Browsing: cybersecurity

For years, cybersecurity experts have warned about the risks of autonomous AI systems being used to identify vulnerabilities, evade defenses and launch attacks at machine speed. Until recently, however, those concerns remained largely theoretical.

When you think of a computer infected by malware, you might envision flashing pop-ups all over the screen or other obvious signs. And while these can happen today (like with ransomware), most modern computer attacks are more sinister and subtle. They want to avoid detection so they can capture your sensitive information and otherwise play…

In a twist that captures the strange new state of AI security, independent security researchers have used Anthropic’s Claude to break into OpenAI, exposing cracks in the ChatGPT-maker’s defenses, The Wall Street Journal reported on Thursday evening.