AI: Startup Vs Incumbent Value - How the Landscape has Shifted

AI: Startup Vs Incumbent Value - How the Landscape has Shifted

The field of artificial intelligence (AI) has seen a significant shift in the distribution of value between startups and incumbents. Historically, in previous waves of technological advancements, startups were able to capture a substantial portion of the value. However, when it comes to AI, incumbents have emerged as the primary beneficiaries, leaving startups with a smaller share of the pie. This unexpected trend has raised questions about the factors that contribute to this disparity and what it takes for a startup to succeed in the AI space.

Looking back at the first wave of the internet, we can observe that the majority of the value generated went to startups such as Google, Amazon, PayPal, and Facebook. These companies were able to leverage the power of the internet to build new and innovative products that disrupted established industries. Incumbents like Microsoft, Apple, IBM, and Oracle also managed to extend their franchises onto the internet, capturing a portion of the value. This wave saw a relatively balanced split of value, with startups possibly claiming a 60:40 or 70:30 ratio.

The mobile revolution, on the other hand, witnessed a different distribution of value. Incumbents like Apple and Google dominated the mobile space, with every mobile version of their apps becoming the go-to choice for users. However, startups still managed to carve out a significant share of the market with platforms like WhatsApp, Uber, and Instagram. In this instance, the split between startups and incumbents may have been closer to 20:80.

Crypto, as a nascent industry, has seen startups completely dominate the value creation. Bitcoin, Ethereum, Coinbase, and Binance are just a few examples of startups that have captured the majority of the value in the crypto space. Existing financial services and infrastructure companies have had limited participation in this value creation. The absence of incumbents in the crypto industry raises questions about their ability to adapt to new technologies and disrupt their own business models.

To beat an incumbent as a startup in the AI space, it is crucial to either build a product that is significantly better than the incumbent's, overcoming their distribution, capital, and pre-existing product advantages, or focus on a new customer segment or distribution moat that the incumbent cannot serve. The key is to offer a product that is at least 10 times better than what the incumbent provides. This may explain why incumbents have historically prevailed, as their data advantage and existing infrastructure gave them a head start. However, this advantage may be diminishing as startups leverage the broader internet as an initial training set and adopt models that work well with smaller data sets.

The AI landscape is evolving, and startups now have a greater chance of capturing value. The current wave of AI technology, with advancements in transformers and unsupervised learning, presents unique opportunities for startups to create products that are exponentially better than what incumbents offer. This technology sea change could level the playing field and allow startups to gain a larger share of the value generated by AI.

One notable development in the AI ecosystem is the emergence of infrastructure-centric companies with broad adoption and growing usage. OpenAI, Stability.AI, Hugging Face, and Weights and Biases are examples of such companies that provide crucial infrastructure for AI development. This infrastructure accessibility opens up more opportunities for startups to leverage AI technology and build innovative products.

Moreover, there are specific use cases where AI can provide significant value. Highly repetitive tasks, such as coding, writing marketing copy, or generating website images, can benefit from AI-powered workflow tools. The ability to summarize or generate text and images in a high-fidelity manner is now made possible by new AI technologies. Startups that can identify these unserved product markets and cater to the needs of end-users have a higher chance of success.

In conclusion, the AI landscape has witnessed a shift in the distribution of value from startups to incumbents. However, with recent advancements in AI technology, startups now have an opportunity to capture a larger share of the value. By building products that are 10 times better than what incumbents offer and targeting unserved markets, startups can overcome the advantages of incumbents. To succeed in the AI space, startups must focus on actual end-user needs and leverage the exciting technology available. The future holds great potential for startups to finally unlock real value from AI.

Actionable Advice:

1. Focus on building a product that is at least 10 times better than what incumbents offer. This will help overcome the advantages of incumbents in terms of distribution, capital, and pre-existing product moats.
2. Identify unserved product markets and cater to the needs of end-users. By addressing actual end-user needs, startups can tap into unexplored opportunities and gain a competitive edge.
3. Leverage the infrastructure-centric companies that provide AI tools and technologies. Access to these resources can enhance the development and scalability of AI-powered products.

Exciting times lie ahead for startups in the AI space. With the right approach and a deep understanding of end-user needs, startups can finally unlock the true value of AI and establish themselves as major players in the industry.