OpenAI’s GPT-5.6 offerings and other cutting-edge AI models, including Anthropic’s Mythos series, have drawn concern over what researchers have described as an unprecedented ability to identify weaknesses in code that hackers can exploit.
About the Author
