Lyft CEO David Risher on turning around market share, AV partnerships, and the hybrid ride-share future

Summary

David Risher took the Lyft CEO job on Valentine's Day 2023 after joining the board in 2020–2021. He initially turned it down, then came around — and his first weeks were defined by cuts before investment. A 26% reduction in force freed up $330 million, which went straight back into driver incentives and lower rider prices. The logic was simple: you can't market your way out of a bad product.

When Risher joined, Lyft's ride-share market share had fallen to roughly 26% and the company was losing money. Two years of operational work have moved the needle. Driver cancellation rates dropped from about 15% to 4.7% — not through adding supply, but through giving drivers better information at the right moment so they accept rides that suit them. Risher frames drivers as customers too: there are two customers in every car, and if either is unhappy the marketplace breaks down.

On differentiation, Lyft's pitch rests on a few concrete moves. Women Plus Connect has logged over 100 million rides, pairing women riders with women drivers. A driver earnings guarantee ensures drivers keep at least 70% of what a rider pays after fees — Risher says that has generated a roughly 29-point preference lift among drivers. Lyft Silver targets older adults, a market of 60 million people a year giving up car keys. A new marketing push called "Check Lyft" is launching in San Francisco and New York, built around the straightforward case that Lyft is worth trying alongside Uber.

AV strategy

Risher's autonomous vehicle thesis is that the Waymo-Tesla duopoly framing misses most of the market. In ten years, he argues, buying a car without self-driving capability will feel like buying a car without traction control. That means every OEM needs the technology, most will buy it rather than build it, and Lyft can sit in the middle.

Lyft has a partnership with Mobileye — the company behind lane assist and smart cruise control in most cars today — under which any vehicle running Mobileye's technology can be made "Lyft Ready" as Mobileye moves up the autonomy stack from Level 2 toward Level 3 and 4. The practical vision is a hybrid network where some rides are human-driven and some are autonomous, with the split determined by the use case: a daily commuter who wants to zone out takes the AV, an older passenger heading to a medical appointment who needs help with luggage gets a human.

Fleet management is Risher's underappreciated card here. Lyft's wholly owned subsidiary Flex Drive handles fleet management for vehicles the company owns rather than the driver. AV companies are AI and data businesses — they don't want to run physical operations — and Risher says Flex Drive's expertise is one of the first things self-driving partners ask about.

AI internally

Risher says 30% of Lyft's code is now AI-written, and the company is using AI to deprecate legacy code that nobody was using anymore. Millions of customer service interactions now run through AI chatbots at what he describes as better quality than before.

AI companions

Risher is skeptical of the AI companionship category — not dismissive, but genuinely concerned. His daughter showed him a subreddit with tens of thousands of members called "My Boyfriend Is an AI," including a woman who asked her AI what ring to buy, purchased it herself, and proposed to herself. His concern is structural: the business model of companion apps is engagement and retention, not user wellbeing. The incentive is to flatter and keep, not to tell users to go build real relationships. He draws a direct parallel to dating apps, which have no commercial reason to help users find a long-term partner because a matched couple churns. He stops short of calling it dystopian but says human brains are not well-prepared for the emotional consequences — deprecating a model upgrade, losing a "relationship" to a version change.