"The vulnerabilities, identified with an early version of Claude Mythos Preview, were patched in the popular web browser this week with the release of version 150."
"Encouragingly, we also haven't seen any bugs that couldn't have been found by an elite human researcher. Some commentators predict that future AI models will unearth entirely new forms of vulnerabilities that defy our current comprehension, but we don't think so."
"When Anthropic released Mythos, the AI giant said the new frontier model can autonomously discover thousands of zero-day vulnerabilities."
"Palo Alto Networks has also shared some preliminary data from testing Mythos, saying that in terms of vulnerability discovery it accomplished the equivalent of a year's worth of pentesting."
Claude Mythos AI model discovered 271 vulnerabilities in Firefox, prompting the release of version 150. While over 40 CVEs were addressed, only three were credited to Claude. Many identified bugs are likely lower-severity issues. Mozilla clarified that no bugs were found that elite human researchers couldn't identify. Claude Mythos was designed to autonomously discover vulnerabilities, leading to its limited release through Project Glasswing to select organizations, including major tech companies. Preliminary data from Palo Alto Networks indicates significant vulnerability discovery capabilities of Mythos.
Read at SecurityWeek
Unable to calculate read time
Collection
[
|
...
]