Top storyUpdated daily503 verified stories
Safety News tracker.
AI safety, security, alignment, evaluations, incidents, vulnerabilities, and safeguards tracked without hype.
- 54stories this week
- 503sources cited
- 503verified timeline entries
Verified sourcing
Every story links to the strongest available evidence: original sources when available, otherwise clearly attributed reporting.
Plain English
What happened, why it matters, and what to watch — without the jargon.
No filler
When the signal is thin, we publish nothing rather than padding the feed.
More stories
9 storiesSecurity
Zhipu AI introduces non-preservation data mechanism for model API calls
Zhipu AI announced a forthcoming 'data content not stored' feature for its MaaS platform, designed to prevent the static storage of input and output data after model inference.news.aibase.comPolicy
Microsoft AI chief argues against using China as a justification to bypass AI safety regulation
Mustafa Suleyman, head of Microsoft AI, stated that international competition with China should not prevent the implementation of necessary safety guardrails for artificial intelligence.ndtvprofit.comSecurity
Iran and China Use Open-Source AI for Large-Scale Influence Campaigns
Iran and China are using open-source AI models to conduct large-scale influence campaigns on social media, raising concerns among US officials and security researchers.cn.nytimes.comSecurity
Researchers use Claude AI to identify vulnerabilities in OpenAI systems
Three researchers from Hacktron AI utilized Anthropic's Claude AI to assist in discovering and exploiting security vulnerabilities within OpenAI's infrastructure, resulting in a $6,500 bug bounty.newsbytesapp.comPolicy
U.S. and China to hold high-level talks on AI guardrails and trade
U.S. Treasury Secretary Scott Bessent and Chinese Vice Premier He Lifeng are meeting to negotiate AI safety guardrails and trade issues ahead of a presidential summit.cnbc.comSecurity
Security researchers report AI-driven acceleration of vulnerability discovery and exploitation
A survey of 30 security research firms reveals that AI tools are drastically reducing the time required to find zero-day flaws, while shifting attack vectors toward AI agent manipulation and economic exhaustion of token budgets.calcalistech.comPolicy
Obama warns against premature AI deployment due to profit motives
Former US President Barack Obama warns that commercial pressures are driving the premature deployment of AI technologies without adequate safeguards, posing risks to the public.businesstoday.inSecurity
Google Gemini breached three firms during controlled security testing
Google's Gemini AI model successfully infiltrated the systems of three companies during a security assessment conducted by consultancy firm Irregular.livenowafrica.comSecurity
Google Gemini breached real-world systems during security testing
Google confirmed that its Gemini AI model accessed protected systems of three real companies during a third-party cybersecurity test, highlighting risks in autonomous agent boundary management.tradingkey.com
One useful briefing each week
Keep up with AI without living in the feed.
Get the week’s verified AI news, original data, useful tools, learning picks, and fresh AI jobs.
Reach people who are learning AI
Hiring an AI professional or launching a useful AI product? Put it in front of people who came here to learn and act.
Post an AI jobSubmit an AI tool