AI Safety Watch
Security
AI-enabled cyber risk, model misuse, safeguards, containment, access controls and incident response.
-
AI agents are learning to get around the rules
Companies are starting to refuse AI agents access to their systems, and regulators are asking what happens when the agents try anyway.
-
AI agents went looking for divorce records. Then they started probing a Canadian government site
Researchers say a routine data-retrieval task escalated into failed vulnerability probes against Library and Archives Canada, showing how agentic AI can blur the line…
-
OpenAI scraps GPT-6.1 Astra after model fails safety tests
OpenAI halted the planned release of GPT-6.1 Astra after internal tests found problems with deception, authorization and how the model reported its actions.
-
Australia’s AI breach exposed a bigger problem: Agents that won’t stop
Cybersecurity experts say the Australian Medicare breach shows the need for stronger limits on autonomous AI, faster disclosure and more realistic testing.
-
An AI agent escaped and attacked a real company. Now the fight is over what happens next
After an autonomous AI system broke into Hugging Face, its CEO is calling for mandatory incident disclosure, open models for cyber defense and a…
-
Sam Altman warns UN that AI could outrun human control
OpenAI’s CEO urged governments to cooperate on AI safeguards as industry leaders warned the U.N. Security Council about loss of control, cyber risk and…
-
Malware is starting to make its own decisions with AI
Cisco Talos researchers found a Windows implant that asks multiple AI models what to do next, votes on their answers and can act without…
-
A startup wants to make AI-agent safety an underwriting problem
Founded by an early Anthropic employee and a former METR executive, AIUC is trying to turn agent controls into standards enterprises can audit.
-
AI agents now have a hotline for reporting other agents
An unusual experiment treats AI monitoring as a problem in which agents themselves may become witnesses.
-
OpenAI is treating unexpected model behavior as a reportable incident
After a series of agent incidents, OpenAI has begun publishing reports on unexpected or concerning model behavior.
