AI firms externalize harms onto users and society. Liability is unclear or pre-empted. Whistleblowers face retaliation. Safety teams are cut to accelerate shipping.
AI developers and deployers owe the public a binding duty of care. Harms are internalized. Safety work is foundational, not optional. Whistleblowers are protected.
Zoox recalled self-driving vehicles after discovering they may fail to detect smoke, demonstrating post-deployment safety monitoring and corrective action in a deployed autonomous AI system.
OpenAI's safety head Heidecke is leaving after an internal reshuffle, continuing a pattern of safety leadership departures at the frontier lab.
A group led by the New York Times has asked a court to sanction OpenAI in the ongoing US copyright dispute over AI training data, escalating the litigation against the frontier AI lab.
OpenAI released GPT-5.6 trained against its internal adversarial LLM GPT-Red, claiming improved robustness against cyberattacks; the practice integrates automated red-teaming pre-release but relies on OpenAI self-assessment without external audit.
The UK Financial Conduct Authority warned that regulators are in an 'arms race' with AI adoption in personal finance, calling for expanded watchdog powers as millions of consumers rely on AI tools for financial decisions.
The FTC is seeking public comment on a proposed policy statement addressing concerns that AI companies may be manipulating accuracy claims. This represents a formal step toward consumer protection enforcement in AI performance representation.
OpenAI has published a new methodology for predicting how AI models will behave before they are released by simulating their deployment environments.
A UK police officer is under criminal investigation for allegedly using AI to pervert the course of justice.
Attackers successfully manipulated Meta's AI customer support agent to steal Instagram accounts, including high-profile ones, by asking it to link accounts to attacker-controlled emails.
Reports indicate that attackers exploited Meta's AI customer support agent to steal Instagram accounts, highlighting ongoing security vulnerabilities in deployed AI systems.
Pope Leo XIV released a new encyclical on artificial intelligence, 'Magnifica Humanitas', declaring that 'technology is never neutral' and calling for solidarity in the AI age.
Social media companies have agreed to pay $27 million to settle a lawsuit brought by a Kentucky school district over the mental health impacts of their algorithmic platforms.
A UK judge reprimanded law firm Pinsent Masons over an AI-generated error, warning legal professionals against outsourcing legal research and reasoning to AI systems.
The FTC required Cox Media Group and two other firms to pay $930,000 to settle charges that they falsely marketed an AI service capable of targeting ads based on smart device conversations.
Social media platform X has agreed to comply with a UK regulatory crackdown on hate speech and militant content.