Server-Side Change
A phenomenon in which a model provider updates model weights, inference configuration, or safety policies in the background while the user interface remains unchanged, leaving users unable to perceive changes in the model version.
Timeline (last 90 days)
The research team identified three server-side changes that left almost no trace on the user interface yet significantly altered the model's judgments.
Server-side changes refer to a model provider updating model weights, inference configurations, safety policies, or system prompts in the background, while the user interface and API version numbers may remain completely unchanged.
All Facts (2)
Server-side changes refer to a model provider updating model weights, inference configurations, safety policies, or system prompts in the background, while the user interface and API version numbers may remain completely unchanged.
50%UnverifiedThe research team identified three server-side changes that left almost no trace on the user interface yet significantly altered the model's judgments.
50%