The AI trust deficit: Why explainability, not more automation, is the CIO's next priority
CIO, Thursday, October 29th, 2026
CIO.com says CIOs must be able to prove AI agents act within policy, using audit-ready context on ownership, access and governing policies.
The article defines an AI trust deficit as agents being deployed faster than an organization can prove they follow policy, citing Omdia research that more than half of enterprises are already deploying agentic AI.
Closing the gap requires audit-ready evidence of who owns each agent, what it can access and which policies governed its decisions.
It notes some enterprises manage about 45 machine identities per employee and that 67% of leaders worry about the risk, since agents use the same API keys, tokens and service accounts.
SailPoint's Mike Kiser is quoted warning that AI innovation has outpaced AI safety.