Heard at .conf26: Trust in AI at Scale Is Earned One Step at a Time
Splunk, Monday, October 5th, 2026
Splunk .conf26 speakers described testing AI agents extensively and expanding their permissions only as they prove reliable.
Splunk's recap of .conf26 conversations finds organizations testing AI agents extensively before production and widening their permissions gradually as they prove reliable.
Clean data proved critical, with one company seeing a 90% false positive rate after rolling out automation on unprepared data.
Companies are adding safeguards such as a second AI agent reviewing decisions, limits on how many actions an agent can take and human approval for critical actions.