Agentic AI vs generative AI in security work

Generative AI produces content a person reads; agentic AI uses a model to choose steps and call tools that change systems. Run on CVE-2026-21589, a KEV-listed Atlassian bug with 19 fixed builds, the same wrong version number is a bad draft in one case and a wrongly closed ticket in the other.

Where the two terms differ

NIST's AI 600-1 Generative AI Profile (July 2024) quotes Executive Order 14110: generative AI is "the class of AI models that emulate the structure and characteristics of input data in order to generate derived synthetic content". The order was revoked on 20 Jan 2025 (White House), so the wording is NIST's quotation, not an order in force. The profile lists confabulation, "confidently stated but erroneous or false content", a risk of any model output.

OpenAI's practical guide to building agents draws the boundary: applications that don't use the model to control workflow execution, "simple chatbots, single-turn LLMs, or sentiment classifiers", are not agents. A summarizing call is generative; the loop around a model that acts is the agent.

One advisory, handled both ways

Atlassian published the advisory for CVE-2026-21589 on 5 Oct 2026: an unauthenticated arbitrary file access flaw (files inside the web application root), CVSS 4.0 score 9.3, affecting all versions before the fixed builds in eight products. Atlassian Cloud is already patched. CISA added the CVE to its KEV catalog on 15 Oct 2026. The advisory's fixed versions, 19 in all:

Fixed versions from Atlassian's advisory for CVE-2026-21589 (5 Oct 2026), read 19 Nov 2026.
ProductFixed versionsBuilds
Bitbucket Data Center9.4.26, 10.2.8, 10.5.13
Confluence Data Center9.2.26, 10.2.192
Jira Service Management Data Center5.12.40, 10.3.26, 11.3.123
Jira Software Data Center9.12.40, 10.3.26, 11.3.123
Bamboo Data Center10.2.24, 12.1.122
Crowd Data Center6.3.7, 7.0.3, 7.1.7, 7.2.44
Crucible4.9.151
Fisheye4.9.151
Total19

Two fixed lines look alike: Bitbucket Data Center 10.2.8 and Confluence Data Center 10.2.19. Suppose an estate runs Bitbucket 10.2.9 (above its fixed 10.2.8, so fixed) and Confluence 10.2.12 (below its fixed 10.2.19, so affected). The right result is one of two instances affected.

Now suppose the model confuses the two products' fixed lines and treats 10.2.8 as the threshold for both. This is a constructed slip to show the mechanism, not a reported model failure.

Advisory facts from the Atlassian advisory for CVE-2026-21589 and CISA's KEV entry, read 19 Nov 2026. The estate and the slip are constructed.
StepGenerative use: one callAgentic use: model in a loop with tools
InputThe advisory text pasted by an analystThe ticket ID only
Model's actionsWrites a summary: "10.2 line fixed at 10.2.8"Calls get_advisory, list_instances (returns 2), compare_versions twice, update_ticket
Where the slip showsIn a paragraph the analyst reads beside the advisory table. The analyst checks the number before actingIn the ticket: Confluence 10.2.12 is marked resolved because 10.2.12 is greater than 10.2.8
CostOne corrected draftA KEV-listed, unauthenticated bug stays open on a Confluence server, with a ticket saying it is closed
How it is caughtBy the readerOnly by a check outside the agent's report: read the version back from the instance (see verify after each change)

The model, the prompt and the mistake are the same in both columns. What differs is that the second form acts on the mistake and has credentials to do it. That moves the questions from "is the summary right" to what the identity may do and who answers when the closure is questioned (human approval points).

What to test differently

A generative use is tested on its answers: run sample advisories and compare the output with the table. An agent is tested on the final state of the system, over repeated runs, because Anthropic's Building effective agents warns that "the autonomous nature of agents means higher costs, and the potential for compounding errors". In the example above, a test set for the agent needs an instance whose version sits between two products' fixed lines, and it passes only if the ticket status matches the instance's real version. AI agent evals covers the setup.

When this is the wrong lens

  • A generative step that feeds an automatic action is already agentic in effect. If a script reads the model's "fixed" or "not fixed" and closes the ticket without a person, the risk matches the agent column, whatever the architecture is called.
  • A fixed workflow with a model step inside it is neither case. Anthropic classes it as a workflow because "predefined code paths" decide what happens next; see agentic AI vs SOAR playbooks.
  • The question that decides the controls is whether anything acts on the output without a person reading it. Where nothing does, a generative tool with a reviewer needs less than an agent; where something does, the label does not lower the risk.

Sources

  1. NIST AI 600-1, Generative AI Profile, July 2024, read 19 Nov 2026
  2. The White House, Initial Rescissions of Harmful Executive Orders and Actions, 20 Jan 2025, read 19 Nov 2026
  3. OpenAI, A practical guide to building agents, What is an agent, read 19 Nov 2026
  4. Anthropic, Building effective agents, read 19 Nov 2026
  5. Atlassian security advisory for CVE-2026-21589, read 19 Nov 2026
  6. CISA KEV catalog entry for CVE-2026-21589, read 19 Nov 2026

Last reviewed 19 Nov 2026