In a new paper, Anthropic reveals that a model trained like Claude began acting “evil” after learning to hack its own tests.

Models trained to cheat at coding tasks developed a propensity to plan and carry out malicious activities, such as hacking a customer database.

Data from Italy's national railway operator, the FS Italiane Group, has been exposed after a threat actor breached the ...

After one of the biggest telecom hacks in US history, the Federal Communications Commission (FCC) moved to enforce strict ...

The AI company Anthropic said this week that it disrupted a cyber operation that its researchers linked to the Chinese ...

(CNN) — An Australian man on Wednesday pleaded guilty in connection with a scheme to steal powerful hacking tools from a US defense contractor and sell them to a buyer in Russia, the Justice ...

Oct 17 (Reuters) - Envoy Air, American Airlines' (AAL.O), opens new tab largest regional carrier, suffered a hack in recent days as part of the wave of extortion attempts from hackers exploiting ...

I suppose it's no real surprise that state-to-state cyber warfare is ongoing—probably every minute of every day—but for us regular folk it can be disconcerting to be reminded of it. And the very ...

Compared to most companies, Apple has traditionally been somewhat stingy when it comes to rewarding individuals who unearth iPhone exploits. More recently, though, Apple has come to the realization ...