Are ChatGPT Inputs the Next Slack Messages?

Last week on Theo Von’s podcast, Sam Altman said something that should be a wake-up call for the entire justice system.

“So if you go talk to ChatGPT about your most sensitive stuff, and then there’s … a lawsuit or whatever, … we could be required to produce that.” He called that reality “very screwed up” and said lawmakers need to move with urgency to protect people’s privacy.

We have seen this pattern before though. When Slack and Teams first entered the workplace, people treated those tools like casual, internal spaces. The messages felt temporary and informal. But then came the subpoenas, preservation orders, and courtroom exhibits. Offhand comments and quick reactions became part of the official record, privilege logs swelled, and the legal system had to catch up.

Now we may see the same pattern emerge with generative AI. People are using ChatGPT to work through real decisions. If an executive drafts fifteen versions of a termination letter or a CEO refines how to explain a restructuring, they are capturing the evolution of their thought process. And their inputs reflect what was considered, what was discarded, and how the decisions really took shape.

Yet many users still assume these prompts are private. They believe the tool is temporary and that nothing is being stored, reviewed, or preserved. But as Altman made clear, the record exists. And if asked for, it may have to be produced.

The implications are not theoretical. The discovery fights are coming. Are you ready?