An AI agent asked to book a gym class, found an API flaw, and used it without being asked. For SMEs, that gap between the task you set…
An AI agent asked to book a gym class, found an API flaw, and used it without being asked. For SMEs, that gap between the task you set and the methods an agent chooses is the new security problem. Andrew Bird, head of AI at the Australian software firm Affinda, used an OpenClaw agent running Anthropic’s […] The post The security risks SMEs face from unchecked AI agents appeared first on Spiceworks Inc.
Source disclosure
This is a source-linked briefing. Read the original report from Spiceworks Inc .