Strava Tightens API Access to Protect User Data

Strava, the well‑known fitness‑tracking platform and social network that lets athletes log runs and share workouts, has decided to put its public data behind a paywall and a login screen. The move aims to curb the relentless scraping by AI firms that have been hammering its servers.

In November 2024, Strava had already banned developers from using any data obtained through its API to train AI models. Yet the companies simply found ways around the restriction, prompting the platform to take further action.

According to a recent announcement on its developer hub, AI companies have been aggressively harvesting user data to fuel no‑code AI tools that generate apps which, in turn, overload Strava’s APIs. This activity has driven a staggering 448% increase in developer applications year‑to‑date, severely degrading performance for everyday users.

Starting today, June 1, Strava will enforce stricter terms by introducing Standard and Extended developer tiers. Developers in the Standard tier can self‑upgrade from their API settings dashboard to access up to ten athletes without a formal review, while those in the Extended tier will receive higher rate limits.

The company also clarified that apps routing data through third‑party intermediaries are now restricted, as these networks obscure how athlete data is ultimately used. Moreover, all new Standard‑tier developers must purchase an active subscription to gain API access. By the end of the month, existing Standard‑tier developers will also need an active subscription to retain access, though they will receive a code granting a three‑month free transition period.

Strava, the popular fitness tracking app, has announced significant changes to its API (Application Programming Interface) on September 1. Key endpoints such as Club Activities, Club Administrators, Club Members, and Segments Explore will be deprecated. Access to the latter will only be granted to Extended tier applications intended for data analysis. To cater to subscribers who wish to analyze their workouts independently, Strava is introducing an official Model Context Protocol (MCP) tool.

The announcement comes after Strava CEO Michael Martin accused Perplexity of routing scraping activities through third-party aggregators to illegally gather athletic data despite being denied licensing proposals by Strava. Previously, Reddit had also filed a lawsuit against Perplexity for employing underhanded tactics to bypass blocklists and directly scrape Reddit forum posts from Google Search results pages.