Unverified50% confidenceCautionExact time
Computer use capabilities generally face the risk of prompt injection attacks in deployment, where malicious instructions hidden in web pages or documents may induce the model to perform unintended operations.
1
Sources
50%
Confidence
Long-term
Relevance
10/26/2026
First Seen
Sources
Related Entities
Related Claims
VerifiedThe core security risks of computer use capability include prompt injection attacks and unauthorized access to sensitive local data.78% similarVerifiedIndirect prompt injection attacks plant hidden instructions within normal business data to induce Agents into executing unauthorized operations, representing a security risk at the semantic level rather than the code level.75% similarVerifiedLLMs are vulnerable to prompt injection attacks, where malicious web pages or files can embed hidden instructions to induce AI to perform unintended actions.75% similarVerifiedOne of the security risks of MCP is prompt injection attacks, where malicious data sources may embed instructions in returned content to induce the model to perform unauthorized operations.73% similarUnverifiedBrowser security mechanisms are shifting from passive blocking to proactive warning, with emerging approaches attempting to assess risk before users visit.73% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1100825API
curl https://kongchang.com/api/v1/knowledge/claims/1100825MCP
get_claim(id=1100825)