Anthropic says its cybersecurity-focused model, Claude Mythos Preview, was able to convert publicly available security patches into functioning exploits within an hour, dramatically compressing the window defenders have long relied on to stay ahead of attackers. In a research exercise, the model built full remote code execution exploits for Mozilla Firefox and forged privilege-escalation chains that reached the highest level of access on Microsoft Windows, illustrating how quickly the gap between a patch release and a usable attack can now close.
Continue reading
The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.
Already purchased? Sign in✓ Signed in — this article isn’t included in your current plan.