Google Halts AI Photo Search Feature in Quiet Rollback

Google has temporarily halted the expansion of its "Ask Photos" AI feature within Google Photos. The tool, which had been gradually rolling out since last autumn, is being paused for improvements. "Ask Photos isn't meeting our standards yet," stated Jamie Aspinall, a Google Photos product manager, in a response to feedback on X. He cited three key areas needing work: response speed, answer quality, and the overall user experience.
This experimental feature is driven by Google's most advanced Gemini AI models. Google clarifies it uses a specialized variant of Gemini built exclusively for the "Ask Photos" function.
Aspinall explained that the rollout has been paused at a very limited scale while the team addresses these concerns. He anticipates a significantly improved version will be released in roughly two weeks, restoring the speed and accuracy of the original photo search.
In a related Tuesday announcement, Google also revealed enhancements to keyword search in Photos. Users can now employ quotation marks to find precise text matches in filenames, camera models, captions, or text within images. Searching without quotes will continue to include visual matches as well.
Originally unveiled at the I/O 2024 event last May, "Ask Photos" was promoted as a tool for asking your photo library practical, everyday questions—the kind you might typically ask another person. Examples include inquiring about past birthday party themes for a child or which national parks you've visited.
Related
- Google CEO Sundar Pichai discusses the upcoming AI platform transition
- Google's AI Super Bowl ad contains an error about cheese
"Gemini's ability to understand different types of content helps it grasp exactly what's happening in each photo, and it can even read text within images when needed," Google explained in the initial announcement. "Ask Photos then formulates a useful response and selects the most relevant photos and videos to share."
This is not Google's first instance of pausing an AI feature rollout. The company is engaged in a rapidly accelerating AI competition with other major tech firms and startups.
Last May, shortly after introducing "AI Overviews" in Google Search, Google suspended the feature. This followed widespread sharing of bizarre and incorrect answers on social media, with users unable to disable the tool. Notable errors included incorrectly labeling Barack Obama as the first Muslim U.S. president and suggesting users add glue to pizza to keep the cheese from sliding off.
Furthermore, in February of last year, Google launched Gemini's image generation tool with significant publicity, only to pause it within the same month. This decision came after users identified historical inaccuracies in generated images, such as depictions of the U.S. Founding Fathers as people of color.
Related article
Notion transforms its workspace into a hub for AI agents
Notion, the productivity software company, is entering the agentic era.During a live-streamed product announcement on Wednesday, Notion—best known for its collaborative note-taking app—unveiled a new developer platform that extends the capabilities o
ElevenLabs names BlackRock, Jamie Foxx, Eva Longoria as new investors
ElevenLabs, the voice AI company, has disclosed additional investors in its $500 million Series D round, originally announced in February. These include institutional investors like BlackRock, Wellington, D.E. Shaw, and Schroders; corporations such a
Google IO 2026 unveils voice interaction with Gmail inbox
Google continues to integrate AI into your inbox. At the IO 2026 developer conference on Tuesday, the company expanded its Gmail "AI Inbox" feature with conversational AI, allowing users to ask questions about their inbox content rather than relying
Related Special Topic Recommendations
Comments (1)
0/500

Google has temporarily halted the expansion of its "Ask Photos" AI feature within Google Photos. The tool, which had been gradually rolling out since last autumn, is being paused for improvements. "Ask Photos isn't meeting our standards yet," stated Jamie Aspinall, a Google Photos product manager, in a response to feedback on X. He cited three key areas needing work: response speed, answer quality, and the overall user experience.
This experimental feature is driven by Google's most advanced Gemini AI models. Google clarifies it uses a specialized variant of Gemini built exclusively for the "Ask Photos" function.
Aspinall explained that the rollout has been paused at a very limited scale while the team addresses these concerns. He anticipates a significantly improved version will be released in roughly two weeks, restoring the speed and accuracy of the original photo search.
In a related Tuesday announcement, Google also revealed enhancements to keyword search in Photos. Users can now employ quotation marks to find precise text matches in filenames, camera models, captions, or text within images. Searching without quotes will continue to include visual matches as well.
Originally unveiled at the I/O 2024 event last May, "Ask Photos" was promoted as a tool for asking your photo library practical, everyday questions—the kind you might typically ask another person. Examples include inquiring about past birthday party themes for a child or which national parks you've visited.
Related
- Google CEO Sundar Pichai discusses the upcoming AI platform transition
- Google's AI Super Bowl ad contains an error about cheese
"Gemini's ability to understand different types of content helps it grasp exactly what's happening in each photo, and it can even read text within images when needed," Google explained in the initial announcement. "Ask Photos then formulates a useful response and selects the most relevant photos and videos to share."
This is not Google's first instance of pausing an AI feature rollout. The company is engaged in a rapidly accelerating AI competition with other major tech firms and startups.
Last May, shortly after introducing "AI Overviews" in Google Search, Google suspended the feature. This followed widespread sharing of bizarre and incorrect answers on social media, with users unable to disable the tool. Notable errors included incorrectly labeling Barack Obama as the first Muslim U.S. president and suggesting users add glue to pizza to keep the cheese from sliding off.
Furthermore, in February of last year, Google launched Gemini's image generation tool with significant publicity, only to pause it within the same month. This decision came after users identified historical inaccuracies in generated images, such as depictions of the U.S. Founding Fathers as people of color.
Notion transforms its workspace into a hub for AI agents
Notion, the productivity software company, is entering the agentic era.During a live-streamed product announcement on Wednesday, Notion—best known for its collaborative note-taking app—unveiled a new developer platform that extends the capabilities o
ElevenLabs names BlackRock, Jamie Foxx, Eva Longoria as new investors
ElevenLabs, the voice AI company, has disclosed additional investors in its $500 million Series D round, originally announced in February. These include institutional investors like BlackRock, Wellington, D.E. Shaw, and Schroders; corporations such a
Google IO 2026 unveils voice interaction with Gmail inbox
Google continues to integrate AI into your inbox. At the IO 2026 developer conference on Tuesday, the company expanded its Gmail "AI Inbox" feature with conversational AI, allowing users to ask questions about their inbox content rather than relying





Home






