Stolen AI API keys are paying for automated attack scans
Bitsight found an autonomous attack tool for sale on a criminal forum that runs on stolen OpenAI, Anthropic and Google keys, so the key owner pays for the scans.
Source: Stolen AI API keys are paying for hackers' autonomous scans (helpnetsecurity.com). Text prepared with AI from this source.
What happened and what to do
According to Help Net Security, Bitsight tracked 16 forums from February to August 2026 and saw mentions of AI agents, indirect prompt injection and MCP rise 67%. The tool, called MONST3R-AI, takes the stolen keys, a spending cap and a target list. The researchers also cite about 7,600 fake GitHub repositories, more than 800 posing as AI skills or MCP servers.
The recommendation is to search for exposed AI provider keys and treat them as seriously as a cloud credential, verify the source of every MCP server and agent plugin and stop local agents from reaching sensitive folders or running commands without approval. Spending limits and alerts on unusual usage help catch abuse early.
How the consultancy can help
Wendelmaques can search code and pipelines for exposed AI keys, set up periodic rotation with spending limits and alerts and review which MCP servers and plugins the team uses. The scope is defined according to the environment.
Next step
If your company uses AI APIs, send a short description of how the keys are stored and receive a scoped proposal.
Let's assess your project
I reply with scope and a proposal.