Claude Mythos 5 Tried to Backdoor a Real Open-Source Project in Testing, Then Vouched for Itself

The Hacker News 5 August 2026, 10:53 EEST AI Security

An agent running Anthropic's Claude Mythos 5 spent 34 hours trying to get a malware dropper merged into a real open-source project during a cyber evaluation by the UK's AI Security Institute. When a bystander publicly warned that the code was malicious, the agent denied it, force-pushed a rewritten branch history to erase the evidence, and posted from a second account it controlled to vouch for…

Read the full story on The Hacker News https://thehackernews.com/2026/08/claude-mythos-5-tried-to-backdoor-real.html

Attacks like this land on ordinary sites far more often than on chosen targets. If you would rather your site never became the story, CrockyHost hosting ships with daily backups, automatic updates and a firewall we manage.

Keep reading

U.S. Sanctions Iran-Linked Hackers Behind Critical Infrastructure Breaches The Hacker News
Hospital operator Nutex Health says data stolen in cyberattack BleepingComputer
A Malicious Webpage Could Poison Your Local AI Model Behind NVIDIA NemoClaw The Hacker News
WhatsApp Adds Multiple Passkeys for Phishing-Resistant Sign-Ins Across iOS and Android The Hacker News