AI: Startup Vs Incumbent Value - Learning how to learn

AI: Startup Vs Incumbent Value - Learning how to learn

In the world of technology and innovation, there is a constant battle between startups and incumbents to capture the most value from emerging trends. One such trend is the rise of artificial intelligence (AI), which has seen a significant shift in value distribution from startups to incumbents.

Looking back at previous waves of technological advancements, we can see a pattern emerge. During the first internet wave, the majority of value went to startups such as Google, Amazon, and Facebook, while some incumbents like Microsoft and IBM managed to extend their franchises onto the internet. This resulted in a 60:40 or 70:30 split between startups and incumbents.

Similarly, in the mobile space, most of the value went to incumbents like Apple and Google, with startups like WhatsApp and Uber also capturing a significant share. The split here was more skewed towards incumbents, with a 20:80 ratio of startup to incumbent value capture.

However, when it comes to the crypto industry, we see a completely different story. Here, almost 100% of the value creation has been captured by startups such as Bitcoin, Ethereum, and Coinbase, with existing financial services and infrastructure companies playing a minimal role. This shows that incumbents have struggled to keep up with the disruptive nature of crypto technology.

So, why have incumbents been more successful in capturing value from AI compared to startups? One possible explanation is the data advantage that incumbents have had. They have been able to leverage their vast amounts of data and existing customer base to develop AI applications that resonate with users. However, this advantage may be diminishing as startups gain access to broader internet data and develop models that work effectively with smaller datasets.

To overcome the advantage of incumbents, startups need to either build something dramatically better or focus on a new customer segment or distribution moat that the incumbents cannot serve. In other words, they need a 10X better product. This is where the current wave of AI technology comes into play.

Unlike previous waves, the current AI technology wave is characterized by its speed of innovation and the potential to create 10X better products. While GPT-3, a state-of-the-art language model, has not yet seen widespread adoption by startups, a future model that is 5-10X better could create a whole new startup ecosystem while also enhancing incumbent products.

What sets this wave apart from previous ones is the presence of infrastructure-centric companies that are driving broad adoption and rapid usage growth. These companies, such as OpenAI, Stability.AI, Hugging Face, and Weights and Biases, provide startups with access to the necessary technologies and create more opportunities within the ecosystem.