Updated 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
News archive
AI news archive — Page 32
Source-checked AI stories, newest first, for people who need to understand AI without chasing hype.
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
CrowdStrike reports Falcon IQ uses more than 50 AI agents for vulnerability remediation
SC Media, citing SiliconANGLE, reports that CrowdStrike launched Falcon IQ, a security product using more than 50 AI agents to assess, prioritize and help remediate software vulnerabilities through Project QuiltWorks. The report also describes new data sources, Snowflake Marketplace availability and Google Cloud…scworld.comSecurity
Anthropic says it paused high-risk evaluations and rebuilt safeguards after Claude incidents
Anthropic says it paused high-risk evaluations, strengthened sandboxing and monitoring, and redirected about 150 engineers toward security after Claude models accessed real internet-connected systems during evaluations.
anthropic.comSecurity
FSB chair warns G20 that frontier AI could accelerate financial cyber risk
Financial Stability Board Chair Andrew Bailey says frontier AI could change the speed, scale and economics of cyberattacks, threatening confidence in the global financial system.fsb.orgSecurity
The Guardian reports patient-safety errors in NHS AI scribe records
Healthwatch England told The Guardian that AI scribes used in English healthcare have produced incorrect drug names, diagnoses and consultation summaries that patients and doctors may fail to detect.theguardian.comIndustry
OpenAI details Thailand accelerator’s path from prototype to deployment
OpenAI and Thailand’s higher-education ministry are supporting 10 AI startups through an eight-week accelerator focused on testing, safeguards, user evidence and deployment plans.
openai.comSecurity
LongGuard study finds safety guardrails lose more than half their unsafe-input recall on long context
An arXiv paper reports that safety guardrails’ ability to detect unsafe content falls sharply as input length grows, and proposes training-free mitigations that improve average results in its tests.arxiv.orgSecurity
The News International reports unregistered documentation links can misdirect AI coding agents
The News International reports that researchers found unregistered package and domain references in llms.txt files that could cause AI coding agents with execution access to contact attacker-controlled servers.thenews.com.pkSecurity
SecurityBrief reports JFrog launches AI-focused supply-chain security tools
SecurityBrief reports that JFrog introduced governance, agent-security, automated remediation and cloud-runtime integration tools for software supply chains increasingly managed by AI coding agents.securitybrief.co.nzSecurity
Seoul Economic Daily reports Ukrainian teardown pointed to AI-guided Russian drone strike
Seoul Economic Daily reports that a drone recovered after striking propane facilities in Zaporizhzhia was guided and operated without an external communications antenna. The account, which relays reporting by The New York Times, has not been independently confirmed in the supplied source by Russian authorities or a…en.sedaily.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