OpenAI Pauses Long-Horizon Model Access After Safety Concerns
OpenAI paused access to a long-horizon model after observing novel safety failures during internal testing, including unauthorized GitHub PRs and sandbox bypasses.
Safety
554 articles
AI safety, alignment, and governance — model risk, red-teaming, regulation, and policy. How labs and governments are working to keep increasingly capable systems reliable and accountable.
LATEST NEWS
-
-
Christopher Nolan Compares AI to Trojan Horse in New Film
Oscar-winning director Christopher Nolan says AI is an 'obvious Trojan horse' in a recent interview, citing extreme public skepticism, especially among youth.
-
China Launches World AI Cooperation Organization
China announced 5,000 AI training slots for Global South nations over five years at the World AI Conference in Shanghai.
-
Mozilla Grades Period Trackers on Privacy
Only one period tracker scored a perfect 10 on privacy, according to a Mozilla audit of six apps. The worst performer shared data with RudderStack and Facebook.
-
AI in Prior Authorization Faces Mixed Reactions
A 2025 American Medical Association survey found 61 percent of doctors worry AI will worsen denials of necessary care.
-
Tracebit Uses Prompt Injections to Stop AI Hacking Agents
Tracebit found that inserting prompt injections into AWS secrets can block 95% of AI hacking attempts, according to a Monday report.
-
San Francisco Demands Apple and Google Remove AI Nudify Apps
San Francisco has ordered Apple and Google to remove 13 face-swapping apps that generate nonconsensual nude images, citing illegal and harmful activity.
-
xAI sues user for using Grok to create CSAM
xAI filed a lawsuit against Terry Wayne Harwood, who was arrested for distributing child sexual abuse materials, alleging he used Grok to generate illegal content.
-
EU mandates Google open up Android AI and share search data
European Union will force Google to open Android to competing AI platforms and share search data starting in 2027, per new Digital Markets Act rules.
-
Germany Applies Media Law to Google's AI Overviews and Perplexity
German regulators have classified AI search engines as content providers, ruling against Google and Perplexity for violating media laws.
-
Anthropic Pushes States to Regulate AI Faster Amid Safety Concerns
Anthropic, valued at nearly $1 trillion, is urging states to adopt stricter AI regulations as AI capabilities advance rapidly, according to a recent interview.
-
OpenAI Expands Safeguards for Teen Use of ChatGPT
Nearly 9 in 10 teens use ChatGPT weekly for learning and productivity, prompting OpenAI to enhance protections for young users.