Wiz counts nine AI developer platform breaches in October as stolen keys fuel LLMjacking
Wiz says at least nine AI developer platforms disclosed incidents in October 2026, with stolen AI keys used for millions of model calls in LLMjacking attacks.
Summary
Wiz Research published a report on 30 October 2026 that counts at least nine AI developer platforms, including Fireworks AI, Lightning AI, OpenCode, Railway, Warp and Northflank, that disclosed security incidents during October. Attackers went after environment variables holding AI provider keys. In two cases Wiz traced, stolen Amazon Bedrock credentials made about 800,000 and 1.7 million model calls, worth more than $100,000 at list price.