Inside the Business Models of Today's Top AI Platforms

Knowledge at Wharton

Ask about Knowledge at Wharton

Episode Description

ABOUT THE EPISODE Stefano Puntoni, Wharton Professor of Marketing, analyzes how OpenAI, Google, and Anthropic are pursuing distinct business models and growth strategies to compete in the rapidly evolving artificial intelligence market.

ABOUT THE PODCAST This Week in Business features interviews with Wharton faculty about the latest news, fascinating trends, and issues impacting both consumers and the business world. Episodes are recorded at the Wharton School and published twice per week.

#AIIndustry #AICompetition #GenerativeAI #AIMarket #TechStrategy ----- The Wharton School of the University of Pennsylvania is committed to sharing its intellectual capital through Knowledge at Wharton, the school’s online business analysis journal.

Launched in 1999, Knowledge at Wharton offers free access to:

- Articles, podcasts, and videos highlighting Wharton faculty research and analysis of current business trends - Interviews with book authors from Wharton School Press - In-depth series of curated content like Wharton Executive Education’s Nano Tools for Leaders - A searchable database of more than 10,000 articles covering all aspects of business - A weekly newsletter that delivers Knowledge at Wharton insights directly to your inbox

Visit the Knowledge at Wharton homepage: https://whr.tn/3qUpSBE Subscribe to the Knowledge at Wharton YouTube channel: https://whr.tn/3DEAgAU

Read the full transcriptHide the full transcript
  1. Well, we certainly know that there is a lot of interest around AI both from the public and from companies. But what's
  2. interesting is about the three main programs that are out there and the design as to how they are supposed to be
  3. best effective. Pleasure to be joined by Stefano Pontoni, professor of marketing
  4. here at the Wharton School who joins us today from California on a foggy morning
  5. out in California. Stefano, great to talk to you today. How are you, sir? I'm great. Thanks, Dan.
  6. So, I mentioned we know that the companies are wide in terms of using it and varied in terms of how they're
  7. using and building AI. But there is also different kind of strategies that are really in play here for these platforms.
  8. Yeah, there's a lot of talk about about not just what kind of algorithms companies are building and the
  9. architecture that they're using and the data that they're using and all of that.
  10. There's also interesting conversation to be had about the business strategy that
  11. different companies, different AI labs are using. So, sure. There's three main ones that we know
  12. about ChatGPT that is the main one seemingly right now. It seems like that to degree ChatGPT
  13. is a case of first mover advantage, correct? Yes, so we have indeed three main US-based AI labs. The first one is
  14. OpenAI, like you said, the ChatGPT is a product that you know, basically made the generative
  15. AI revolution kick off in a way in a big way, you can say. And then you have Google which
  16. is a company that invented the transformer model that is powering all of these generative AI tools, the
  17. language-based ones. So, they are very much you know, at the forefront too with the
  18. chatbot Gemini and other tools that they're embedding in various products. And also Anthropic with its
  19. chatbot Claude has also been very, very prominent and very successful. On top of
  20. that, you have players in China, you have you know, other other players. But I think those
  21. three are indeed the main ones. Now, ChatGPT indeed was the first mover and that enabled them to build an enormous
  22. customer base. We're talking about over a billion people of active users. So, the massive use user base. But the other
  23. ones aren't kidding either. Should we expect that as more users join and try it each of these platforms that
  24. the level of usage for all three will somewhat level out or does the advantage that ChatGPT have already
  25. one that continues on for a longer period of time? Yeah, looking for the three main looking
  26. at the strategy of the three main labs in the US, there's actually a working paper that
  27. just came out by a team of academics including Daniel McCarthy who's a Wharton PhD student now professor in
  28. Maryland. They're looking at mobile app downloads and trying to explore the niches and spaces that these different
  29. companies are occupying. And what they find is that so far you've had a company releasing a model every few weeks, you
  30. know, doing very well out of that and then everybody's now excited about, you know, what Google is doing, what
  31. Anthropic is doing, what OpenAI is doing. But that doesn't seem to have come at the cost of the other companies.
  32. So far, leaps in forward in terms of AI capabilities have led to an expansion of
  33. the pie rather than then you know, direct you know, taking customers away from from other competitors. But we'll have
  34. to see as the market continues to mature and you know, we reach a point where almost everybody has these
  35. accounts, then it becomes a little bit of a matter of what kind of habits people are forming, how they're getting
  36. used to using this technology. And so Gemini is perceived to be something that will
  37. be used within kind of an ecosystem, correct? Yes, so if you look at the three the
  38. three main labs, Google is the only one that you can say is fully vertically integrated in so far as
  39. they deal with the you know, the model building parts, but they also are embedding AI into
  40. into everything they do and also they own the cloud infrastructure that you need in order to power those tools.
  41. Well, that is not the case or at least you know, OpenAI is trying to build it, but obviously it's a huge effort. And um
  42. in case the case of Gemini it's basically language models in general will be tools that are used in
  43. every Google product. They're used in Gmail, they're used in Google Maps, they're used everywhere. Gemini is the
  44. main chatbot product and so far you've seen Google using it more as a as a ChatGPT-like tool, more of a
  45. productivity getting stuff done, also images and other things. But they haven't so far embedded so much
  46. into their advertising which is kind of interesting considering that Google's business has been based on advertising.
  47. In contrast what Google has been doing is to build AI overviews at the top of search pages. And so they are looking at
  48. monetizing chatbot through advertising via the search portal which will change with time. We're not
  49. entirely sure how. But in contrast taking ChatGPT is not the only product that OpenAI has basically,
  50. at least the two you know, consumer products. So, they are will need to monetize their product directly, not as
  51. part of an ecosystem the way that Google is doing, but more as a standalone product. And for them therefore, it's
  52. very important to think about what is the revenue per user that you can build for that chatbot. So, they have a free
  53. plan which is what you know, large majority of people are using and that doesn't make them any money. Like they
  54. they are losing money on that free plan. And then you have the paid plan with different tiers and they're making money
  55. on that, but there aren't enough of those who convert from the free to the paid plan and therefore, they announced
  56. that they're going to be doing advertising on ChatGPT. And so that's that's an I you know, an example of how
  57. the strategy pursued by the three firms is getting them to make different decisions. And so Claude is viewed as a
  58. premium niche strategy, meaning what specifically? So, Claude has a much smaller number of users,
  59. much smaller compared to ChatGPT but also compared to Gemini. However, the revenue per user is
  60. extremely high and Dan McCarthy team estimates that the revenue per user for Claude is about 40 times higher than the
  61. revenue per user from Gemini for example. So, massive difference. And the reason for that is that they have
  62. pursued a more professional market, basically coding through Claude code and poor work and tools like that. And that
  63. obviously is a market that is willing to pay because it's so easy to justify spending money on the chatbot if you
  64. find it that it really helps you create code with much faster speed or much higher quality. So, in that sense indeed
  65. it's a niche strategy. Niche not to mean very small, we're still talking about many millions of users and huge revenues
  66. and growing fast. But a little bit less focus on the consumer. Although you can see that a
  67. little bit changing. So, Anthropic was doing for example, you know, advertising during the Super Bowl. I think it
  68. indicates a more of a mainstream mass strategy. So, do the strategies that these
  69. platforms have right now and these companies have for them give us an idea of how each one of these
  70. could potentially develop in the years ahead? Yeah, it's very interesting because
  71. the strategies are very different and yet it's not clear to me that one is right and one is wrong. I think
  72. probably there's going to be different ways of being successful and it seems to me that the strategies that the
  73. companies have used is a good fit for their companies. So, OpenAI is going to go for this mass market and try to
  74. build revenues out of the free users using advertising while at the same time they keep building at the high end for
  75. paid subscribers and for enterprise. You've got Gemini is going to be basically part of a ecosystem play by
  76. Google to leverage AI to build a strong customer value across a set of tools and platforms and monetize them
  77. through advertising along the way. And then you're going to have Anthropic continuing this more niche strategy and
  78. trying to basically nail the coding and professional market. All of them are being successful. All of them will
  79. continue likely to be successful, but in their own different way. Because there's enough of a marketplace
  80. for all three to kind of continue to develop in their own landscape. The I mean, the the market didn't exist
  81. three and a half years ago. So, it's still still still pretty fresh and there's still room to grow in many
  82. different ways. You can grow the number of users, you can grow the revenue per users, you can grow how much users are
  83. doing with their platforms. So, we expect that to continue evolving. And by the way, the fact that one has a say
  84. ChatGPT or Gemini subscription doesn't mean that we not also want to a closed subscription. This is not just
  85. a for some users will be a need one choose a one and go with that. But for many people they'll have more. I mean
  86. they're using all of them. I think many will do that too. So if we have if we can have five streaming
  87. services we can have at least one AI service, right? Right. I do that with my students. I ask them you know who has
  88. a Netflix subscription and everybody raise a hand. I ask him who has you know a Spotify or everybody has those
  89. subscriptions and I think we'll end up having many of these chat subscriptions too. Stefano, great to talk to you.
  90. Enjoy California. We'll see you back here in Philadelphia. Thanks, Dan. You got it. Stefano
  91. Petroni, marketing professor here at the Wharton School.

Brief Summary

Stefano Pontoni discusses AI strategies of OpenAI, Google, and Anthropic, highlighting their unique approaches and market positions.

Key points

  • The Rise of AI Platforms. AI interest is booming, with major players like OpenAI, Google, and Anthropic leading the charge.
  • Different Strategies for Success. Each AI company is pursuing unique strategies to capture market share and revenue.

Tension

40/ 100Low tension

Related episodes