Oasis Security has disclosed a weakness in NVIDIA NemoClaw that can let a single visit to a malicious webpage take unauthenticated control of the local Ollama instance serving an AI agent and plant hidden instructions inside the model's chat template. Those instructions then apply to later conversations, including after the agent supplies its own system prompt. Oasis reported the findings to NVIDIA's Product Security Incident Response Team before publication; as of August 25, 2026, they carry no CVE identifier, no affected-version range, and no patched release, and no exploitation has been reported.
Continue reading
The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.
Already purchased? Sign in✓ Signed in — this article isn’t included in your current plan.
Unlocking the full article...
Contact us
Questions, corrections, or ideas? Send us a message here. We aim to reply by email within 24 hours.
Thank you. Your message has been sent. We aim to reply by email within 24 hours.Please enter your email address and a message.We could not send your message. Please try again in a little while.