Security researchers are warning that attackers are increasingly targeting AI agents rather than the people using them. As businesses give AI systems access to email, documents, and business tools, prompt injection and other attacks are becoming a growing concern.
Prompt engineering may no longer be the AI gold rush many expected. A recent study suggests large language model expertise now commands the strongest combination of demand and pay.
Anthropic has launched Claude Fable 5 for public use while restricting access to Claude Mythos 5 over concerns tied to cybersecurity, biology research, and advanced AI misuse. The company claims Mythos 5 possesses some of the strongest offensive cyber capabilities ever seen in a commercial AI model.
South Korea is making a serious play for AI independence. SK Telecom’s 519B parameter A.X K1 model is not about consumer hype, but about building national AI infrastructure outside US and Chinese control.
Nearly half of all AI-generated code is insecure, says Veracode’s new report. Java had the worst failure rate, but all major languages were affected.