AI: Startup Vs Incumbent Value - We Need a New Science of Progress

AI: Startup Vs Incumbent Value - We Need a New Science of Progress

The value generated by AI has mostly been captured by incumbents rather than startups, despite the significant startup activity in the field. This is in contrast to previous waves of technological advancements, such as the internet and mobile, where startups managed to capture a significant portion of the value created. The question arises: why have incumbents been so successful in the AI space?

One possible explanation is that incumbents have an advantage in terms of distribution, capital, and pre-existing product moats. To beat an incumbent, a startup typically needs to build a product that is dramatically better or focus on a new customer segment or distribution moat that the incumbent cannot serve. In other words, a startup needs a 10X better product to overcome the advantages of incumbents.

However, there is reason to believe that startups will take a bigger share of the value generated by AI this time around. Unlike previous waves of AI startups, there are now infrastructure-centric companies with broad adoption and rapidly growing usage. These companies, such as OpenAI, Stability.AI, Hugging Face, and Weights and Biases, provide the necessary infrastructure for startups to build AI-powered products.

Another factor that may contribute to startups' success in the AI space is the speed of innovation. The current wave of AI technology seems to be much stronger, making it easier to create products that are 10X better than what incumbents offer. Furthermore, there are highly repetitive, highly paid tasks that can be automated using AI, and workflow tools for these tasks are either weak or nonexistent. This creates an opportunity for startups to leverage AI to provide solutions in these areas.

However, it is important for startups to avoid the "hammer-looking-for-a-nail" problem. Instead of focusing solely on the capabilities of AI technology, startups should identify actual end user needs and unserved product markets that can benefit from AI. By addressing these needs, startups can create products that provide real value to users.

On a broader scale, there is a need for a new science of progress. Progress, defined as the combination of economic, technological, scientific, cultural, and organizational advancement, has transformed our lives and raised standards of living over the past couple of centuries. Yet, there is no broad-based intellectual movement focused on understanding the dynamics of progress or speeding it up.

A new focus on progress would involve studying various aspects, such as identifying and training brilliant young people, facilitating the exchange and sharing of ideas among small groups, understanding the drivers of productivity differences among organizations, and determining effective ways to select and fund scientists. This would require a decentralized shift in priorities among academics, philanthropists, and funding agencies.

The success of progress studies would depend on its ability to identify effective interventions to increase progress and ensure their adoption by various institutions. It is not enough to merely understand progress; the goal is to treat and improve it. By investing in research and initiatives that promote progress, humanity can get better at knowing how to get better.

In conclusion, while incumbents have historically captured most of the value generated by AI, there are reasons to believe that startups will have a bigger share in this wave of technological advancements. The presence of infrastructure-centric companies and the speed of innovation in AI provide opportunities for startups to create products that are significantly better than what incumbents offer. However, startups must focus on addressing actual end user needs and unserved markets to create products that provide real value. Moreover, there is a need for a new science of progress to understand and accelerate the dynamics of progress in various fields. By investing in progress studies, society can work towards a better future.

Actionable advice:

• 1. Focus on building a product that is 10X better than what incumbents offer to overcome their advantages.

• 2. Identify actual end user needs and unserved product markets that can benefit from AI technology.

• 3. Invest in research and initiatives that promote progress to improve society's ability to innovate and advance.