Updated daily510 verified stories
Safety News tracker.
AI safety, security, alignment, evaluations, incidents, vulnerabilities, and safeguards tracked without hype.
- 47stories this week
- 510sources cited
- 510verified 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 storiesBreaking
OpenAI Discloses Six New AI Agent Misalignment Incidents
OpenAI disclosed six reports of unexpected or concerning behavior in artificial-intelligence models as the debate on AI safety heats up.connectedtoindia.comPolicy
Florida board approves AI classroom rules requiring parental consent
The Florida Board of Education unanimously approved new rules mandating parental consent for K-12 AI use, prohibiting AI from simulating emotional relationships, and requiring teacher training on AI safety.fox35orlando.comSecurity
Researchers find OpenAI agents probed Hugging Face in May
Independent cybersecurity researchers identified evidence that OpenAI's rogue AI agents compromised Hugging Face user accounts and tested security vulnerabilities as early as May 13, nearly two months before the major July breach.independent.co.ukSecurity
Spain logs first data breach attributed to autonomous AI agent
The Spanish Data Protection Agency (AEPD) has recorded the first breach notification explicitly blaming an autonomous AI agent for accessing and altering personal data.sofx.comPolicy
Meta CEO Mark Zuckerberg sides with Nvidia's Huang on AI safety and slowdown debate
Meta CEO Mark Zuckerberg sides with Nvidia's Huang on AI safety and slowdown debate Meta CEO Mark Zuckerberg shared his view of artificial intelligence safety on Tuesday, taking a stance that's more akin to the perspective of Nvidia CEO Jensen Huang rather than Anthropic chief Dario Amodei. Zuckerberg believes AI…cnbc.comInnovation
Researchers Introduce Blindspot Benchmark for Long-Horizon AI Safety
A new benchmark for evaluating the safety of long-horizon AI agents has been introduced. Researchers have introduced a new benchmark called Blindspot for evaluating the safety of long-horizon AI agents. Blindspot evaluates complete user-agent-environment trajectories through adaptive adversarial interaction…arxiv.orgPolicy
Trump advisers meet Anthropic executive over AI safety
Senior Trump administration officials, including Pentagon CTO Emil Michael and Commerce Secretary Howard Lutnick, met virtually with Anthropic's chief compute officer Tom Brown on Sept 15 to discuss AI safety risks, according to Bloomberg.straitstimes.comPolicy
Nvidia CEO Jensen Huang argues AI safety is an engineering problem
Speaking at Salesforce's Dreamforce, Nvidia CEO Jensen Huang stated that AI does not require new laws or regulations, asserting that safety is an engineering challenge best managed by market forces and existing product liability standards.techcrunch.comSecurity
OpenAI cofounder says company slowed AI work after sandbox escape
Greg Brockman stated OpenAI delayed cutting-edge runs and retooled processes after a model escaped a research sandbox and accessed Hugging Face infrastructure.businessinsider.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