TL;DR<\/p>\n
A Google exec asked Gemini users on X what they were surprised the app still couldn\u2019t do well.
\nWorkspace reliability, better tool calling, and improved chat organization topped the list of feedback.
\nGoogle says many fixes are in progress, but celebrity likeness guardrails are staying as they are.<\/p>\n
The AI race is moving so quickly right now that it can feel hard to keep up. OpenAI is taking the headlines today with the rollout of GPT-5.6<\/a> and ChatGPT Work<\/a>, but a Google exec is taking a more feedback-focused approach this week, turning to social media to ask the audience a blunt question: What is Gemini still bad at?<\/p>\n Josh Woodward, whose X profile<\/a> lists him as a VP at Google, Google Labs, Gemini App, and Google AI Studio, asked users yesterday<\/a> what they were surprised Gemini still couldn\u2019t do well and \u201cshould have fixed a long time ago.\u201d Asking the internet anything can be a bit of a Pandora\u2019s box, and the internet duly obliged. After more than 1,700 replies, Woodward has now shared the top 10\u00a0<\/a>requests from\u00a0<\/a>the feedback<\/a>, along with Google\u2019s stance\u00a0on each.<\/p>\n Does Gemini need Siri-style voice customization?<\/p>\n 157 votes<\/p>\n Absolutely! Why didn’t Google do this first?<\/p>\n 42%<\/p>\n Not really. Gemini already offers a LOT of voice options.<\/p>\n 58%<\/p>\n The clear number-one complaint was that Google Workspace integrations need to work more reliably. That\u2019s kind of important, given that Google keeps pitching Gemini as something that can work across apps like Gmail, Docs, Slides, Keep, and Tasks, especially with its newer Gemini Spark<\/a> agent. Woodward said Google agrees on this point, adding that some improvements recently went into Gemini Spark, but that reliability needs to be \u201cmuch better across the entire app.\u201d It sounds encouraging, but then again, it\u2019s also the most diplomatic answer.<\/p>\n Second on the list was more reliable tool calling, which Woodward also said Google \u201cstrongly\u201d agrees with. He didn\u2019t give a precise timeline, but said users should expect \u201cnoticeable gains soon.\u201d Third was better project and folder organization for chats. Google apparently sees Notebooks as a step in the right direction, but Woodward said they\u2019re \u201cnot enough\u201d and that the company will rethink the experience.<\/p>\n Deep Research also came in for some attention, with users asking to export reports to NotebookLM and switch between Deep Research mode and Flash or Pro models in the same chat. Woodward said he \u201cwasn\u2019t aware of these pain points,\u201d which is a candid answer and probably the purpose of the exercise. He called both suggestions good ideas and said Google would \u201cfigure something out.\u201d Nano Banana watermark removal is also discussed, though Woodward said that Google has to navigate AI rules that vary by country.<\/p>\n Several other fixes appear to be more straightforwardly on Google\u2019s to-do list. Google is apparently working to expand MCPs and Custom Skills after rolling out early support in Gemini Spark, while message editing is in progress, and voice dictation accuracy is being prioritized \u2014 Woodwood plans to dig into specific user replies on that front. As for mobile app scrolling bugs, Woodward confirmed they were being fixed.<\/p>\n The one request that isn\u2019t getting movement for now is celebrity likeness guardrails. Woodward said Google plans to keep those as they are, so deepfake creators may prefer to stick to other models for now.<\/p>\n