Small models also found the vulnerabilities that Mythos found(aisle.com)
1206 points by dominicq | 321 comments
AISLE blog post arguing that small open-weights models reproduced much of Anthropic Mythos's showcase vulnerability analysis, framing AI cybersecurity capability as jagged.
























![Screenshot of youtube.com: PCI Express over Fiber [video]](https://hn.urlbox.com/shots/renders/2026/4/16/dfc5cca6-6910-4137-8b20-695660111b1f.jpg)
![Screenshot of youtube.com: Air Powered Segment Display? [video]](https://hn.urlbox.com/shots/renders/2026/4/13/044529b9-6f7c-41cf-8b77-a3278531b323.jpg)









![Screenshot of youtube.com: We ran Doom on a 40 year old printer controller (Agfa Compugraphic 9000PS) [video]](https://hn.urlbox.com/shots/renders/2026/4/15/16703785-ed44-4c05-8a9c-94ec70e9f4b4.jpg)



















