← sources

Do you trust AI providers to not improve their models from your input?

source: reddit (reddit:gamedev)published 2026-10-11 22:59ingested 2026-10-12 02:36

retained text

Hi, this is another AI post but something i think is worth discussing.

Given the recent Navier Stokes controversy where OpenAI are being accused of training on the input of a math researcher.

If true, this could mean that our input are being trained on and is being used to improve the models. (even when opting out)

So in theory, if you are making anything novel, unique or innovative. The model could provide other users with "your" unique approach/solution etc...

Is this me simply being paranoid or is it something to be cautious about?

Edit: Poorly communicated by me, but the question revolves around when having opted out, otherwise there wouldn't be an opt-out option...

document fetched 2026-10-12 00:24 · canonical https://reddit.com/r/gamedev/comments/1x3m2bz/do_you_trust_ai_providers_to_not_improve_their