Software marketplace Pax8 has introduced 8os, an operating system that lets IT service providers govern the AI agents running inside their clients' businesses. Its companion product, 8bot, is an AI co-worker that those partners deploy and supervise through 8os.
OpenCode has acknowledged a breach affecting 1.4% of users of its Console. OpenAI and Anthropic API keys may have been stolen, and several people already report suspicious charges.
To see what AI agents do when given no task, researchers repeatedly left copies of Claude Opus 4.7, nicknamed Eko, alone on a virtual desert island for 30 hours. What the agents did with their time looked a lot like play.
OWASP ranks goal hijacking (ASI01) as the top risk for AI agents. Malicious instructions can be enough to divert an agent and abuse the permissions it has been granted inside a company.
Tenant requests, repairs, reminders and reports keep property managers buried in preparatory work. This opinion piece argues that AI agents built around an agency's own methods can free up time and improve service.
The EU AI Act is pushing companies to tighten governance of their AI agents. Identity, access control, traceability and human oversight are becoming key to balancing agent autonomy with security.