I recently posted a detailed discussion about the planned retirement of Custom GPTs and the accessibility implications for neurodivergent users who rely on bounded, repeatable cognitive environments.
An OpenAI representative replied by restating that Custom GPTs are being retired and that workflows can be migrated to Plugins, then closed the discussion.
That did not address the central point of the post.
The issue I raised was specifically that Plugins and Projects do not reproduce the same context topology as Custom GPTs.
I provided a concrete example showing that a brand-new Project with project-only memory still receives broader account-level context about my identity, authorship, preferences, and prior behaviour. For my workflow, that means Projects cannot replace the “knows the project, does not know me” environment I currently use for bias-controlled writing and analysis.
I also raised concerns about accessibility for neurodivergent users who rely on compartmentalised cognitive environments, and about the possibility that migrated functionality could become restricted to Work mode, Codex, credits, or another separately metered surface.
The response did not address any of those distinctions.
I am not asking for private support on an account-specific problem. I am trying to continue a public product discussion about whether the proposed replacement preserves the actual capabilities of the feature being retired.
Is there a way to have the original topic reopened, or another appropriate public category where this discussion can continue without simply being closed after a migration FAQ is repeated?
I do not want to cross-post the original issue or violate forum guidelines. I want to continue the existing discussion because the core accessibility and workflow concerns remain unanswered.
