Where did the Alpha Go?
From Wikipedia, the free encyclopedia - Alpha is a measure of the active return on an investment, the performance of that investment compared with a suitable market index
The following was not written or embellished with any form of AI - straight from my head to your eyes.
I recently had a phenomenal conversation on where there’s alpha left in starting a new company. Previously I mentioned that in a world where anyone can build anything with AI, the only thing that matters is the ability to deploy and distribute product effectively. This still holds true. That said, I recently had a conversation that made me look at things from a new perspective and I had to share.
While I’m rooted in my answer for the ‘what matters when we can solve seemingly any problem with AI’ question, someone I spoke with had a fresh perspective that answered it differently. His belief, was that product actually matters most today. At first I disagreed (deployment and distribution, duh). But, I really wanted to understand his why. After going back and forth on it, it finally clicked for me.
The unsolved and unknown problems are where the biggest alpha remains today. Solve what can’t be solved and build what can’t be built (easily, that is).
While it’s true that there’s a ton of opportunity in the AI Services and Application layer, a lot of it continues to erode with companies of all sizes entering the market (Palantir, Distyl AI, Percepta, AI for Vets, AI for Doctors, AI for Billing, AI for Restaurants, AI for X, Y, Z and onwards). I’m sure Peter Thiel wouldn’t want to touch any of these viciously competitive markets with a 10ft pole anymore (see “competition is bad” from Peter Thiel’s Zero to One). The companies building in spaces where problems are misunderstood, if understood at all, will stand to generate the greatest alpha in the medium term future. (Yes this is also a bull case for SpaceX).
Now, what are examples of unsolved problems? These would be problems that AI can’t solve today because it’s unable to interact with the physical world, in a cost-effective way. For example, take the company Anello. Anello, has created a technology which allows for inertial navigation in GPS-denied environments. As it exists today…AI Models cannot solve this problem end-to-end, just yet.
On the other hand, what are examples of unknown problems? These would be problems that people don’t know they have, which may be solved by technologies that people don’t even know exist right now. For example, take the company Ixana. Ixana has Wi-R Chips which safely allow data transfer using the electro-magnetic field around humans. Think of touching your phone and a speaker to play music without wires or Bluetooth. Think opening your apartment door by touching the handle, with your phone sending a signal to your lock with your finger!
In the past, great impact and wealth was made by solving problems with novel technology that people couldn’t even imagine. Similar to the technological transitions from film photography to digital or even from flip phone to smartphone, those that can solve the unknown will stand to make the greatest alpha versus other companies.
For early career graduates, young adults, and investors alike that lack the credentials, experience, and credibility to advance rapidly in the highly competitive AI Application and Services world, it’s worth your while to embark on the pursuit of solving unknown and unsolved problems with strong commercial viability. That’s where the biggest alpha will lie going forward.
(Hint - DeepTech + Reindustrialization Problems!!!)
P.S. Thanks Rishi.