Strava has rolled out an optional Google Gemini integration for U.S. subscribers, allowing users to connect their workout history to the AI assistant for conversational training insights.
Strava announced a new integration with Google Gemini this week, giving subscribers a conversational tool to examine their training logs and progress. The rollout builds directly on the company’s recent launch with Claude, expanding the ways athletes can query their logged activities.
Once connected, users can ask the chatbot specific performance questions. Sample prompts provided by the platform include Am I getting faster at the same heart rate?
Can you give me a detailed analysis of how I performed in my last race?
and How is my strength training impacting my running?
Setup Process and Privacy Permissions for U.S. Subscribers
The Gemini integration is rolling out to U.S. subscribers beginning now. Setting up the feature requires opening Personal Intelligence within Gemini, selecting Strava under connected apps, signing into a Strava account, and authorizing the connection.
The link is entirely optional and read-only. Users can disconnect the service at any time from Connected Apps in Gemini. However, the consent screen explicitly warns that Gemini will access workout activities, performance metrics, training plans, gear, clubs, and profile information. It also notes that parts of the conversation, along with the user’s name and email, may be shared with Strava and handled according to its privacy policy.
Pagination Limits and Marathon History Retrieval Errors
Testing the integration reveals immediate friction when querying long-term athletic histories. When asked for a detailed analysis of every marathon-distance effort since 2019, Gemini reported that the user had logged exactly one: the Marine Corps Marathon on October 25, 2026. In reality, the user had logged at least eight.
When questioned about the missing data, the AI explained that its access is restricted to recent history, roughly from summer 2026 onward, due to pagination limits on past activities. A workaround requires providing specific dates and locations manually, though data reach across years like the 2025 NYC Marathon and 2024 Philadelphia race remains uneven.
Weather Variables and General Training Advice
AI interpretations also require careful human oversight regarding environmental variables. When asked about efficiency gains, Gemini initially confirmed that the runner was getting faster at the same heart rate, leaving out the impact of temperature.
The assistant required prompting to point out that heat and temperature are variables. Similarly, when tested with broad queries like What am I doing right and what am I doing wrong as a runner?
the tool offered flattering, conventional guidance about adding speed workouts and polarizing training that most recreational runners could spot independently.
