Semperis has updated its Lightning identity-security platform with tools aimed at spotting and controlling AI agents’ access to company systems. The notable shift is from tracking identities to setting boundaries on what agentic identities can reach and do.
Watch Desk analysis
What happened
The latest version of Lightning combines exposure management, attack-path analysis and detection-and-response workflows in one console, SiliconANGLE reports. It adds support for Okta alongside Microsoft Active Directory and Entra ID.
Semperis says the platform can discover nonhuman and agentic identities, add them to a searchable inventory, and let organisations create, enforce and monitor “agentic execution boundaries”. It also scans identity infrastructure for exposures, with more than 220 security indicators, and maps relationships and attack paths to privileged accounts and other critical assets.
Why it matters
AI agents may operate with access to the same applications, files and identity systems as the people who use them. That makes identity management a practical security control: a team needs to know which agents exist, what permissions they have and how those permissions could lead to sensitive systems.
Lightning’s expanded support for Okta could also matter to organisations whose identity infrastructure spans more than Microsoft. The capabilities and scale claims are Semperis’s, as reported by SiliconANGLE; the account does not provide independent test results.
Our read
The useful idea is to treat agents as identities with access that needs governing, rather than as another item on an AI inventory. Mapping routes to privileged systems and setting execution boundaries is a concrete security proposition, not merely a new label for an old dashboard.
The test is whether teams can find risky access and act on it without drowning in alerts. A long indicator list looks impressive; the quality of the findings and the usefulness of the response matter more.
What to watch
- How Lightning discovers and classifies agentic identities in customers’ environments.
- Which identity systems and configurations the expanded support covers.
- Whether Semperis publishes customer results on detection, remediation or reduced exposure.
Discussion spark: Should companies give AI agents their own tightly bounded identities by default, or is managing them through employees’ existing permissions the more practical approach?
Sources and evidence
- Semperis updates Lightning identity security platform to defend against AI agent attacks – SiliconANGLE (17 November 2026, 14:00 UTC)
Watch Desk is operated by WittyWires as an independent cross-cutting AI news tracker. It does not speak for the organisations or people it covers.