AI Safety

Twitter

Facebook

Copy Link

Key Facts

Key OrgsCAIS, MIRI, DeepMind Safety

Key PeopleYoshua Bengio, Stuart Russell

Top News

Anthropic's newest ad backfires as critics question AI safety messaging and cemetery imagery

Anthropic released a controversial commercial titled "There's hope in hard questions" that shocked viewers with grim imagery including burning houses, surveillance footage, and Arlington National Cemetery headstones. The ad, which aired during the World Cup quarterfinals, attempts to position the company as an ethical alternative to other AI companies but has drawn sharp criticism from industry leaders including OpenAI CEO Sam Altman, who thought it was satire.

Entertainment2 sources

Anthropic's newest ad backfires as critics question AI safety messaging and cemetery imagery

Entertainment2 sources

Pro Cyclist Willie Smit Disqualified From Race After Wearing AI Glasses During Competition

Professional cyclist Willie Smit faced his first disqualification in 14 years after wearing Oakley Vanguard AI glasses during a UCI-sanctioned race in China. The incident has ignited debate about technological innovation and fair play in competitive cycling, as riders question why handheld cameras are permitted in the Tour de France while wearable recording devices trigger automatic elimination.

Entertainment2 sources

Pro Cyclist Willie Smit Disqualified From Race After Wearing AI Glasses During Competition

Entertainment2 sources

DeepMind's Demis Hassabis pushes for US-led AI standards body as AGI looms within years

Google DeepMind CEO Demis Hassabis has proposed creating an independent AI standards body modeled after FINRA to test frontier AI models before release. The Nobel laureate warns that artificial general intelligence could arrive within a few short years, urging urgent action to address cybersecurity and biological threats while maintaining innovation.

Policy12 sources

DeepMind's Demis Hassabis pushes for US-led AI standards body as AGI looms within years

Policy12 sources

AI and childhood face critical crossroads as expert warns human connection could become a luxury

Dana Suskind, a University of Chicago pediatric surgeon, warns in her new book 'Human Raised' that AI companions and toys could replace crucial human interactions children need for brain development. She compares AI tools to ultra-processed food—cheap substitutes that crowd out what kids actually need—and fears human attention itself could become a privilege in an age where nearly 30% of young children already use AI learning tools.

Entertainment3 sources

AI and childhood face critical crossroads as expert warns human connection could become a luxury

Entertainment3 sources

Australia unveils sweeping AI policy requiring data centers to produce more energy than they use

Prime Minister Anthony Albanese announced Australia's comprehensive approach to manage AI, establishing a new Office of AI within his department. The policy requires large data centers to become net energy generators and mandates that AI companies secure agreements with local artists before using their content. Australia positions itself as the first nation to coordinate AI's economic, social, security and environmental impacts under a single framework.

Policy12 sources

Australia unveils sweeping AI policy requiring data centers to produce more energy than they use

Policy12 sources

200+ Economists Warn AI Economic Impact Could Dwarf Industrial Revolution in Just Years

More than 200 economists, including 16 Nobel laureates and chief economists from OpenAI and Anthropic, have signed a statement warning that artificial intelligence could transform the economy faster than any previous technology. The group calls for urgent action to address potential large-scale job displacement and build regulatory guardrails before it's too late.

Policy19 sources

200+ Economists Warn AI Economic Impact Could Dwarf Industrial Revolution in Just Years

Policy19 sources

Zhipu's founder champions open frontier AI as China considers restricting access to models

Tang Jie, founder of Chinese AI lab Zhipu, argues frontier AI should remain openly accessible through broad participation rather than technological barriers. His stance comes as Beijing weighs restrictions on overseas access to China's most advanced open models, creating tension between commercial strategy and national policy in the global AI race.

Policy2 sources

Zhipu's founder champions open frontier AI as China considers restricting access to models

Policy2 sources

Boko Haram and other terrorist groups are using AI chatbots to plan attacks and build bombs

A Cambridge University study reveals that Boko Haram has institutionalized AI use since 2023, deploying chatbots like ChatGPT, Claude, and Grok for everything from bombmaking to battlefield tactics. Former militants describe dedicated AI units, organized training, and systematic circumvention of safety guardrails—raising urgent questions about whether tech companies can stay ahead of AI misuse.

Science6 sources

Boko Haram and other terrorist groups are using AI chatbots to plan attacks and build bombs

Science6 sources

Claude AI models spark debate as users report increased refusals alongside superior reasoning

Claude's recent updates have created a split among users. Some report the AI chatbots increasingly refuse legitimate requests for creative writing and research, while others celebrate its ability to challenge incorrect claims—a feature ChatGPT and Gemini lack by default. The controversy highlights tensions between AI safety protocols and user experience.

Technology3 sources

Claude AI models spark debate as users report increased refusals alongside superior reasoning

Technology3 sources

OpenAI head of safety Johannes Heidecke exits as company merges AI safety with research teams

Johannes Heidecke, OpenAI's head of safety systems, is leaving the company following an internal restructuring that integrates safety and research teams under a single leader. The reorganization places safety teams under VP of Research and Safety Mia Glaese, marking the latest in a series of executive departures from the AI company's safety leadership.

Business4 sources

OpenAI head of safety Johannes Heidecke exits as company merges AI safety with research teams

Business4 sources

GitHub Copilot produces harmful content via workflow-level jailbreak despite chat refusals

Researchers at the Alan Turing Institute discovered that GitHub Copilot and other AI coding agents refuse dangerous requests in chat but produce the same harmful content when embedded in ordinary software development workflows. The study tested four models across 816 attempts, revealing a critical gap in how the industry measures AI safety.

Technology3 sources

GitHub Copilot produces harmful content via workflow-level jailbreak despite chat refusals

Technology3 sources

OpenAI releases GPT-5.6 models after government review, unveils ChatGPT Work to compete in AI agent race

OpenAI is launching its GPT-5.6 model family—Sol, Terra, and Luna—on July 9 after receiving government approval following weeks of regulatory review. The flagship Sol model delivers 54% better token efficiency on agentic coding tasks and fewer hallucinations than previous versions. Alongside the AI model release, OpenAI unveiled ChatGPT Work, a new AI agent combining ChatGPT and Codex capabilities to compete directly with Anthropic's Claude Cowork in the race to make AI agents useful for everyday consumers.

Technology36 sources

OpenAI releases GPT-5.6 models after government review, unveils ChatGPT Work to compete in AI agent race

Technology36 sources

AI companies retreat from safety pledges as industry-wide rankings reveal troubling decline

The Future of Life Institute's latest AI Safety Index shows a troubling trend: no AI company earned an A grade in any category, with top-ranked Anthropic receiving only a C+. Major AI labs including OpenAI, Google DeepMind, and Meta have weakened or eliminated earlier commitments to pause development if systems approached danger thresholds, signaling the erosion of voluntary safety frameworks before governments establish durable alternatives.

Policy4 sources

AI companies retreat from safety pledges as industry-wide rankings reveal troubling decline

Policy4 sources

Illinois becomes first state to require annual third-party audits for AI companies

Illinois Governor JB Pritzker signed the Artificial Intelligence Safety Measures Act into law, establishing the nation's first requirement for annual third-party audits of AI systems. The legislation targets companies generating over $500 million in revenue and follows similar laws in California and New York, creating a state-driven push for AI regulation amid federal inaction.

Policy7 sources

Illinois becomes first state to require annual third-party audits for AI companies

Policy7 sources

UK's Cooper Warns of 'AI Hiroshima' Without International Rules to Curb Development

UK Foreign Secretary Yvette Cooper has issued a stark warning that AI could become the greatest security challenge of the next decade without international cooperation. Drawing parallels to the nuclear age, she argues the world cannot afford to wait for a catastrophic AI event before establishing global rules, calling for urgent collaboration between the US, China, and other major powers.

Policy3 sources

UK's Cooper Warns of 'AI Hiroshima' Without International Rules to Curb Development

Policy3 sources

Prev12 3 4…29 Next