Understanding Recent Challenges with Project Management and User Feedback in AI Tools

As long-time enthusiasts and power users of AI-driven platforms, many of us rely heavily on features that facilitate complex workflows, extensive worldbuilding, and detailed documentation. Recently, however, significant regressions in core functionalities have begun to impact productivity and workflow stability—particularly concerning are issues with project referencing and user feedback submission, especially on mobile devices.

Core Functionality Disruption: Project Referencing and Attached Files

Historically, project management features within these AI systems served as versatile, scoped knowledge bases. Users could attach reference materials such as canonical documents, PDFs, or other files, ensuring consistency and rapid access to authoritative sources during long-term projects. This setup supported writers, researchers, and collaborative teams by maintaining context and reducing manual cross-referencing.

Alarmingly, recent updates—specifically around version GPT-5.2—have introduced notable issues:

  • Inconsistent File Referencing: The system no longer reliably scans or retrieves attached files within projects.
  • Contradictions in Canonical Data: Previously maintained canonical documents get overwritten or contradicted unless users manually re-insert references each session.
  • Neglect of Large or Key Documents: Larger, authoritative files tend to be ignored or overlooked, undermining the project’s integrity.

These issues reset the utility of Projects from dynamic, knowledge-scoped repositories to mere themed chat folders—diminishing their foundational purpose and complicating workflows that depend on persistent, reference-based knowledge.

Challenges in Mobile Feedback Submission

Compounding these issues is the difficulty users face when attempting to provide feedback via iOS apps. The process has become cumbersome, discouraging constructive input:

  • Limited Entry Points: Few intuitive options exist for submitting detailed feedback.
  • Inadequate Feedback Tools: The interface offers minimal support for elaboration or contextual reporting.
  • Flow Interruptions: Redirects and complex steps pull users out of their workflow, further discouraging feedback.

As a result, valuable insights from mobile users are increasingly lost, reducing the platform’s ability to adapt and improve based on user experience.

Why These Changes Matter

These aren’t trivial usability quirks—they represent a substantial erosion of the platform’s capabilities. Many users, myself included, have built complex workflows around features that are now inconsistent or unreliable. Over time, such regressions erode trust in the platform’s stability and future viability.

In response, I’ve taken measures such as exporting canon and archives to external systems like

Leave a Reply

Your email address will not be published. Required fields are marked *