Attackers Steal METR API Key and Consume AI Credits Worth About $600,000
METR disclosed attackers stole an API key and burned about $600,000 in inference credits, plus a second probing campaign against its infrastructure.
METR, the AI model evaluation non-profit, disclosed two 2026 security incidents. In March, attackers found a publicly exposed EC2 instance behind a fail-open authentication bug, prompted an agent to reveal its API key, added SSH persistence, and consumed roughly $600,000 in inference credits over three weeks. In May, a likely financially motivated actor systematically probed METR's public infrastructure using agents for vulnerability discovery, credential stuffing, OAuth token grants and staff phishing, with no confirmed access to non-public data.
- March attackers harvested API keys from vibe-coded, publicly exposed agent dashboards
- ~$600,000 in credits consumed undetected due to normal high token spend
- May campaign used agents to automate credential stuffing and OAuth token abuse
- Exposed read-only SQL mechanism in transcript viewer risked unpublished eval data
- No sensitive information believed accessed in either incident
- ATT&CK
- T1552.001 Credentials In FilesT1110.003 Password SprayingT1098 Account ManipulationT1190 Exploit Public-Facing ApplicationT1566.002 Spearphishing Link
- Vendors
- Amazon Web ServicesGoogle
- Products
- EC2
- Victims
- METR
- Organizations
- METR
From the original article
3 min read at thehackernews.com
Ravie Lakshmanan Sep 01, 2026 Cyber Attack / Artificial Intelligence METR (short for Model Evaluation and Threat Research and pronounced "Meter"), a research non-profit that evaluates frontier artificial intelligence (AI) models for their ability to carry out long-horizon, agentic tasks, disclosed that it suffered "two notable security incidents" where external actors attempted to gain unauthorized access to…
By @TheHackersNews · thehackernews.com
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.