” It’s here, embedded in our cities, our homes, and our daily lives. Personal data is the critical resource underlying much of this innovation. In turn, the paradigm of rights-based data protection laws is now competing with one driven by national security, to protect ‘our citizens’ personal data’ from hostile exploitation. This distinct shift from protecting “everyone’s personal data” is now “our people’s data,” which implies something different for “their data.” – https://www.techpolicy.press/the-tech-arms-race-is-reshaping-our-lives-and-threatening-democracy/
From “black box AI” to operational AI transparency: How the HAIP Reporting Framework can play an important role in global AI governance
(Nicolas Miailhe, Amanda Craig, Hector de Rivoire, Robert Trager – OECD.AI – 21 October 2025) As the “black box challenge” remains central to the AI revolution, transparency has shifted from being a goal to a fundamental governance objective, shaping into an operational discipline for responsible and trustworthy innovation. Expectations are rising rapidly, and so too is the range of policy instruments to address them. As the number of instruments grows, fragmentation and interoperability gaps could emerge across AI value chains, creating complexities for business, developers and other actors. – https://oecd.ai/en/wonk/from-black-box-ai-to-operational-ai-transparency-haip-role-global-ai-governance
Geostrategies
Why Europe’s Resistance to Big Tech Matters for the Future of Democracy
(Courtney C. Radsch – Tech Policy Press – 22 October 2025) Over the past several years, Europe has enacted a set of laws, regulations and antitrust remedies intended to bolster democratic oversight and accountability, compliance and competition, and ultimately the health of its information ecosystems by redressing the outsized power of very large online platforms (VLOPs). Over the past year, the United States government and dominant tech firms launched a campaign to undermine digital legislation in Europe, equating regulation with censorship to protect American dominance while undermining the democratic sovereignty of its allies. This campaign targets the Digital Services Act (DSA) and Digital Markets Act (DMA) as well as other laws that apply to the very largest digital platforms. The administration has inaccurately portrayed these laws as “foreign censorship” and “taxes on American firms.” – https://www.techpolicy.press/why-europes-resistance-to-big-tech-matters-for-the-future-of-democracy/
Security and Surveillance
15th EDEN Event: Data Protection and Cybersecurity in Law Enforcement
(Europol – 22 October 2025) The 15th Europol Data Protection Experts Network (EDEN) conference took place from 21 to 22 October 2025 at the University of Malta’s Valletta Campus. Jointly organised by the Academy of European Law (ERA) and EDEN, and co-hosted by the Maltese Police Force and the University of Malta, the event gathered senior policymakers, data protection officers, law enforcement officials and academics from across Europe. This year’s conference title, ‘Red Light, Green Light! Reduce Complexity, Increase Speed, Defend EU Values’, drew inspiration from a term used in data protection and cybersecurity to signify compliance or non-compliance. This phrase also echoes the children’s game popularised by the Netflix series Squid Game, where the stakes are extremely high – life or death. Similarly, data protection and cybersecurity in law enforcement can have severe, real-world implications for individuals. – https://www.europol.europa.eu/media-press/newsroom/news/15th-eden-event-data-protection-and-cybersecurity-in-law-enforcement
Reconfiguring U.S. Cyber Strategy in the Wake of Salt Typhoon
(Alistair Simmons – Lawfare – 22 October 2025) In a multiyear campaign called Salt Typhoon, threat actors from the People’s Republic of China (PRC) have broken into many major telecom providers, including Verizon, AT&T, and T-Mobile. Collectively, 397.1 million users subscribe to these three telecom providers, indicating that Salt Typhoon could impact hundreds of millions of people. Due to the magnitude of this data breach, Sen. Mark Warner (D-Va.), vice chairman of the Senate Select Committee on Intelligence, claimed Salt Typhoon is the “worst telecom hack in our nation’s history.” As former FBI director, Christopher Wray claimed Salt Typhoon is the “most significant cyber espionage campaign in history.” – https://www.lawfaremedia.org/article/reconfiguring-u.s.-cyber-strategy-in-the-wake-of-salt-typhoon
US cyber policy goals have regressed during Trump 2.0 in ‘unprecedented setback,’ landmark report says
(David DiMolfetta – NextGov – 22 October 2025) A landmark progress report tied to a congressionally-mandated cyber policy body shows that the federal government’s cyber policy posture has regressed by roughly 13% on advances made in prior years, marking the first major reversal since the group’s creation. The analysis — delivered Wednesday by the Cyberspace Solarium Commission 2.0 in the Foundation for Defense of Democracies — illustrates the high-level consequences of hollowing out the federal cyber workforce amid broader efforts to reorganize the government and make it more efficient in the eyes of the White House and top agency leaders. Since February, the Trump administration has moved to shrink the federal workforce and shutter offices it claims undermine American ideals. – https://www.nextgov.com/cybersecurity/2025/10/us-cyber-policy-goals-have-regressed-during-trump-20-unprecedented-setback-landmark-report-says/408990/?oref=ng-home-top-story
Iranian hackers targeted over 100 govt orgs with Phoenix backdoor
(Bleeping Computer – 22 October 2025) State-sponsored Iranian hacker group MuddyWater has targeted more than 100 government entities in attacks that deployed version 4 of the Phoenix backdoor. The threat actor is also known as Static Kitten, Mercury, and Seedworm, and it typically targets government and private organizations in the Middle East region. Starting August 19, the hackers launched a phishing campaign from a compromised account that they accessed through the NordVPN service. – https://www.bleepingcomputer.com/news/security/iranian-hackers-targeted-over-100-govt-orgs-with-phoenix-backdoor/
Meta strengthens protection for older adults against online scams
(DigWatch – 22 October 2025) The US giant, Meta, has intensified its campaign against online scams targeting older adults, marking Cybersecurity Awareness Month with new safety tools and global partnerships. Additionally, Meta said it had detected and disrupted nearly eight million fraudulent accounts on Facebook and Instagram since January, many linked to organised scam centres operating across Asia and the Middle East. – https://dig.watch/updates/meta-strengthens-protection-for-older-adults-against-online-scams – https://about.fb.com/news/2025/10/cybersecurity-awareness-month-helping-older-adults-avoid-online-scams/
Judge bars NSO Group from using spyware to target WhatsApp in landmark ruling
(DigWatch – 22 October 2025) A US federal judge has permanently barred NSO Group, a commercial spyware company, from targeting WhatsApp and, in the same ruling, cut damages owed to Meta from $168 million to $4 million. The decision by Judge Phyllis Hamilton of the Northern District of California stems from NSO’s 2019 hack of WhatsApp, when the company’s Pegasus spyware targeted 1,400 users through a zero-click exploit. The injunction bans NSO from accessing or assisting access to WhatsApp’s systems, a restriction the firm previously warned could threaten its business model. – https://dig.watch/updates/judge-bars-nso-group-from-using-spyware-to-target-whatsapp-in-landmark-ruling
Defence, Military, and Warfare
World’s first nuclear carrier built for 6th-gen fighters could be deployed by Italy
(Interesting Engineering – 22 October 2025) Italy will start studying the possibility of building the country’s first nuclear-powered aircraft carrier in 2026. This project is an important step in modernizing Italy’s Navy and could place the country among a few nations that operate nuclear aircraft carriers. The project, known as Portaerei di Nuova Generazione, or New Generation Aircraft Carrier, will be developed under a long-term plan by the Italian Navy (Marina Militare) that extends to 2040. – https://interestingengineering.com/military/worlds-first-nuclear-jet-carrier
US firm’s Star Wars-style humanoid robot soldier brings sci-fi to battlefield
(Interesting Engineering – 22 October 2025) A San Francisco-based robotics company named Foundation has developed what could be the world’s first humanoid robot explicitly designed for warfare. Called the Phantom MK-1, the machine stands 5 feet 9 inches tall, weighs 175 pounds, and can carry loads up to 44 pounds. The company’s move breaks ranks with most major robotics manufacturers, which have pledged not to weaponize their technology. – https://interestingengineering.com/military/real-life-star-wars-battle-droid
Massive Chinese stealth drone spotted in flight fuels bomber rivalry with US
(Interesting Engineering – 22 October 2025) A mysterious stealth drone, possibly China’s next-generation bomber, has taken to the skies for the first time, marking a significant step in the country’s aerial weapons development. The large flying-wing aircraft, unofficially dubbed GJ-X, was spotted in a video circulating on Chinese social media since October 19, as reported by South China Morning Post (SCMP). Military analysts quickly connected the aircraft to satellite images captured in August over Malan Airbase in China’s Xinjiang Uyghur Autonomous Region. The drone seen in the video appeared nearly identical to the one previously observed there. – https://interestingengineering.com/military/mysterious-chinese-stealth-drone-spotted
Frontiers
Boston Dynamics, Field AI, and Honeywell Vets Launch AI & Robotics Advisory Firm
(AI Insider – 22 October 2025) Nascent Scale, founded by robotics veterans Tim Dykstra and Chris McCasky, launched as an advisory firm to help organizations move robotics and AI initiatives from pilots to enterprise-scale deployment with measurable ROI. The founders bring leadership experience from Boston Dynamics, Field AI, and Honeywell, positioning the firm to bridge the gap between nascent technologies and scaled operational outcomes. Service lines target technology providers, industrial end users, and investors, covering go-to-market acceleration, pilot design and rollout, partner/channel strategy, due diligence, and interim GTM support. – https://theaiinsider.tech/2025/10/22/boston-dynamics-field-ai-and-honeywell-vets-launch-ai-robotics-advisory-firm/
China’s Lianhe Sowell to Invest $105M to Establish AI Robot Manufacturing and R&D Base in UAE
(AI Insider – 22 October 2025) Lianhe Sowell International Group (NASDAQ: LHSW) signed a non-binding $105 million term sheet with Excellent Capital Investments to finance its first AI-powered robot manufacturing and R&D base in the UAE as part of its global expansion. The UAE project totals ~$132.5 million, with Lianhe Sowell contributing 20% equity and ECI providing 80% as structured senior secured credit, pending due diligence and final approvals. Targeting an 18-month build, the base aims to produce several thousand AI robots annually, create 2,000+ skilled local jobs, add a robotics training center, and support “Industry 4.0” adoption across sectors like automotive, construction, aviation, rail, marine, and new energy. – https://theaiinsider.tech/2025/10/22/chinas-lianhe-sowell-international-group-to-invest-105-million-to-establish-ai-robot-manufacturing-and-rd-base-in-uae/
MIT unveils SEAL, a self-improving AI model
(DigWatch – 22 October 2025) Researchers at the Massachusetts Institute of Technology (MIT) have unveiled SEAL, a new AI model capable of improving its own performance without human intervention. The framework allows the model to generate its own training data and fine-tuning instructions, enabling it to learn new tasks autonomously. – https://dig.watch/updates/mit-unveils-seal-a-self-improving-ai-model
BlastPoint Closes $10.6M Growth Funding to Advance Responsible AI for Utilities and Financial Services
(AI Insider – 21 October 2025) BlastPoint raised $10.6 million in growth funding led by MissionOG, with participation from Curql Fund, Michigan Capital Network, and existing investors to expand its AI-driven customer intelligence platform. The company operates in a $13 billion market across 13 sectors and will use the funds to enhance its auditable, equity-focused AI prediction technology built only on permission-granted data. With over 2X annual growth and 100% customer retention, BlastPoint now reaches nearly one-quarter of U.S. households, helping utilities and financial institutions use AI responsibly to improve customer engagement and access to financial assistance. – https://theaiinsider.tech/2025/10/21/blastpoint-closes-10-6m-growth-funding-to-advance-responsible-ai-for-utilities-and-financial-services/
Appy.AI Launches AI Business Creation Platform After Raising $5M Seed Funding
(AI Insider – 21 October 2025) Appy.AIlaunched a public beta after a $5M seed round led by Four Rivers (Dan Scholnick) and Founder Collective, offering a platform to build monetizable AI businesses without coding. Unlike code-gen tools, it delivers production-ready AI agents with full business infrastructure — payments (Stripe), auth, subscriptions, analytics, white-label sites, support, and continuous improvement — via a conversational builder. Early users report going from idea to paying customers in minutes; access is free with no token limits during October 2025, ahead of simple, transparent pricing after beta. – https://theaiinsider.tech/2025/10/21/appy-ai-launches-ai-business-creation-platform-after-raising-5m-seed-funding/
Realm.Security Redefines Security Data Pipelines with AI, Secures $15M to Accelerate Next-Gen SOC Operations
(AI Insider – 21 October 2025) Realm.Security raised $15 million in Series A funding led by Jump Capital, with participation from Glasswing Ventures and Accomplice, to accelerate its AI-native Security Data Pipeline Platform (SDPP). The platform uses AI and machine learning to filter and structure security data in real time, cutting costs, reducing noise, and helping SOC teams focus on critical threats rather than raw data overload. Early enterprise users report rapid deployment and significant savings, with Realm poised to redefine how organizations manage, analyze, and secure their data pipelines. – https://theaiinsider.tech/2025/10/21/realm-security-redefines-security-data-pipelines-with-ai-raises-15m-to-accelerate-next-gen-soc-operations/
Arcjet Raises $8.3M to Tackle Surge in AI-Powered Cyberattacks and Malicious Bots with In-Code Defense
(AI Insider – 21 October 2025) Arcjet raised $8.3 million in Series A funding led by Plural and Ott Kaukver, with participation from Andreessen Horowitz, Seedcamp, and other investors, bringing total funding to $12 million. The company embeds AI-powered threat detection directly into application code, enabling developers to identify and block sophisticated attacks in real time without slowing deployment. Already protecting over 500 production apps across AI and e-commerce, Arcjet is positioning its in-code AI security model as the next-generation defense against adaptive, automated cyber threats. – https://theaiinsider.tech/2025/10/21/arcjet-raises-8-3m-to-tackle-surge-in-ai-powered-cyberattacks-and-malicious-bots-with-in-code-defense/
Versaterm Partners with Vantage Robotics to Expand Public Safety Drone Portfolio
(AI Insider – 21 October 2025) Versaterm partnered with Vantage Robotics to integrate the U.S.-made, NDAA-compliant, Blue UAS–cleared Vesper and Trace drones into its DroneSense by Versaterm public safety platform. The integration delivers secure real-time video/telemetry into OpsHub, unified flight/compliance management, and city-scale coordination across UAVs and ground assets for rapid ISR and DFR operations. Vesper is available now and Trace is coming soon, with Versaterm to be at IACP 2025. – https://theaiinsider.tech/2025/10/21/versaterm-partners-with-vantage-robotics-to-expand-public-safety-drone-portfolio/
Researchers Develop System to Verify Safety and Stability of AI-Driven Control Systems
(AI Insider – 21 October 2025) Researchers at the University of Waterloo have developed a new framework combining mathematics and machine learning to verify the safety and stability of AI systems that control critical infrastructure such as power grids and autonomous vehicles. The study, published in Automatica and supported by Waterloo’s TRuST Scholarly Network, uses neural networks to learn Lyapunov functions—key mathematical tools in control theory—paired with logic-based verification to ensure AI controllers behave safely under dynamic conditions. By automating the generation and validation of mathematical safety proofs, the system could help engineers design more trustworthy AI for real-world applications, with the team planning to release the framework as open-source software for broader adoption. – https://theaiinsider.tech/2025/10/21/researchers-develop-system-to-verify-safety-and-stability-of-ai-driven-control-systems/
Starship Technologies Raises $50M Series C to Scale Autonomous Delivery Across U.S. Cities
(AI Insider – 21 October 2025) Starship Technologies raised $50 million led by Plural, bringing total funding to $280+ million to scale autonomous delivery from European cities and U.S. campuses into American urban markets. The company reports 9 million+ deliveries across seven countries, operates at Level 4 autonomy with positive gross margins, and claims a city-scale logistics network backed by a continuously growing dataset. Partnerships with Grubhub (U.S.) and European platforms including Bolt, Wolt, and Foodora position Starship as an autonomous last-mile backbone as it expands its robot fleet and U.S. footprint. – https://theaiinsider.tech/2025/10/21/starship-technologies-raises-50m-series-c-to-scale-autonomous-delivery-across-u-s-cities/
Romanian Startup .lumen Gets €11M Grant to Develop Autonomous Delivery Robots
(AI Insider -21 October 2025) Romanian startup .lumen secured a €11 million grant for “PABLO – Autonomous Robots for Urban Delivery,” a 2025–2028 program to build humanoid and quadruped sidewalk delivery robots with a national consortium. The project totals 56,856,649.01 RON with 34,987,312.27 RON in non-reimbursable funding (21,309,568.52 RON EU + 13,677,743.75 RON national), running Sept. 16, 2025–Sept. 15, 2028. Led by .lumen with IMSAR, BreadCrumbs Interactive, and Linnify, the effort extends PAD AI navigation to last-mile logistics to cut delivery costs and times, reduce environmental impact, and produce a full software-hardware stack plus new R&D roles and IP. – https://theaiinsider.tech/2025/10/21/romanian-startup-lumen-gets-e11m-grant-to-develop-autonomous-delivery-robots/
Chipmind Launches From Stealth With $2.4 Million For Chip-Making AI Agents
(AI Insider – 21 October 2025) Chipmind, a European startup specializing in AI agents for chip design, has launched Chipmind Agents to help semiconductor companies accelerate the path from specification to manufacturing. The agents integrate directly into existing electronic design automation (EDA) workflows, using proprietary customer data to autonomously execute complex design and verification tasks while keeping engineers in control. Backed by a $2.4 million pre-seed round led by Founderful, Chipmind aims to reduce development cycles and address rising chip complexity through human–AI collaboration. – https://theaiinsider.tech/2025/10/21/chipmind-launches-from-stealth-with-2-4-million-for-chip-making-ai-agents/