Understand this story
At the Elevate event, Ryan Nigro, head of the Security Trust and Enablement team at monday.com, proposed that permission management is key to securing AI agents, and that well-designed guardrails can let teams move faster. He advocates gradually opening up AI capabilities according to the principle of least privilege, granting each newly deployed tool only the minimum access it requires, and expanding the scope after verifying the results. He also cited IBM research stating that 80% to 90% of data breaches stem from human error related to excessive authorization. This view is currently reported by SiliconANGLE, with no further developments yet.
AI-generated from public reports
Latest development · Report
monday.com security lead: Permission management lets AI agent teams move fasterDevelopments
1 confirmed developments
Ryan Nigro proposes that permission management is key to AI agent security
Source:SiliconANGLE
Reports and discussion timeline
Browse reports and discussion by time. Their evidence levels remain distinct.
Oct 30
SiliconANGLE
monday.com security lead: Permission management lets AI agent teams move faster
Ryan Nigro, head of the Security Trust and Enablement team at monday.com, said at the Elevate event that permission management is key to securing AI agents, and that well-designed guardrails can let teams move faster. He advocates gradually opening up AI capabilities according to the principle of least privilege, granting each tool only the minimum access it needs when it goes live, and expanding scope only after verifying the results. He cited IBM research saying that 80% to 90% of data breaches stem from human error related to excessive permissions.