{"id":47,"date":"2021-04-03T16:24:33","date_gmt":"2021-04-03T16:24:33","guid":{"rendered":"https:\/\/aitoolkit.test\/guideline\/update-and-adapt-cautiously\/"},"modified":"2023-05-24T20:06:52","modified_gmt":"2023-05-25T03:06:52","slug":"update-and-adapt-cautiously","status":"publish","type":"guideline","link":"https:\/\/www.microsoft.com\/en-us\/haxtoolkit\/guideline\/update-and-adapt-cautiously\/","title":{"rendered":"Update and adapt cautiously"},"content":{"rendered":"
\n\t\n\n

Guideline 14: Update and adapt cautiously<\/h2>\n\n\n\n
\"green<\/figure>\n\n<\/div>\n\n\n\n

Limit disruptive changes when updating and adapting the AI system\u2019s behaviors.<\/h2>\n\n\n\n

When updating the AI system, consider the scale and rate of changes. Conduct user research to understand how changes may disrupt or impede the user.  <\/p>\n\n\n\n

When doing model updates, avoid introducing new errors for tasks where the AI system previously performed well. Use tools such as Backward Compatibility<\/a> to check for such errors. <\/p>\n\n\n\n

Use Guideline 14 design patterns to update the AI system cautiously, considering the scale and rate of changes:<\/p>\n\n\n\n