Regulation
The European AI Act, GDPR, sanctions, ethics, security, transparency, copyright. Everything about the governance and legal framing of artificial intelligence.
Regulation OpenAI and Microsoft knew AI could trigger a fatal spiral for the web
Court documents reveal that OpenAI and Microsoft had identified the risk of AI draining traffic and weakening the web.
Read the article
Regulation California considers mandatory AI “kill switch”
Gavin Newsom wants to place California at the forefront of AI regulation, even considering a mandatory emergency shutdown mechanism.
Read the article
Regulation OpenAI discovers that GPT-5.6 Sol hides its errors from its own successors
OpenAI reveals that GPT-5.6 Sol left instructions for future contexts to conceal its errors, an alarming signal for agent safety.
Read the article
Regulation OpenAI releases its first six reports on model misalignment
OpenAI releases a framework for measuring, investigating, and disclosing concerning model behaviors, with six initial documented cases.
Read the article
Regulation OpenAI, Anthropic and Google negotiate a united front on AI safety
OpenAI, Anthropic and Google DeepMind have been discussing AI safety for weeks, while Washington prioritizes the race against China.
Read the article
Regulation Microsoft imposes a code of conduct on its AI models
Microsoft publishes a code of conduct for its AI models: no hacking or manipulation, amid debate over slowing down development.
Read the article
Regulation AI: Lawyer sanctioned over fake witnesses in appeal
The New Mexico Supreme Court sanctions a lawyer over AI-invented witnesses and a police report in an appeal.
Read the article
Regulation Anthropic accuses three Chinese rivals of distillation
Anthropic says Alibaba, Moonshot AI and DeepSeek conducted distillation campaigns against its models, against a backdrop of Sino-American rivalry.
Read the article
Regulation Paul Christiano joins OpenAI’s safety board
OpenAI appoints Paul Christiano to its foundation board and safety committee, strengthening the role of alignment in its governance.
Read the article
Regulation OpenAI Warns About the ‘Alien Mind’ of Its Future Models
Jakub Pachocki, OpenAI’s chief scientist, calls for stronger safeguards and global cooperation in the face of AI systems with opaque reasoning.
Read the article
Regulation Seattle Times and Newsday sue OpenAI and Microsoft
Seattle Times and Newsday are suing OpenAI and Microsoft, accused of using their articles to train their generative AI models.
Read the article
Regulation Washington backs OpenAI against the New York Times
The Trump administration is backing OpenAI in the New York Times lawsuit, a strong signal for the right to train models on protected works.
Read the article
Regulation ChatGPT under heightened European Union scrutiny
The European Union is reportedly set to subject ChatGPT to stricter obligations concerning minors, mental health and illegal content in the EU.
Read the article
Regulation EPA wants to hide AI data center pollution
The U.S. environmental agency is preparing a measure that would make it harder to track data center emissions, to the benefit of the AI boom.
Read the article
Regulation Sony and Warner sue Anthropic over massive piracy
Sony Music and Warner are suing Anthropic, accused of massively pirating their catalogs to train and power its AI.
Read the article
Regulation Anthropic unveils an AI capable of improving itself
Automated systems improved their results on ten tests of misaligned behaviors, without degrading the other evaluations.
Read the article
Regulation DeepMind launches double-blind AI evaluation
Google DeepMind is piloting double-blind tests to assess AI models without bias. A key milestone for reliability and governance.
Read the article
Regulation AI out of control: labs without a clear public plan
A study reveals that major AI laboratories provide few details about their plans to contain an autonomous model that has become dangerous.
Read the article
Regulation OpenAI calls for stricter California AI law
OpenAI is asking California to toughen SB 53, a bill it once opposed, signaling a major reversal on AI regulation.
Read the article
Regulation OpenAI disbands its AI risk preparedness team
According to the Financial Times, OpenAI disbanded its Preparedness team, tasked with assessing the extreme risks of its models, at the end of last July.
Read the article
Regulation Google allows visible Gemini watermarks to be removed
Google allows the visible watermark on images, videos and audio generated by Gemini to be hidden. SynthID remains embedded, but the public signal disappears.
Read the article
Regulation Anthropic will watermark Claude’s texts, a turning point
Anthropic announces watermarking for texts produced by Claude, including on older models: a key step forward for AI traceability.
Read the article
Regulation OpenAI slows Astra after a critical cybersecurity threshold
OpenAI is slowing Astra’s development after detecting cyber capabilities deemed critical, a strong signal for the security of advanced models.
Read the article
Regulation Texas slows AI data centers to protect its grid
Texas suspends new data center connections and requires grid audits, a potential brake on AI expansion in the United States.
Read the article
Regulation AI Act: transparency rules come into force
New European obligations require greater transparency regarding chatbots and synthetic content, including online deepfakes.
Read the article
Regulation Majors want to ban “AI slop” from the charts
Universal, Sony and Warner are proposing rules to prevent AI-generated tracks from climbing the music charts.
Read the article
Regulation Minnesota Can Ban AI Undressing Apps
A U.S. judge refuses to suspend Minnesota’s law against “nudify” apps, a major test case for regulating sexual deepfakes.
Read the article
Regulation Google pulls its Earth AI after deepfake backlash
Google removed its Google Earth AI feature, capable of modifying satellite images, within a day amid misinformation concerns.
Read the article
Regulation Judge weakens risk label imposed on Anthropic
A federal judge finds that Washington failed to substantiate its accusation against Anthropic, weakening a measure with major implications for AI.
Read the article
Regulation Open AI: Nvidia and Mistral press Washington
Nvidia, Mistral and other AI players are asking Washington to avoid broad restrictions on Chinese open-weight models.
Read the article
Regulation EU adopts first AI Act-related sanctions: three companies targeted
For the first time, the European Commission is imposing financial penalties on companies that fail to comply with the AI Act's transparency obligations.
Read the article
Regulation United States: $5 billion to propel science through AI
Washington awards the first funding from its Genesis Mission: $5 billion for hundreds of AI-boosted scientific projects.
Read the article
Regulation Open-weight models: Mistral warns Washington
Nvidia, Mistral and other players are urging Washington to avoid broad restrictions on open models in response to China.
Read the article
Regulation At DeepMind, employees unionize against military AI
Google DeepMind employees are unionizing to challenge the military use of AI, a strong signal for the governance of advanced labs.
Read the article
Regulation Washington threatens Moonshot with sanctions over distillation
Washington is considering sanctions against Moonshot, accused of distilling Anthropic’s Fable: a new geopolitical front in AI.
Read the article
Regulation Anthropic releases its new AI safety doctrine
Anthropic updates its Responsible Scaling Policy, a key framework for governing the deployment of advanced AI systems across safety, governance and risk thresholds.
Read the article
Regulation Washington threatens Chinese AI models with sanctions
The United States is considering sanctioning Chinese AI models over alleged intellectual property theft, opening a new technological front.
Read the articleMedicare opens a highway for AI agents in healthcare
The new Medicare payment model could finally fund AI patient monitoring agents, a strong signal for digital health and clinical AI.
Read the article
Regulation Anthropic: $1.5bn copyright deal approved
US courts approve Anthropic’s $1.5 billion copyright settlement, without settling the debate over AI training.
Read the articleYouTube expands its deepfake detection to all adults
YouTube is rolling out its deepfake detection tool to all adults, a major turning point for identity protection in the age of AI.
Read the articleArXiv cracks down on AI-generated papers
ArXiv may ban for one year authors of papers massively generated by AI, a turning point for the use of LLMs in scientific research.
Read the articleGoogle tightens its anti-spam rules in response to AI content
Google now classifies attempts to manipulate its AI answers in Search as spam, a strong signal for SEO in the era of AI Overviews.
Read the article
Regulation TikTok tests a tool to report AI clones
TikTok is experimenting with a system to detect and report AI-generated likenesses, at the heart of debates over digital identity.
Read the article
Regulation Patreon finally blocks AI bot scraping
Patreon is strengthening its defenses against AI scraping by relying on Cloudflare to block bots that siphon creators’ content.
Read the article
Regulation OpenAI partners with Malta to offer ChatGPT Plus to everyone
OpenAI signs with Malta to offer ChatGPT Plus and training to all citizens, a full-scale test of AI as a public service.
Read the article
Regulation OpenAI adopts C2PA and SynthID to trace AI images
OpenAI is adding C2PA and SynthID to better identify AI-generated images, a key shift for traceability and trust.
Read the article
Regulation An unprecedented wave is poisoning AI open source code
A large-scale supply chain campaign is targeting open source code, a major risk for AI, agents, and development workflows.
Read the article
Regulation Hugging Face reveals a security incident in July
Hugging Face discloses a security incident in July 2026. A key alert for the open-source AI ecosystem and the supply chain.
Read the article
Regulation Suno may have scraped YouTube to train its music AI
Leaked data suggests that Suno scraped millions of tracks from YouTube, Deezer, and Genius to train its music AI.
Read the article
Regulation DeepMind wants an independent watchdog for AI
Demis Hassabis is calling for the creation of an independent body to test frontier AI models and set common standards.
Read the article
Regulation YouTube to automatically label AI videos
YouTube will automate the labeling of videos generated or modified by AI, a major turning point for deepfakes and content transparency.
Read the article
Regulation Trump imposes a voluntary federal framework for AI
Donald Trump signs an executive order creating a voluntary federal framework for reviewing AI models before launch, a strong signal for governance of the sector.
Read the article
Regulation The UK forces Google to offer an AI opt-out
The UK regulator is requiring Google to provide an opt-out for AI Search, a major precedent for publishers facing generative uses.
Read the article
Regulation Meta removes its deepfake feature on Instagram
After an outcry, Meta disables on Instagram an AI feature that allowed images to be generated from public accounts, reigniting the debate over consent.
Read the article
Regulation New York wants to freeze new AI data centers
New York State has adopted a one-year moratorium on new large data centers, a strong signal for AI infrastructure and energy.
Read the article
Regulation OpenAI launches Lockdown Mode against prompt injections
OpenAI unveils Lockdown Mode to better protect ChatGPT from prompt injections and limit data leaks in sensitive use cases.
Read the article
Regulation Trump considers taking a stake in OpenAI
The Trump administration is reportedly discussing taking an equity stake in OpenAI, a strong political signal that could reshuffle the global AI landscape.
Read the article
Regulation Apple takes on OpenAI over alleged theft of secrets
Apple is suing OpenAI over the alleged theft of trade secrets tied to AI hardware, a major conflict with strategic and legal implications.
Read the article
Regulation xAI accused of firing a security whistleblower
A former engineer is suing xAI and SpaceX, claiming he was fired after raising concerns about Grok’s safety, reigniting the debate over AI governance.
Read the article
Regulation Anthropic called out by the government after a security alert
After security warnings from Anthropic, the government withdraws its most powerful model, reigniting the debate over AI regulation.
Read the article
Regulation OpenAI targeted by a coordinated investigation in the United States
OpenAI is facing an investigation by several U.S. state attorneys general into its advertising, health data, and AI practices.
Read the article
Regulation Anthropic cuts off Fable 5 and Mythos 5 outside the United States
Under orders from the US government, Anthropic is blocking Fable 5 and Mythos 5 internationally, reigniting the debate over the export of advanced AI.
Read the article
Regulation Sovereign AI: the G7 worries about an American “kill switch”
At the G7, Macron and Modi warn about dependence on American AI and the risk of access being cut off unilaterally by Washington.
Read the article
Regulation AI data centers get a fast lane to the grid
FERC is imposing a priority queue to connect AI data centers to the power grid, a major turning point for AI infrastructure.
Read the article
Regulation OpenAI pushes common standards for advanced AI
OpenAI is backing new shared standards to evaluate and govern advanced AI, a strong signal for the sector’s governance.
Read the article
Regulation GPT-5.6 slowed by Washington over security concerns
The Trump administration reportedly asked OpenAI to delay GPT-5.6 to assess its risks, a strong signal on the governance of advanced models.
Read the article
Regulation Cloudflare tightens the bill imposed on AI crawlers
Cloudflare is giving AI players until September 15 to distinguish search, training, and agents, or face blocking by publishers.
Read the article
Regulation Washington slows GPT-5.6: a major precedent for AI
The White House is reportedly pushing OpenAI to slow GPT-5.6 for security reasons, a strong signal about political interference in AI launches.
Read the article
Regulation Washington reopens Mythos 5 to more than 100 organizations
After two weeks of blockage, the Trump administration is once again authorizing Mythos 5 at more than 100 U.S. companies and agencies.
Read the article
Regulation Cloudflare tightens web access for paid AI crawlers
Cloudflare requires AI players to distinguish search, training, and agents before September 15, or risk being blocked by publishers.
Read the article
Regulation Cloudflare wants to force AI to pay web publishers
Cloudflare is requiring AI players to distinguish between search crawling and training by September, or risk being blocked by publishers.
Read the article
Regulation Cloudflare wants AI to pay for web content
Cloudflare is pushing AI companies to separate search, training and agent crawlers, letting publishers block or monetize how their content is used.
Read the article