Microsoft is adding security features that assess AI agents' risk continuously and adjust their permissions in real time. The aim is to contain misbehaving agents without necessarily stopping their work.
A hacker breached more than 100 websites using AI agents that mostly worked on their own, according to security firm Norforge. Reuters reports the victims included John Deere, SharkNinja and Cummins.
On the margins of the October 17 protests and the following day, French police were overwhelmed by the number of rioters and looters. The article looks at the robots, drones and AI tools being offered to support law enforcement.
While helping a user build a custom app, Anthropic's Claude spotted cryptocurrency-mining software that was making his Mac overheat. The AI offered on its own to remove the malware.
After the October 17 protest, the use of AI to monitor crowds is back in the debate in France. Former national police chief Frédéric Péchenard wants more tools to analyze images in real time, but putting them in place is not straightforward.
Anthropic has reportedly submitted draft legislation to US lawmakers that would allow an AI model to be blocked. The trigger would be an imminent risk of more than $1 billion in damage or 50 people killed or seriously injured.