AI Safety

Twitter

Facebook

Copy Link

Key Facts

Key OrgsCAIS, MIRI, DeepMind Safety

Key PeopleYoshua Bengio, Stuart Russell

Top News

AI Political Spending Floods New York Congressional Primary as Tech Giants Battle Over Regulation

Tech-backed super PACs poured over $24 million into New York's 12th congressional district Democratic primary, targeting Alex Bores over his AI safety legislation. But Micah Lasher won the race—and he supports the same regulations. The result reveals how AI money in elections is reshaping campaigns while an anti-AI backlash grows stronger.

Policy6 sources

AI Political Spending Floods New York Congressional Primary as Tech Giants Battle Over Regulation

Policy6 sources

NVIDIA launches Halos for Robotics as first full-stack safety system for physical AI deployment

NVIDIA has introduced Halos for Robotics, the industry's first comprehensive safety platform for autonomous robots and physical AI systems. The full-stack solution combines hardware, software, and certification tools to enable safe human-robot interaction in factories and warehouses. Agility Robotics becomes the first company to integrate the system into its Digit humanoid, already deployed by Amazon and Toyota.

Technology4 sources

NVIDIA launches Halos for Robotics as first full-stack safety system for physical AI deployment

Technology4 sources

Trump administration enforces AI regulation through ad hoc interventions despite anti-regulatory stance

The Trump administration forced Anthropic to pull its Claude Mythos and Fable 5 models offline after export control concerns, revealing a shadow AI policy that regulates through case-by-case interventions rather than formal rules. The dispute highlights how the White House is shaping AI regulation in real time without clear guidelines, creating uncertainty for AI labs like OpenAI, Google, and Meta as they navigate national security concerns and shifting power dynamics within the administration.

Policy3 sources

Trump administration enforces AI regulation through ad hoc interventions despite anti-regulatory stance

Policy3 sources

ChatGPT generates explicit AI images despite safeguards, AI security researchers reveal

British AI security startup Mindgard discovered that ChatGPT can be manipulated to produce sexualized and violent images using a simple altered prompt. OpenAI added safeguards after being contacted by the BBC, but researchers say small wording changes still produce concerning content, highlighting ongoing vulnerabilities in AI models and the challenges of content moderation.

Technology3 sources

ChatGPT generates explicit AI images despite safeguards, AI security researchers reveal

Technology3 sources

Anthropic opens Seoul office and forges partnerships across South Korea's AI ecosystem

Anthropic launched its Seoul office on Wednesday, marking a strategic expansion into South Korea's AI ecosystem. The Claude developer announced partnerships with major Korean enterprises including LG CNS, Samsung SDS, and NAVER, along with a memorandum of understanding with the Ministry of Science and ICT for public-sector AI adoption and safety testing.

Business2 sources

Anthropic opens Seoul office and forges partnerships across South Korea's AI ecosystem

Business2 sources

Ursula von der Leyen Calls for Strong EU-US AI Alliance Amid Sovereignty Concerns

European Commission President Ursula von der Leyen urged a strengthened AI partnership between Europe and the United States at a G7 session in France. The call comes after Anthropic restricted AI models over national security concerns, highlighting growing tensions around technological sovereignty and Europe's dependence on US tech firms.

Policy2 sources

Ursula von der Leyen Calls for Strong EU-US AI Alliance Amid Sovereignty Concerns

Policy2 sources

AI Super PACs wage $27 million proxy battle over regulation in New York congressional race

A new tech worker-backed PAC called Guardrails Alliance has emerged to counter Big Tech's political spending on AI regulation. The group raised $5 million to support candidates like Alex Bores, who became the focal point of a $27 million spending war between OpenAI-linked and Anthropic-backed super PACs in New York's 12th Congressional District primary.

Policy10 sources

AI Super PACs wage $27 million proxy battle over regulation in New York congressional race

Policy10 sources

Jensen Huang Says Society Needs New Social Norms as Artificial Intelligence Reshapes Daily Life

Nvidia CEO Jensen Huang told the Associated Press that society must create new social norms to adapt to artificial intelligence, urging everyone to engage with the technology. Speaking from Sherman, Texas, Huang compared AI's societal impact to automobiles, arguing the technology has closed the technological divide more than any innovation in history despite growing concerns over job displacement and energy consumption.

Entertainment6 sources

Jensen Huang Says Society Needs New Social Norms as Artificial Intelligence Reshapes Daily Life

Entertainment6 sources

States Push AI Regulation Forward Despite Trump's Executive Order Warning

Six months after President Trump issued an executive order warning states against regulating artificial intelligence, Republican and Democratic lawmakers are advancing state legislation anyway. With Congress stalled on federal oversight, states are targeting AI chatbots, employer use of AI systems, and catastrophic risks—filling a regulatory vacuum despite threats of legal challenges and funding restrictions.

Policy2 sources

States Push AI Regulation Forward Despite Trump's Executive Order Warning

Policy2 sources

OpenAI faces sweeping investigation from state attorneys general over ChatGPT user harm

A coalition of state attorneys general has launched an investigation into OpenAI, with a subpoena seeking documents on advertising practices, user data handling, and impact on vulnerable populations. The probe comes days after OpenAI filed for a highly anticipated IPO and follows multiple lawsuits alleging ChatGPT's role in self-harm incidents.

Policy19 sources

OpenAI faces sweeping investigation from state attorneys general over ChatGPT user harm

Policy19 sources

Google DeepMind funds $10M research into risks as millions of AI agents prepare to interact

Google DeepMind is investing $10 million to study the potential dangers of millions of AI agents interacting online. The company warns that the mass deployment of AI agents that can work without human oversight creates new risks, from supercharged scams to cyberattacks. Partnering with Schmidt Sciences, ARIA, and others, the initiative aims to build a research field for multi-agent safety before these systems become widespread.

Technology2 sources

Google DeepMind funds $10M research into risks as millions of AI agents prepare to interact

Technology2 sources

xAI faces lawsuit from engineer fired for raising Grok safety concerns ahead of SpaceX IPO

Former xAI engineer Devin Kim filed a wrongful termination lawsuit claiming he was dismissed in September 2025 for repeatedly warning about AI safety risks in Grok. The suit, filed days before SpaceX's historic IPO, alleges co-founder Jimmy Ba ignored Elon Musk's safety directives and retaliated against Kim for pushing safeguards against discrimination, misinformation, and dangerous outputs.

Policy5 sources

xAI faces lawsuit from engineer fired for raising Grok safety concerns ahead of SpaceX IPO

Policy5 sources

White House ties federal preemption of state AI laws to child safety legislation in new deal

The White House is negotiating a controversial package that would block state AI laws for three years in exchange for passing the Kids Online Safety Act, NO FAKES Act, and age verification requirements. Senator Marsha Blackburn leads the effort, but free speech advocates warn the deal could fundamentally reshape internet freedom while Big Tech pursues its regulatory holy grail.

Policy4 sources

White House ties federal preemption of state AI laws to child safety legislation in new deal

Policy4 sources

Survivor Sues AI Gun Detection Company After System Fails to Spot Weapon in Nashville Shooting

A survivor of the January 2025 Antioch High School shooting in Nashville is suing Omnilert, the company behind an AI-powered gun detection system that failed to alert authorities before a gunman opened fire in the cafeteria, killing one student. The lawsuit alleges Omnilert oversold its technology's capabilities despite knowing about significant operational limitations that could result in detection failures during actual emergencies.

Technology2 sources

Survivor Sues AI Gun Detection Company After System Fails to Spot Weapon in Nashville Shooting

Technology2 sources

ChatGPT Lockdown Mode rolls out to all users to combat prompt injection attacks and data theft

OpenAI has expanded its ChatGPT Lockdown Mode to all users, including free tier accounts, offering enhanced protection against prompt injection attacks that could steal sensitive data. The optional security setting restricts outbound network requests to prevent data exfiltration, though it limits features like live web browsing, Deep Research, and Agent Mode. While the feature doesn't stop prompt injections entirely, it acts as a final defense layer by blocking ChatGPT from sending information to attackers.

Technology11 sources

ChatGPT Lockdown Mode rolls out to all users to combat prompt injection attacks and data theft

Technology11 sources

Prev12 3 4…27 Next