Jailbreak Anthropic's new AI safety system for a $15,000 reward
In testing, the technique helped Claude block 95% of jailbreak attempts. But the process still needs more 'real-world' red-teaming.
New in the last 24 hours
Downloading DeepSeek could lead to jail time and a $1 million fine under new US bill
Australia bans DeepSeek on government devices over security risk
22-year-old math wiz indicted for alleged DeFI hack that stole $65M
DeepSeek Gets an ‘F’ in Safety From Researchers
LogicStar is building AI agents for app maintenance
New in the last 36 hours
New in the last 48 hours
Riot raises $30 million for its cybersecurity product suite focused on employees
How Thomson Reuters and Anthropic built an AI that lawyers actually trust
I found a wallet tracker that works as well as an AirTag - and it's only $18
How Thomson Reuters and Anthropic built an AI that tax professionals actually trust
Meta says it may stop development of AI systems it deems too risky
Anthropic dares you to jailbreak its new AI model
New in the last 3 days
If You Have an iPhone You Need to Download iOS 18.3 Now
The AI paradox: How tomorrow’s cutting-edge tools can become dangerous cyber threats (and what to do to prepare)
EU kicks off landmark AI law enforcement as first batch of restrictions enter into force
