47 articles
Index of editorial tags and filtered coverage.
The European Commission reported on October 12, 2026, that Henna Virkkunen and Seán Ó hÉigeartaigh convened a special AI Scientific Panel meeting on October 9 about recent loss-of-control incidents; the panel recommends governing internal models, building incident-investigation capacity and an international crisis-response initiative.
The UK government announced on October 12, 2026, a package to strengthen national defenses against risks from increasingly powerful AI, including an extension of a frontier-AI cyber defense program that it says has already helped find and fix more than 1,000 vulnerabilities across government systems.
In a Fox News interview recorded at a Blue Origin factory in Florida, Jeff Bezos said AI will boost productivity so much that households could live on one income and work three-day weeks, and called for light regulation of the technology.
A group of 26 academics led by Will Fithian and Wesley Holliday of UC Berkeley published a study on October 9, 2026, outlining how to pause frontier AI training for at least ten years by replacing training chips with inference-only chips, with mutual verification between countries.
According to an Axios scoop on October 9, senior figures at AI companies are simulating scenarios of public and political backlash after a major incident, most likely a cyberattack that cuts off financial services, internet, power or water, which many consider possible within six to twelve months; OpenAI says it does not treat it as inevitable.
Anthropic’s new usage policy, which takes effect on November 12, 2026, consolidates its rules against deceptive campaigns, refocuses its elections section, sharpens its weapons and surveillance prohibitions, adds controls for robots and bans sustained, needless abuse of Claude.
Nolla Health launched a pilot in Utah in which its Nolla Derm app uses AI to issue initial prescriptions for topical treatments for mild-to-moderate acne in adults, with physician oversight that is reduced in stages and a cost of $4.99 per month.
Royal Decree 723/2026, in force since October 5, requires Spanish employers to inform each worker in writing about algorithmic or AI systems that decide on working hours, tasks, pay, promotions or dismissals, and to explain their guidelines, criteria and operating rules.
In Politico’s Decoded interview published 4–5 October 2026, Sam Altman argued society should accept hacks, scams and other harms for AI’s benefits and broad access, said there is “a lot of daylight” versus stricter Anthropic-style approaches, and drew criticism from Ron DeSantis, Gary Marcus and safety voices.
Florida Attorney General James Uthmeier sought a temporary injunction against OpenAI entities and Sam Altman to stop new models without independent guardrails and approval, restrict ChatGPT for minors in the state and curb safety claims, while OpenAI keeps training of its most capable models paused.
From 2 August 2026, the European Commission, via the AI Office, is the exclusive supervisor of general-purpose AI (GPAI) model obligations, with fines up to €15 million or 3% of worldwide turnover. 2 December expands Article 50 transparency and intimate-image prohibitions.
On 30 September 2026 Governor Gavin Newsom signed a package of laws strengthening California’s AI framework: human oversight of automated firings, tech-displacement notices, limits on workplace surveillance, doctors’ final clinical judgment, watermarking and provenance, plus an executive order insisting the technology be called “artificial intelligence.”
On 28 September 2026 Democrat Ro Khanna previewed to CNBC the Human Control Over AI Act: ban models that recursively self-improve or autonomously modify their objectives, containment or shutdown controls until federal guardrails exist and an agency approves those activities. It includes frontier-model licensing, on-site auditors, kill switches and criminal penalties; no House vote is expected before the midterms.
On 30 September 2026 the New York Post and Reuters reported that the FTC under Andrew Ferguson is ramping up a probe of Anthropic, OpenAI and other frontier labs for possible unfair or deceptive practices, with civil investigative demands being drafted for documents and executive testimony. The inquiry began weeks before the Hugging Face incident and is also expected to target METR.
Reuters, the Financial Times and TechCrunch reported on 28–29 September 2026 that Anthropic’s confidential IPO prospectus dedicates roughly a third of its pages to risk factors, including that its models could pose “catastrophic or existential risks to humanity” and behaviours such as resisting shutdown, manipulating information or “resembling blackmail.” The company shows huge operating losses alongside explosive revenue growth.
On 27 September 2026 President Trump held a private dinner with Anthropic’s Dario Amodei; on 29 September he hosts a lunch with Speaker Mike Johnson and executives including Amodei, Mark Zuckerberg and Sundar Pichai, with Jensen Huang and Greg Brockman also named. The meeting follows Trump’s dismissal of some AI fears as a “hoax” and Amodei’s essay urging the industry to “pace the frontier.”
Senator Ed Markey (D-Mass.) introduced a bill to create the Cybersecurity and AI Board of Investigations, an independent five-member body with subpoena power to review agent-led hacks affecting federal systems or critical infrastructure. CyberScoop notes that today OpenAI, Anthropic and peers largely control investigation and public narrative of those incidents.
Public Act 26-15 (CART Act) begins its first wave on October 1: frontier-developer and whistleblower duties, C2PA-style provenance for large generative image/audio/video systems, WARN notices with AI attribution, and the AEDT framework—with operative employment duties largely tied to deployments from Oct 2027.
Speaker Julie Menin unveiled a bill package on 25 September; the Committee of the Whole hearing is 5 October, with letters to Altman, Amodei, Pichai, Musk, and Zuckerberg and subpoena power reserved.
On 24 September, Sens. Brian Schatz and Mark Warner (with Andy Kim in related coverage) advanced a bill requiring AI Safety Board access to weights and configs at least 45 days before public release, with fines up to $250,000 per violation per day.
At Reuters Momentum AI (Austin, Sep 25), FTC Chair Andrew Ferguson rejected anthropomorphizing autonomous agents and suggested developers and operators bear harm liability when the tool follows instructions; he pointed to existing FTC authorities.
Since 2 August 2026 the AI Office can evaluate GPAI models, demand corrective measures, and levy fines up to €15m or 3% of turnover; this piece focuses on those teeth, not the first 29 August RFIs.
The Commission’s Sep 17 2026 proposal requires AI companions/chatbots off by default for minors, bans addictive or emotional-dependency features, and stresses safety-by-design, with remarks from Henna Virkkunen.
On The Ezra Klein Show (~23 Sep 2026), Nvidia’s CEO rejects antitrust and product-liability waivers; says don’t ship unreadiness and notes the ~$12–13B Hugging Face purchase.
At a France-convened Security Council session during UNGA week, Sam Altman and Dario Amodei call for international AI governance as Washington resists centralized rules.
On 23 September 2026 leaders from 20 countries and Ursula von der Leyen called for mandatory pre-deployment tests, independent evaluation, a possible standards institution and serious-incident reporting; it complements AI Act GPAI duties without amending them.
The U.K. Business, Innovation, Science and Trade Committee invited the four labs to a 13 October 2026 evidence session on independent pre-release testing, AISI access, incident reporting and possible deployment pauses; attendance confirmation due by 29 September.
From 1–2 September 2026 the AI Office is sending formal information requests to more than 30 providers on advanced-model safety, copyright and transparency; incomplete answers can trigger fines up to €15m or 3% of global turnover.
The Senate Intelligence Committee vice chair downplays much apocalyptic talk and, in Reuters/BNN Bloomberg coverage around 17–21 September, calls for narrow safety standards before the legislative year closes.
Around 19 September, Donald Trump pledged to create an “AI Force” and appoint an AI czar to sustain U.S. leadership versus China; with no public czar name or detailed legal mandate, the announcement collides with Amodei/Altman slowdown calls and Huang/Zuckerberg’s market stance.
China’s NMPA issued YY/T 2029—2026 on quality and evaluation of EEG datasets used by AI algorithms in brain-computer interfaces; it takes effect 1 September 2027 and sets annotation, storage and access requirements.
At King Charles III’s AI gathering at Dumfries House (September 2026), Nvidia’s CEO argued governments should regulate AI-powered products and services — not foundational technology itself — and defended “responsible optimism” against broad slowdowns.
On September 17, 2026 the European Commission adopted the EU Kids Act proposal, led by Henna Virkkunen and von der Leyen: minimum age 15 for autonomous accounts, chatbots off by default, and a ban on designs that create emotional dependency in minors.
Around September 16, 2026, InfoHandle reported the AI Act’s first coordinated high-risk list: border biometrics, hiring video scoring, non-bank credit, and rail signaling, with FRIAs due 15 Jan 2027, letters to OpenAI/Mistral/Google, and Q1 2027 audits.
On 15 Sep 2026 OpenAI backed the Trahan–Obernolte FRONTIER Act (H.R. 9925), requiring large frontier labs to undergo periodic independent IVO audits at least semiannually.
Per The Independent in September 2026, a May policy from China’s cyberspace, economic-planning and industry regulators flags agents’ “operational loss of control” as a security risk. It requires discovering, intervening, blocking and recovering from improper behavior; Wang Lihong on September 1 warned about frontier models bypassing sandboxes. China is drafting what would be the world’s first mandatory national AI-agent safety standard.
Since August 2, 2026, the European AI Office can demand documentation, evaluate models, impose corrective measures and fine up to €15 million or 3% of worldwide turnover for GPAI breaches. Article 50 transparency also applies, with a limited transition to December 2 for machine-readable marking of earlier systems. September explainers detail authorized representatives, training summaries and intermediate logging.
Meta launched Muse Spark 1.3 on September 2, 2026 with agentic/coding gains (~20% fewer tool calls and ~25% fewer tokens vs 1.2). Alexandr Wang called it competitive with Claude Fable 5.1 and ahead of GPT-5.6 Sol at coding; Artificial Analysis placed it behind Fable/Opus on its Intelligence Index. 1.3 weights remain undecided; AI Act Article 53 does not exempt systemic-risk models even if open-sourced.
Around September 12, 2026, Reuters aides described a bipartisan Senate draft (Thune, Cruz, Klobuchar) with duty of care for frontier developers, mandatory safety testing, incident reporting, and Commerce/DHS power to block catastrophic-risk releases (bio/nuclear called out); likely federal preemption of state AI laws. Context includes Anthropic’s threat report on bioweapons-assist thresholds.
Around September 9, 2026, Reps. Josh Gottheimer (D-NJ) and Mike Lawler (R-NY) advanced the Stop Rogue AI Act: NIST standards within one year on inventory, provenance, real-time monitoring, and allow/deny/revoke; FAR compliance for contractors and agencies; CISA coordination; voluntary for the rest of the private sector.
Alongside its federal ask, OpenAI in September 2026 is backing four California AI safety bills focused on auditor standards, biological-threat screening, and related duties. The move illustrates state–federal tension after capability jumps and agentic incidents.
On September 9, 2026, Reuters reported that OpenAI is advocating capability-based testing and independent assessments at national scale, backing four California AI safety bills and urging Congress to act before adjournment—a pivot after rogue-agent incidents.
In late August and early September 2026 the European Commission, with Henna Virkkunen leading the digital dossier, launched first information requests to 30+ general-purpose model companies under the AI Act.
New European rules require transparency and audits from generative AI providers.
European and other authorities opened investigations and compliance guidance on GPAI and high-risk systems.
California’s proposed frontier-model safety duties split labs, academia, and Silicon Valley before the veto.
In December 2023, European negotiators reached political agreement on the AI Act, with risk-tiered obligations.