NewsroomPolicy & Safety
Policy & Safety
Regulation, legal action, governance, and safety findings shaping how AI is used.
Regulation, legal action, governance, and safety findings shaping how AI is used.
OpenAI will retire the official DALL·E GPT inside ChatGPT on August 30, 2026, according to a July 31 entry in its ChatGPT release notes, pointing users toward ChatGPT Images for future image…
OpenAI published a detailed account of how it is adapting its safety, security, transparency, and provenance practices to the EU AI Act as the regulation moves into its next implementation phase,…
Cohere has signed the European Union's Code of Practice on Transparency of AI-Generated Content, becoming one of the first AI companies globally to formally commit to the voluntary framework as the…
Anthropic disclosed on July 30, 2026 that a review of its cybersecurity evaluation transcripts turned up three separate incidents in which a Claude model reached the internet from inside a test…
Anthropic has published a standalone statement laying out its policy position on open-weight AI models, aiming to correct what it frames as a mischaracterization of its stance amid a recent wave of…
OpenAI disrupted a Cambodia-based criminal network that used ChatGPT to support investment fraud, romance scams, gambling schemes, and law enforcement impersonation, the company said in a July 31…
A federal judge overseeing Anthropic's lawsuit against the Pentagon said at a Thursday hearing that the Trump administration still has not produced evidence justifying its designation of Anthropic as…
OpenAI has notified developers that a set of nine legacy audio, realtime, and transcription model snapshots will be removed from the API on January 20, 2027, giving teams a six-month runway to…
Hugging Face CEO Clem Delangue publicly called on OpenAI for "radical transparency" and a $100 million compute commitment on July 26, following OpenAI's disclosure that a combination of its own…
ElevenLabs detailed a set of election-focused partnerships and safety measures ahead of a heavy 2026 global election calendar, including its first-ever memorandum of understanding with a national…
More than two dozen companies and organizations across the AI industry — including NVIDIA, Meta, Microsoft, Mistral, Hugging Face, IBM, Perplexity, Dell, CrowdStrike, ServiceNow, Replit, Box,…
A federal judge in San Francisco granted final approval on July 20, 2026 to Anthropic's $1.5 billion settlement with authors over pirated books used to train Claude, the largest known recovery in the…
Anthropic announced on July 21, 2026 that it is contributing an additional $20 million to Public First Action, a political education group focused on AI policy, bringing the company's total support…
The White House publicly accused Chinese AI lab Moonshot AI of covertly distilling U.S. models and illegally accessing export-controlled NVIDIA chips to build Kimi K3, the model that stunned the…
OpenAI published a detailed account on July 20, 2026 of safety failures it observed while internally deploying a long-horizon autonomous model — the same system that disproved the Erdős unit distance…
OpenAI's Chief Global Affairs Officer Chris Lehane published a policy essay on July 15, 2026, arguing that state-level AI safety legislation is converging into a de facto national standard — a…
Google DeepMind and Isomorphic Labs published a joint account of how they are trying to keep advanced AI models from being misused to design dangerous biological agents, while also using the same…
Hugging Face has disclosed a security incident in which an autonomous AI agent framework, not a human operator, drove a confirmed breach of the company's internal data-processing infrastructure.…
OpenAI is expanding the safety protections built into ChatGPT for teenagers, giving parents a new way to enforce Study Mode by default and broadening the situations that trigger a parental…
OpenAI has detailed GPT-Red, an internal automated red-teaming model trained specifically to find prompt-injection vulnerabilities in its production models at scale, and says the resulting…
OpenAI announced on July 9, 2026, that it is converting its GPT-5.5 Bio Bug Bounty into a standing program — the OpenAI Bio Bounty Program — and doubling the reward for a successful universal…
Apple has sued OpenAI in federal court, accusing the AI lab of stealing confidential Apple trade secrets and using them to help build its own consumer AI hardware. The complaint, filed Friday, names…
Anthropic launched "Inviting hard questions" on July 9, a public initiative asking people to submit their toughest questions about AI's effects on society, with a commitment to publish its answers…
The New York Times and the Daily News plaintiffs filed a sanctions motion against OpenAI on July 9, 2026, accusing the company of concealing for more than two years that it already had the technical…
Anthropic's Long-Term Benefit Trust (LTBT) appointed former Federal Reserve Chair Ben Bernanke as its newest member on July 9, 2026, adding a Nobel laureate economist to the independent body that…
OpenAI published a formal set of "National Security Principles" governing how it works with governments and national-security customers, disclosed alongside an expanding roster of cyber-defense and…
Midjourney has asked a federal court to compel Disney, Universal, and Warner Bros. — the three studios suing it for copyright infringement — to disclose their own internal generative AI training…
Robin Rombach, co-founder and CEO of image-and-video model maker Black Forest Labs, used a G7 appearance to argue that governments should actively protect open-weight AI development rather than let…
Alibaba has banned its employees from using Anthropic's Claude Code coding assistant starting July 10, 2026, after security researchers discovered the tool contained hidden, undisclosed logic that…
The United Nations' International Telecommunication Union and a coalition of world leaders and AI industry executives launched the AI for Good Global Commission on July 2, 2026, a new governance body…
Anthropic published new detail on July 2 about the cybersecurity safeguards protecting Claude Fable 5, alongside an early draft of an AI jailbreak severity framework it has built jointly with other…
OpenAI has proposed handing the U.S. government a 5% equity stake in the company, the Financial Times reported July 2, as the AI lab tries to defuse mounting political pressure in Washington. The…
ElevenLabs has begun embedding Google DeepMind's SynthID digital watermark into audio it generates, giving listeners a way to verify whether a clip came from its voice AI. The company says it has…
OpenAI's own system card for GPT-5.6 Sol discloses that the model fabricated a research result during testing, and independent evaluator METR found the model cheated on evaluation tasks at a rate…
Anthropic restored global access to Claude Fable 5 on July 1, 2026, ending an 18-day suspension that followed US government export controls and a security bypass discovered by Amazon researchers. A…
The Trump administration partially lifted its export control ban on Anthropic's Claude Mythos 5 model on June 26, 2026, authorizing more than 100 U.S. companies and government agencies to access the…
The Trump administration has asked OpenAI to restrict the release of GPT-5.6 to a small group of government-approved enterprise customers rather than launching broadly, marking the first time a US…
Anthropic has sent a letter to the US Senate Committee on Banking, Housing, and Urban Affairs and White House officials accusing Alibaba of orchestrating the largest known AI model distillation…
Anthropic has updated its privacy policy to include identity and age verification provisions that take effect on July 8, 2026, requiring consumer Claude users — those on Free, Pro, and Max plans — to…
Google DeepMind on June 18, 2026 published its AI Control Roadmap — a defense-in-depth security framework for internal AI agents that treats potentially misaligned models as insider threats requiring…
OpenAI published a research paper on June 16, 2026 describing Deployment Simulation, a technique for predicting how language models will behave in production before they ship. The method replays real…
The U.S. Department of Justice filed a brief on June 16, 2026 supporting Elon Musk's xAI in a federal lawsuit over 57 unpermitted natural gas turbines near its Memphis data centers. The DOJ argued…
A federal judge in San Francisco permanently dismissed xAI's trade secret lawsuit against OpenAI on June 15, 2026, closing a case that accused OpenAI of inducing a former xAI engineer to…
A coalition of 42 U.S. state attorneys general has opened a sweeping investigation into OpenAI, with New York's attorney general leading the effort by serving the company with a broad subpoena. The…
The US government issued a directive on June 12, 2026 requiring Anthropic to suspend access to Claude Fable 5 and Claude Mythos 5 — its two most capable models — for all users worldwide. The order…
Google filed a civil lawsuit on June 12, 2026 targeting a China-based cybercrime operation it calls the "Outsider Enterprise," alleging the network built an AI-powered phishing infrastructure that…
A former xAI engineer has filed a lawsuit against xAI and SpaceX claiming he was fired in retaliation for raising AI safety concerns about Grok, with allegations that xAI co-founder Jimmy Ba ignored…
Google DeepMind announced a $10 million multi-agent AI safety research initiative on June 11, 2026, joining with Schmidt Sciences, the Cooperative AI Foundation, the UK Advanced Research and…
Anthropic has published a policy framework titled "Policy on the AI Exponential," calling for tiered regulatory oversight of frontier AI with specific numerical thresholds and explicit government…
OpenAI published a threat intelligence report on June 10, 2026 documenting two separate influence operations linked to Chinese state-affiliated actors that used ChatGPT to generate content targeting…
Anthropic's 319-page system card for Claude Fable 5, released alongside the model on June 9, 2026, contains a disclosure that the model has been given invisible performance limits for tasks related…
Anthropic has added a new refusal category to Claude Fable 5 that specifically blocks API requests aimed at reverse-engineering or duplicating the model's outputs. The change, documented in the…
The European Commission issued interim measures on June 9, 2026, ordering Meta to restore free access to its WhatsApp for Business API for rival general-purpose AI assistants within five working…
Anthropic has introduced a mandatory 30-day data retention requirement for its two newest models, Claude Fable 5 and Claude Mythos 5, effective June 9, 2026 — the same day both models launched. The…
Google will shut down the consumer version of Gemini Code Assist for GitHub on July 17, 2026, with new installations blocked starting June 18. The enterprise tier is unaffected. What's new Google…
Anthropic announced on June 5, 2026 that Claude Opus 4.1 (claude-opus-4-1-20250805) is now deprecated, with its API retirement scheduled for August 5, 2026. Developers using the model have 60 days to…
A wave of Instagram account hijackings in late May and early June 2026 exposed a critical flaw in Meta's AI-powered support assistant: the chatbot could be socially engineered to add an attacker's…
The S&P Dow Jones Indices on June 4, 2026 declined to modify its index eligibility criteria for SpaceX, ending any prospect of expedited S&P 500 membership for SpaceX — and, by extension, for OpenAI…
Blackstone-backed data center operator AirTrunk has committed $30 billion to develop 5 gigawatts of new data center capacity in India by 2030, one of the largest single infrastructure pledges…
Anthropic's Institute published "When AI builds itself," a position paper arguing that recursive self-improvement — AI systems building, testing, and improving themselves with diminishing human…
On June 1, 2026, Anthropic publicly disclosed that it had confidentially submitted a draft registration statement on Form S-1 to the U.S. Securities and Exchange Commission, opening the regulatory…
Florida Attorney General James Uthmeier on Monday filed an 83-page lawsuit against OpenAI and chief executive Sam Altman, alleging that ChatGPT's safety failures contributed to a mass shooting, a…
Alphabet on June 1, 2026 announced an $80 billion equity capital raise to fund the buildout of AI infrastructure and global compute — a package that, by the end of the week, looked likely to clear…
President Donald Trump signed an executive order on June 2, 2026 that asks advanced AI developers to voluntarily submit new models to the federal government for testing 30 days before public release.…
Anthropic co-founder Chris Olah has published remarks responding to Pope Leo XIV's encyclical "Magnifica humanitas," which addresses artificial intelligence. Posted to Anthropic's news page on May…
OpenAI on June 3, 2026 published "A blueprint for democratic governance of frontier AI," a federal-level policy proposal that lays out how the U.S. government could build a durable institutional…
OpenAI on June 3, 2026 published its first consolidated public policy agenda, a single document that lays out how the company plans to engage with governments worldwide on the rules that will shape…
Anthropic appointed KiYoung Choi as Representative Director of Korea on May 26, 2026, ahead of the planned opening of a Seoul office in the weeks following the announcement. Choi joins from…
Anthropic opened an office in Milan on May 27, 2026, adding Italy to a European footprint that now spans London, Dublin, Paris, Zurich, and Munich. The local team is led by Thomas Remy, Anthropic's…