On October 16, 2026, OpenAI announced that ChatGPT Slides is rolling out to eligible Pro, Business and Enterprise users with Space access on the web and desktop. OpenAI says users can create and refine a presentation from a conversation, files or an outline; edit its text, images, shapes and charts; collaborate in real time; leave comments; present; and export to PowerPoint, PDF or Google Slides. For QA teams, that makes a deck a potential delivery surface for test evidence—not just a summary artifact.

What OpenAI announced

  • ChatGPT Slides can start from a conversation, uploaded files or an outline.
  • OpenAI says decks can be edited directly or refined through ChatGPT, including text, images, shapes and charts.
  • Space collaborators can edit in real time and leave shared comments.
  • On the web, decks can be exported to PowerPoint, PDF or Google Slides; the desktop app can open a deck in PowerPoint or Keynote when installed.

OpenAI describes this as a rollout, so access depends on an eligible plan, Space access, client and workspace settings. Treat account and rollout variation as test context rather than assuming every user will see the feature at the same time.

Why this matters for QA engineers

Test reports often turn dense evidence into a decision-ready story. When an AI-assisted slide workflow is used for that step, the primary risk is not whether a deck looks polished—it is whether the chart, claim, link and export still faithfully represent the underlying run. Collaboration adds another quality boundary: reviewers need the right evidence, changes need attribution, and sensitive test data must stay within the intended workspace.

ChatGPT Slides QA collaboration checklist

  1. Reconcile claims with source evidence. For every pass rate, defect count or trend, compare the slide with the CI run, test-management record or approved query that supplied it.
  2. Use controlled fixtures for charts. Seed a small known dataset with boundaries, missing values and a deliberate failure. Verify labels, totals, ordering and narrative text against the fixture.
  3. Review collaboration permissions. Confirm that only the intended Space members can view, comment or edit, and that a reviewer can identify the latest approved evidence.
  4. Test export fidelity. Open the same deck after export to PowerPoint, PDF and Google Slides where those paths are used. Check fonts, charts, links, speaker notes and slide order.
  5. Keep an audit trail. Store the source prompt, input files, deck revision, reviewer decision and export timestamp with the release evidence.

A compact release gate

Given a slide deck built from approved test evidence
When the deck is reviewed and exported
Then each numeric claim reconciles to its source
And each reviewer sees the intended revision
And the required export preserves charts and links
And no restricted test data appears outside its approved workspace

This is editorial QA guidance, not a statement about OpenAI’s internal implementation. It gives teams a practical boundary between a useful AI-generated presentation and an auditable release artifact.

Bottom line

ChatGPT Slides QA collaboration should put evidence fidelity ahead of visual polish. Verify the source, collaboration scope and exported result separately, and an AI-assisted test-reporting deck can remain both fast and trustworthy.

Source