If you are building or investing in AI-based products and services, this piece from Greylock is well worth a read to navigate the fairly nascent area of AI monetization.
As I read this, I had a few thoughts to add:
a) Considering that foundation models in AI are heavily data-dependent compared to previous tech stacks, I wonder if we might see increased segmentation within this market along domain or vertical lines. For example, while a technology like CRM could span across retail and manufacturing industries in the previous tech stack, the AI era with its substantial reliance on specific data might lead to stronger delineation in the tech stack.
b) The dynamics of big vs. small businesses and economies of scale are bound to have unique implications in AI due to the necessity for domain-specific data. The impact of these different data sources on performance, and the potential for varying points of diminishing returns, might introduce subtleties that we haven’t seen below in the market.
c) Lastly, I can’t help but wonder about the potential for new types of ‘moats’ that could emerge or be amplified in the AI era. Just as the advent of the internet significantly amplified network effects, could we see something similar or even entirely new with the advancement of AI?
These are my initial thoughts, and I’m keen to hear your thoughts as well. The world of AI is continually evolving, and the more perspectives we can gather, the better equipped we’ll be to drive its future.
Leave a Reply