As the CEO of a company where machine and learning and AI play an important role, I am always interested in seeing how other organizations choose to implement these technologies into their organization. It is fascinating to see how Uber vs Lyft has been battling for supremacy — through tellingly different approaches to AI.
How both Uber and Lyft have moved forward says a lot about their overall goals as a company. Thanks to their recent IPOs and the amount of information that has been publicized as a result.
On the face of it, Uber and Lyft look similar.
Uber and Lyft look like fairly similar companies: both have made their names in the ride-sharing business while generating billions of dollars in revenue. Take a closer look, however, and it becomes apparent that each company has its own very different approach to success. They each have their own method of applying AI in order to achieve it.
Uber, as the larger company of the two, has more resources at its disposal, and therefore has the ability to spend more time building out a comprehensive machine learning platform and platform.
Lyft, on the other hand, might be smaller but the company still has access to a tremendous amount of customer and driver data that it uses to optimize its own ride-sharing platform.
Despite their size differences, they have also publicly explained the differences in their corporate philosophies, and those differences have carried over into their approaches to AI.
Much of the information available about the two companies.
The valuable information come from their respective SEC filings, made prior to IPO. These filings are a treasure trove of information for anyone looking to learn more about how companies such as Uber and Lyft structure their businesses. Where do they think their greatest value lies?
As ZDNet writer Larry Dignan notes, Uber’s filings “[make] it clear that its data science and algorithms are the key to its marketplace technologies,” whereas Lyft’s IPO documents do not actually mention artificial intelligence directly.
The Lyft angle.
What Lyft takes care to highlight, however, is the data it has collected “from over one billion rides and over ten billion miles driven,” which has been used to develop machine learning algorithms and inform data science engines.
According to Lyft, the insights from all this information are used “to improve the product experience for riders. Riders are presented with personalized transportation options,” as well as to “anticipate market-specific demand” and “create customized incentives for drivers in local markets.”
Lyft focus is less on innovation and more on how to perfect the service that it offers.
Based on all of this information, it’s clear that Lyft sees machine learning as a means to improve their existing services. Uber – as I will explore shortly – has a different take.
What does Lyft consider to be the core elements central to the company’s success? Overwhelming focus is on its leadership, core values, and the relationships it has with employees, drivers, customers, and partners.
Uber, on the other hand, emphasizes its “deep technology advantage.”
Uber looks at the numerous proprietary systems it has built, and its stated goal for the coming years is to use this technology “to redefine the massive meal delivery and logistics industries.”
For Uber, technology is not merely the means for providing a better ride-share experience – it’s also the opportunity to expand into other industries and completely transform them.
- Uber leverages artificial intelligence and machine learning.
- Uber uses machine learning to predict demand for rides, as well as determine optimal routes based on the time of day, traffic and weather conditions, and any other factors that could affect movement.
- The company has also integrated AI into virtually every aspect of the business, from customer service to fraud detection to marketing spend to the onboarding process that drivers have to go through.
- Uber has its own AI research team, whose findings are regularly published and presented at conferences around the world, and has released open-source software for outsiders to use.
According to Uber’s SEC filings, the company uses machine learning in many different ways. The use of natural language processing has, in the company’s words, helped to “simplify and enhance interactions” on their platform. It allows the company to save money by forgoing human customer service agents for automated ones.
Uber uses computer vision to verify drivers’ licenses and other important documents (again reducing the need for human interaction), and what it terms “sensor processing algorithms.” These help to improve its location accuracy in crowded areas.
Looking beyond the rideshare business, its algorithms are used to estimate food preparation and arrival times, as well as to generate personalized recommendations based on a person’s ordering history.
Uber does a lot with machine learning.
This is not to say that Lyft is indifferent to the technology; rather, it speaks to the very different long-term strategies that each company has decided to employ. Uber, one could argue, has positioned itself as a tech company, whereas Lyft’s mission, to “improve people’s lives with the world’s best transportation,” is more focused on the personal impact that the company can make.
It’s for this reason that Lyft claims to use machine learning to make the experience for consumers and drivers better, whereas Uber touts the increased efficiency and automation that using machine learning can bring.
Which approach is better?
That depends on your definition of success. On the one hand, Uber is a much larger company than Lyft, with a current market cap of over $68 billion (compared to Lyft’s $15.5 billion) and operations in over 70 countries.
On the other hand, that doesn’t seem to have had a positive impact on its stock price, which has fallen significantly since the company went public. It also doesn’t help that Uber has consistently made headlines for all the wrong reasons, from endemic sexual harassment to a gaffe-prone CEO.
The question then becomes, is it more important for a company to have a “human touch,” or to focus on technology first and people second? The answer will ultimately be decided not by the financial markets, but by users themselves.
Jeremy Fain is the CEO and co-founder of Cogntiv. With over 20 years of interactive experience across agency, publisher, and ad tech management, Jeremy led North American Accounts for Rubicon Project before founding Cognitiv. At Rubicon Project, Jeremy was responsible for global market success of over 400 media companies and 500 demand partners through Real-Time-Bidding, new product development, and other revenue strategies, ensuring interactive buyers and sellers could take full advantage of automated transactions. Prior to Rubicon Project, Jeremy served as Director of Network Solutions for CBS Interactive. With oversight of a $30 million+ P&L, Jeremy was responsible for development, execution and management of data-driven solutions across CBS Interactive’s network of branded sites, including audience targeting, private exchange, and custom audience solutions. Prior to CBS, Jeremy served as Vice President of Industry Services for the IAB, where he shaped interactive industry policy, standards, and best practices, such as the first VAST standard and the Tc&Cs 3.0, by working on a daily basis with all the major media companies as well as all the agency holding companies.