The premier artificial intelligence company isn't a public company at all, but rather private start-up OpenAI, the parent of ChatGPT. ChatGPT's debut exactly two years ago was an earth-shaking event, showing consumers and investors alike the astounding possibilities of generative AI.
The only way to access OpenAI's growth today is through cloud giant Microsoft (MSFT -1.73%), which has invested almost $14 billion in the start-up.
The catch was that until recently, OpenAI was a partially non-profit company, with Microsoft being excluded from some of the most advanced technological capabilities -- that is, until now.
Microsoft was initially limited in its OpenAI investment
The massive investment in OpenAI allowed Microsoft to obtain exclusive technology licenses, but the financial payoff was limited in its original deal.
For one, OpenAI was founded as a non-profit, then moved into a "capped profit" structure. The thinking by Altman and OpenAI's other founders was that they didn't want the technology to be beholden to capitalist forces, but rather dictated according to the overall benefits to humanity.
That of course complicates things for OpenAI's outside investors. When OpenAI attracted outside funding, the non-profit portion of the company exerted control and limited the input of outside investors, including Microsoft. That control aspect is what led to the temporary firing of OpenAI founder and CEO Sam Altman last year, although Altman was later reinstated.
Meanwhile, the capped return aspect of initial contracts limited any investor's return to 100 times their initial investment. That sounds like a lot, but for a disruptive company like OpenAI, the 100 times returns not is only possible but perhaps necessary for Microsoft's investment to make a difference in Microsoft's current $3.37 trillion market cap.
In the original setup, Microsoft was also limited in the technology it could license. For instance, it restricted Microsoft's ability to license OpenAI's advanced models once OpenAI achieved artificial general intelligence, or AGI. While the definition of AGI is imprecise, it would basically be when OpenAI's AI models become capable of performing valuable tasks better than humans without supervision.
OpenAI gives in
Because of OpenAI's need for massive computing investments, Altman and company appear to be altering its original mission. At a recent conference, Altman said, "When we started, we had no idea we were going to be a product company or that the capital we needed would turn out to be so huge. If we knew those things, we would have picked a different structure."
Back in September, OpenAI announced it was beginning a restructuring into a for-profit model, with the non-profit side receiving either a minority stake or non-voting shares. While the non-profit side would still retain a sizable stake in the company, the restructuring would also eliminate the investor profit cap, benefiting Microsoft.
Additionally, it was recently reported by Reuters on December 6 that OpenAI was in talks to remove the AGI clause in its contract with Microsoft. While no final decision has been made, it appears things are going in the direction of Microsoft gaining access to AGI capabilities once OpenAI gets there. Given Microsoft's stature as the dominant enterprise software platform, implementing AGI into its offerings could yield "must-have" products for businesses.
But how soon really will AGI be achieved? Maybe sooner than you think. In an interview for Y Combinator last month, Altman said AGI could come as soon as next year. While some may disagree on that timeline, if Altman sees next year, the likely timeline to AGI is probably much closer than the 10-year time-frame others have predicted.
So, if OpenAI hits its target and the AGI clause is removed, it could be a big benefit to Microsoft.
And there were other OpenAI announcements too
Since the calendar flipped to December, OpenAI made several new product announcements that should also benefit Microsoft as well.
On Dec. 5, OpenAI launched its new ChatGPT o1 model, as well as a new ChatGPT Pro tier for $200 per month, 10 times the cost of the $20 per month Plus tier. Then on December 9, Altman announced Sora, a feature within ChatGPT that allows users to generate videos based on text prompts.
These are all really positive announcements. The quick release of new models is good for OpenAI and Microsoft, as o1's improved multimodality (including images, videos, and unstructured data) will preserve and extend OpenAI's standing in the industry. Additionally, the day after Sora was released, OpenAI had to limit access due to overwhelming demand for the service.
The new $200 per month pro-tier also shows that there could be an audience willing to pay top-dollar for unlimited usage and the most advanced models. There have been lots of questions as to whether AI companies will generate a sufficient return on their massive spending for AI compute, and the new tier seems to show there very well could be.
As models become more advanced, that Pro tier could become table stakes for individual business owners and companies seeking a competitive advantage over rivals.
This all bodes well for Microsoft
As OpenAI improves and opens itself up for more Microsoft investments, it benefits Microsoft in two ways. First, licensing OpenAI's AGI models when that happens will improve Microsoft's products, and two, Microsoft's investment in OpenAI will gain in value, especially if the new $200 per month tier catches on.
Given that OpenAI was recently valued at over $150 billion in its latest funding round, it's not a stretch to think that as AI becomes more and more important to society, it could become a trillion-dollar company in the near future. If OpenAI reaches that valuation, Microsoft's stake could even begin making a difference to its already-giant valuation as well.
