If you were to ask twenty of our peers and founders, 'What does it mean to be an ‘AI’ company today?' I think you would receive twenty different answers. And yet, all of them are likely correct.
You could argue that it's generative AI, specifically foundational models like OpenAI, Dall-E, Copilot, or even MoLFormer, that are operating at the bleeding edge. These models ingest vast amounts of data, have billions of parameters, and are fine-tuned to produce specific outcomes. In OpenAI’s GPT-4 (an LLM), the model is trained to take a user-written prompt and generate an AI-written response. Others maintain that even companies that just incorporate off-the-shelf machine learning or pattern recognition could be classified as AI companies.
Setting AI semantics aside, the short answer is that all of these companies utilize artificial intelligence. Yes, there's a difference between companies enabled by AI and those building AI (i.e., the next foundational model). However, the overwhelming majority of companies fall into the former category, not the latter.
This brings us to our point. The paradigm has shifted and investing in AI as a category-specific goal for a portfolio, in our opinion, is beleaguered. The reality is AI is now table stakes for any company. If you’re not evaluating how AI applications, either off the shelf or built internally, can fundamentally change the growth trajectory of your business, you’ve either already missed the train, or are hoping to catch it at the next stop. As Goldman Sachs recently wrote, “AI’s huge benefits will eventually trickle down to [all] companies that leverage it to boost productivity, improving their revenues and potentially enhancing their margins.”
Right now, AI is a constant buzzword bingo in pitches we take and it can be easy to get caught up in the fervor. But taking time to step back and understand the whole picture is incredibly important. The AI matters, but not as much as the people, business, and opportunity of the company. Does the founder(s) have an outlier opinion on the market? Does that outlier approach make sense? Is the GTM strategy sound? Do we have conviction in the thesis? Is the company creating and capturing value?
Everyone is trying to tell you they’re an AI company, and why that's the reason they will outperform. Color us skeptical. Artificial intelligence will accomplish many things for startups, it will democratize founder-ship, increase the pace of innovation, and drive down the cost to founding and building companies. It becomes a catalyst for net new company creation.
We believe the best approach is to invest in fundamentally sound startups, creating step functions of value that are enabled by innovations in AI.
What do you think? We would love to hear from you!
