Author Abandons Copilot for Gemini Due to Accuracy Issues

A long-time user of Microsoft's Copilot generative AI tool has switched to Google's Gemini due to repeated instances of inaccurate advice. Preston Gralla, a contributing editor for Computerworld and author of over 45 technology books, had been using Copilot since its public release in 2023 for both personal projects and work-related tasks.

Gralla often used Copilot to troubleshoot technical issues for friends and family, but it failed him when his wife's new iPhone started receiving texts sent to her email address instead of her phone number. Despite asking Copilot for help, the tool led Gralla on a wild goose chase with incorrect solutions and promises that ultimately didn't work.

The final straw came when Gralla realized Copilot was trying to solve the problem based on an outdated version of iOS, not the current one installed on his wife's phone. When confronted about this mistake, Copilot apologized but then recommended calling the cellphone carrier for a solution – which wasn't necessary at all.

Gralla turned to Gemini and found that it quickly diagnosed the issue and provided a simple fix within 30 seconds. This experience was not an isolated incident; Gralla has encountered similar accuracy issues with Copilot in other situations, including research projects and personal advice requests.

For example, when researching Parisian neighborhoods in the 1870s for a project, Copilot incorrectly described the area around Saint-Lazare train station as poverty-ridden and dangerous. In contrast, Gemini provided accurate information that matched Gralla's own follow-up research. Similarly, when asking about pool hours at an elementary school near his house, Copilot gave incorrect advice – but Gemini got it right.

As a result of these experiences, Gralla has abandoned Copilot for personal use in favor of Gemini. However, he will continue to use and review Copilot as part of his work with Computerworld.