Addressing Gemini's Custom Instruction Challenges in Google Workspace
Recent reports from the Google support forum highlight growing concerns among Google Workspace users regarding a significant decline in Gemini's performance. This decline is particularly evident concerning its ability to adhere to custom instructions, often referred to as GEMS. The issue is causing considerable frustration and impacting daily productivity for many organizations relying on Google Workspace for their operations.
The discussion thread, initiated by gemini_platform, paints a clear picture of this escalating problem. What were once reliable GEMS, functioning 90% of the time, now frequently fail, demanding constant re-instruction. This is not merely a minor inconvenience; it represents a fundamental breakdown in the expected functionality of an AI tool specifically designed to enhance efficiency.
