AI companies have become data-hungry businesses as they require increasingly large datasets to train their models. To meet that need, many AI startups are actively collecting data, ignoring long-standing internet conventions such as respecting robots.txt files, which signal to automated crawlers which parts of a website are off-limits. This requires websites to restrict access to data and, in some cases, enter into licensing agreements with AI companies. Fitness and social running company Strava is moving in this direction by restricting its website and introducing fees for developer access.
To prevent scraping, the company has strengthened security around its website and will now only allow authorized users to view certain data. Previously, users could view details such as public profiles and health club listings without logging in. The company puts all of its data behind authentication to protect it from fraudulent AI scraping.
On the API side, developers were previously able to start building apps on Strava through a free tiered access program. They started by requesting basic access and then requested additional access as their app grew. The company is currently adding a flat monthly fee of $11.99 to all developers, but says the price may vary by region.
Strava says its developer community has grown from 185,000 members last year to 241,000 members this year, and the company plans to continue supporting them. As part of this, Strava will also be adding support for Model Context Protocol (MCP). This is a new standard that allows AI assistants and apps to access external data in a structured way, giving you control over exactly what and how Strava shares it.
The company also plans to deprecate some API endpoints (separate access points that allow external apps to retrieve certain data such as club details) to protect user data. Strava had already tightened its API rules in 2024, banning its use for AI training and restricting third-party apps from viewing other users’ data. These changes sparked a backlash from developers, who said their apps would be severely affected.
Some developers may accept subscription fees, but deprecating certain API endpoints may impact dependent apps. Strava is giving developers a 90-day grace period before making these changes.
Strava CEO Michael Martin said in an interview with TechCrunch that unchecked AI scraping could spell the end of the public internet.
“AI companies are relentlessly scraping public websites because they need endless amounts of training data, slowing down the overall performance of the site,” Martin said. Over the past few months, we have experienced multiple instances of poor performance and, in some cases, failures. Not only are they scraping public sites, they are ignoring API terms and trying to access data using our APIs. ”
He noted that Strava has rejected offers from major AI labs for data licensing agreements. He singled out Perplexity in particular, saying the AI search startup scraped through an aggregator service to hide its origins despite the rejection. This is consistent with Perplexity being accused of similar conduct in the past.
Martin also warned of server overload caused by poorly built and vibe-coded apps, where API calls are often configured inefficiently, creating a disproportionate load on the Strava system. This is a pattern. Last year, Meta made a similar argument about system overhead when it banned third-party chatbots from WhatsApp.
The timing is probably no coincidence. Strava secretly filed for an IPO earlier this year, and the move to protect data may be aimed at demonstrating data discipline to prospective investors. Comparisons to Reddit’s 2024 crackdown on API access were one of the issues Martin was quick to bring up. Unlike Reddit, which priced API access by the number of calls made (out of reach for many app developers), Strava is betting that flat fees won’t hurt its developer ecosystem.
“We want our users to feel like they own their data and have peace of mind in how we manage and protect it. But we also want our developers to continue to thrive and grow,” Martin said.
If you buy through links in our articles, we may earn a small commission. This does not affect editorial independence.
Source link
