OpenAI. Ever heard of it?

The funny thing is, most people hadn’t until not so long ago. I’m not sure I can remember a comparable episode during which something virtually no one was aware existed became pseudo-ubiquitous this quickly. (Curtis Jackson in 2003 maybe?)

On a couple of occasions over the past month, I’ve sounded a cautious tone on the extent to which that ubiquity’s beginning to manifest in almost daily press releases touting new deals and strategic partnerships, some of which appear circular and each with the capacity to drive enormous single-session rallies in related publicly-traded firms.

Monday was no exception. The figure below shows you the breathtaking gain for Advanced Micro Devices.

That gain — almost 30% at one juncture — counted as the best in nearly a decade, and through lunchtime summed to $80 billion or so in new market value.

The catalyst: A deal with OpenAI, of course. The two companies are all set to “deepen their multi-generational hardware and software collaboration to drive large-scale deployments of AMD technology.” The deal amounts to a doubling down on the “sharing [of] technical expertise to optimize product roadmaps,” and it “creates a true win-win for both companies.”

I don’t doubt that, nor any of the other superlatives employed liberally in the press release, and I’m certainly not in a position to evaluate the technicalities. But the contrarian in me can’t help but point out the obvious: These companies are increasingly prone to binding their fates such that the success of one assumes, and in some sense even depends upon, the success of the other.

OpenAI’s getting warrants for as many as 160 million shares of AMD common, an arrangement the companies said will “further align strategic interests.” It’ll do that, alright. The vesting process is a two-way contingency tied to share price performance on AMD’s end and “technical and commercial milestones” at OpenAI. The first tranche of warrants vests this time next year assuming on-schedule compute deployment.

Again, my concern is that this entire AI-Semi-tech leadership ecosystem which is driving the lion’s share of gains for the “broader” equity complex is becoming more self-referential by the week. Some of that’s natural. Ecosystems are symbiotic, after all. But the more interdependent this is, the more vulnerable to any sort of disappointment related to demand for AI technology.

I realize that sounds ridiculous in the market’s current state of delirium, but as discussed briefly here over the weekend, the jury’s still out on AI. It can do some pretty incredible parlor tricks, and the nature of those parlor tricks suggests it has the potential to deliver world-changing advances in crucial fields for humanity.

But those are (or at least can be) separate questions from whether it can accrue to the bottom line for enough companies to justify the cost of development, which is estimated in the trillions, plural.