Interview Of The Week

Interview Of The Week: Sangeet Paul Choudary

Sangeet Paul Choudary is the founder of Platformation Labs and the best-selling author of Platform Revolution and Platform Scale. He has advised the leadership of more than 40 of the Fortune 500 firms and has been selected as a Young Global Leader by the World Economic Forum.

Choudary’s work on platforms has been selected by Harvard Business Review on four occasions in the HBR Top 10 Must Reads compilations.He is a frequent keynote speaker at leading global forums including the G20 Summit, the World50 Summit, and the World Economic Forum. Choudary was also formerly a member of the Forum’s Global Future Council and the co-chair of the MIT Platform Strategy Summit.

He recently spoke to The Innovator about Generative AI and competitive advantage.

Q: Many large corporates are racing headlong into adopting Generative AI and many are throwing all caution to the wind. If everyone is adopting the technology, where is the competitive advantage?

SPC:. AI, like every other technology, is going to be commoditized.

In today’s hype cycle, CEOs are scrambling to effectively communicate how AI will change their competitive advantage to investors. If you believe, as I do, that AI is increasingly getting commoditized, the question they should be asking is what is our competitive ecosystem and how does AI impact our ability to play in that ecosystem?

I’ll give a simple example. Let’s say you’re a company like Intercom, in the business of creating software for call centers. Intercom has changed from a SaaS-based strategy to a cost-per-resolution strategy that leverages AI. So, Intercom is innovating, right? It has used AI to its advantage. But the problem is that Intercom is no longer playing in just the contact center software space. It’s also playing in the larger customer journey management space, where it may have to interact with sales software, marketing software and other types of software. These other software providers – Salesforce, ServiceNow, and others – are also looking to use AI to create a decision-making hub around which all other downstream decisions are being made. All these software companies play within the larger enterprise workflow space; even Amazon is getting into contact center management software with Amazon Connect. My point is that your competitive ecosystem is no longer what you thought it would be because AI reduces the boundaries between competitive ecosystems by virtue of its ability to harness intelligence. If you take over key decisions for the user, you gain the right to take over workflows that are upstream and downstream from that decision.

The second thing that’s important is to really see if there are multiple competitive ecosystems in which you play and then determine which ones are primed for an AI advantage:  what’s the level of critical knowledge in this space that AI can organize well? Today, we are all enamored by large language models. What we don’t realize is that large language models are not intelligent. They are based more around affinity between words, and they derive context from that affinity, whereas, if you are playing in a space in a competitive ecosystem where context is already encoded, say in a knowledge graph, or in some other format, then you don’t need to rely just on the generic extraction of a large language model to get to context. So if AI is getting commoditized, what will become valuable is context. If something gets commoditized, its complement becomes more valuable. So, you need to first think about the playing fields, and secondly, think about which of these playing fields has sufficient encoded context for you to operate well using AI.

Then you need to evaluate three things: Will AI give me a relationship advantage, a decision advantage or a workflow advantage? The relationship advantage is the ability to compete on the basis of a strong relationship or channel:  a proprietary and defensible way to engage and serve your customer or key stakeholder. If you already have the decision advantage, you can strengthen it further. If somebody else has the decision advantage, then you need to ask yourself how can AI help you enable a more important decision for the user? If someone else has a workflow advantage, you can get yourself in that workflow by virtue of empowering the most important decision that the user is making within it.

Q: Can you give a concrete example?

SPC: Sure, in a value chain of manufacturers and resellers, resellers have a relationship advantage, in the sense that they have access to the customers, and they offer highly customized services. For that bundle of advantages, they claim a significant margin, if the relationships matter in that value chain. Now if in that value chain if manufacturers provide a sales AI assistant, which encapsulates all the knowledge about creating configurations of the product, they can, with that, potentially increase the reseller base, so that anybody who doesn’t understand the product very well can now also start selling it with the help of the AI assistant. The more they use the assistant, the more the manufacturer gains information about customer preferences. Eventually it can create sufficient customer knowledge to create direct channels to the customer and the customer can start solving the problems themselves, because the assistant has bridged the product knowledge and the customer knowledge gap. That is how you can change the relationship advantagQ: What are some of the biggest mistakes that companies are making right now when embracing Gen AI?

SPC: There are two very widely different ways in which companies are embracing Gen AI: through cost leadership or differentiation. Some, like Klarna, are deciding that 70% of their customer care workforce will no longer be needed because AI can solve the problems just as well.  Others are using AI to improve their margins. And then there are the companies which are trying to bundle an AI with their product, and through that trying to create differentiation.

On the cost leadership side, companies are not necessarily thinking about the trade-offs. For example, one big American company wants to create an AI architect that would allow you to say ‘I want to design a stadium with A, B, C, D and E attributes and I want it to be designed in the style of Gaudi, and with that prompt, the whole design gets created. With things of that sort, you are commoditizing creativity, and you’re losing some of the aesthetics that can come when a human is involved in the creation.

On the differentiation side, the most common mistake that companies are making is that they are slapping AI on top of anything that they’re doing. It’s a bit like saying Michael Jackson used to be popular. So here, this is his album. You know what? Taylor Swift is popular now, let’s put a few songs from Taylor Swift onto the Michael Jackson album. It does not make sense. Bolting on AI onto a product without considering the customer-desired attributes is quite similar. Every product is a bundle of value. It’s been created with certain customer attributes in mind. It’s been created to deliver towards certain kinds of customer requirements, and now you’re just taking AI and slapping it on top. You’re not really unbundling and asking, ‘Have the core attributes changed? Do we need more attributes? Does AI help us deliver relevant attributes in a better way? Does AI create trade-offs?’ So you need to think about it at that level and then rebundle it into a new product. And the problem right now is that many companies are taking an existing bundle, ignoring the design considerations that went into creating that bundle, and then slapping on AI and creating a new bundle, which may or may not make sense, because it doesn’t align with attributes that the customer cares about.

Q: What do you want The Innovator’s readers to take away from this interview?

SPC: We’ve reached a point where we have created markets which favor commoditization and I think that this is a vastly underappreciated aspect of how AI fits in the 2024 tech landscape. Take the case of Spotify. It constantly pushes songs in its playlist which are best from an economics perspective. It sources some songs very cheaply from independent sources and starts adding them to different types of playlists because the number of streams that those independent songs have reduces how much it needs to pay to the labels. My point is that Spotify, as a result, cares less about artist creativity or user interest and more about increasing its margin, which creates the incentives for increasing commoditization. If you keep in mind that we’ve created incentives for increasing commoditization in the economy, that is where AI coming in 2024 creates all sorts of questions, because AI is a tool for commoditization, and if you’ve already created systems and structures that favor commoditization, then you are not going to use AI to enhance creativity. I believe we are at a point where even if we solve for the right solution in one corner, systemically, we will be moving towards commoditization, and I think that’s the real issue. That’s the real challenge that we have.

To access more of The Innovator’s Interview Of The Week articles click here.


About the author

Jennifer L. Schenker

Jennifer L. Schenker, an award-winning journalist, has been covering the global tech industry from Europe since 1985, working full-time, at various points in her career for the Wall Street Journal Europe, Time Magazine, International Herald Tribune, Red Herring and BusinessWeek. She is currently the editor-in-chief of The Innovator, an English-language global publication about the digital transformation of business. Jennifer was voted one of the 50 most inspiring women in technology in Europe in 2015 and 2016 and was named by Forbes Magazine in 2018 as one of the 30 women leaders disrupting tech in France. She has been a World Economic Forum Tech Pioneers judge for 20 years. She lives in Paris and has dual U.S. and French citizenship.