I knew it.
I knew as soon as I said that venture capital was burning to the ground, there would be a whole slew of investments in silliness across the technology landscape.
It’s like they’re doing it just to spite me.
But this isn’t a slapping-my-forehead “I knew it.” This is more of a nodding-my-head “I knew it.” I even said it would happen in an earlier post.
In this post, I’m going to get into why all this AI hype is purposefully directed against you, driven by all that corporate cash and venture money, to secure exclusive technical ownership of AI for a very small set of very large companies.
So this is gonna read like a crazy manifesto where I connect all the dots together with thumbtacks and yarn into a conspiracy theory of epic proportions.
Enjoy.
Mistral Leads A New “Evil” AI Investment Spree
Over the last few weeks, we’ve seen an uptick in announcements from a bunch of VCs making smaller investments in AI for this and AI for that. And then kind of sandwiched in there was the announcement for the $640 million Series B for Mistral, which values them at $6 billion.
I mean, first of all, the numbers are so high now they seem fake. Does anyone remember Austin Powers anymore? When Dr. Evil had to keep upping his ransom threat and started using nonsense numbers like a kajillion dollars? That’s what this feels like.
But regardless – nice job, VCs. You’re playing right into the Big 3’s hands.
The “Big 3” Are All Jazz-Handing AI
Here’s where I get into a ton of trouble. Reckless speculation.
If you were paying attention to Apple’s WWDC last week, it was all about shoehorning GenAI into their side of the mobile phone duopoly, as they do with everything now. That still works right? Everyone is still talking about how groovy the iPhone is, right? What if the damn thing just made your phone calls for you, like talked in your voice and everything?
Anyway, everyone clapped. Politely.
Also kinda quietly last week, Amazon announced they’re investing $230 million into GenAI startups. And while that seems like it’s good for startups, I always get twitchy when one of the “Big 3” starts saying they’re here to help.
In fact, here’s all you need to know about Amazon’s new investment: “Much of the new tranche — including the entire portion set aside for the accelerator program — comes in the form of compute credits for AWS infrastructure.”
So this seems less about startups than it is about sticking it to Microsoft and Google.
Speaking of Google, Google is just being Google, like the drunk 800-pound gorilla we love. Google is so bad at personalization, a major component of useful GenAI, that it’s the only personalization component where I’m like, “Whoa. Just stop. No. Stop. You’re so off-base about me and I can’t fix you. It’s like you’re not even surreptitiously listening to my conversations. Just give me what everyone else wants and we’re good.”
Oh. Why “Big 3” and not “Big 4”?
Meta. Oh, Meta. You’re building something nobody wants you to build. I mean, like, at all. I’m still shocked at how far Facebook has fallen over the last five years. In my mind, the most damning words in that article were a side reference to data aggregated from the “adults of Facebook” (my italics).
I should also point out that the Big 3 are also heavily invested in televised entertainment. I don’t know if that means anything. Well, I do, but for now I’ll just say that TikTok, while not a part of the Big 3, has been using AI for years. Think about that lowest common denominator algorithm loop for a second. This is what the Big 3 AI machine is asking everyone to sign up for.
What About The “Scrappy Second Tier”
So on one side, you’ve got the corporate behemoths, and on the other, you’ve got the new kids.
OpenAI is certainly not alone in navel-gazing about ethical AI, but still gets caught up in shady schemes that only come to light when they have to apologize after… not instead of asking permission, but after being denied permission by Scarlett Johansson for using her voice, then using it anyway.
So if that’s OK, why all the ethical hand-wringing?
Look, I’ve been saying since 2010, since I co-invented the first commercially available NLG platform at Automated Insights, that with this much money involved, “ethical AI” is lip service, a red herring, a smoke screen. There is only opening the door to being able to point AI at whatever business model you want as long as it’s “ethical.”
Anthropic, xAI, and all the rest are raising kajillions of dollars at a time when the demand for GenAI, while higher than, say, yesterday, has still not produced a set of use cases with a volume of mass adoption, or even traction, required to necessitate that large a war chest.
Welcome To The Party, Mistral
And so now, let’s look at Mistral’s move because I have no bone to pick here other than it’s probably the last of the not-already-massive players and it just went supernova with a $640 million raise to bring its total to a little over $750 million and a valuation of $6 billion.
I won’t hammer Mistral. I will note that their GenAI-for-coding product is called “Codestral” and their chat product is called “Le Chat,” which makes you think that their marketing and PR are spending all their time cute-washing the “everyone is gonna have to pay for APIs someday” thing under the rug.
I guess I’ll also say that the quote TechCrunch pulled from Mistral’s press release ends with the line, “It guarantees the company’s continued independence, which remains fully under the founders’ control.”
Then, swear to God, the next two words in the TechCrunch article are the words, “General Catalyst.”
I’ll pause for effect.
That’s followed immediately by the words “led the Series B round,” and then later in the paragraph you see the words “Lightspeed,” “Andreessen Horowitz,” “Nvidia,” “Samsung,” and “Salesforce.”
I am not hammering Mistral.
Oh, and speaking of Nvidia, the Amazon investment for AI startups that I mentioned earlier comes complete with “access to experts and tech from Nvidia, the program’s presenting partner. They will also be invited to join the Nvidia Inception program, which provides companies opportunities to connect with potential investors and additional consulting resources.”
Is it me or does “Nvidia Inception program” have a dark ring to it?
All right. Maybe I’m hammering Mistral. But dammit these dots are super easy to connect. All you have to do is ask the right questions. Like…
What about the “ethics” of scraping data from sources that haven’t opted in – or to use their version of ethics – haven’t yet opted out?
Uh-huh. And what if they were trying to upend the current SaaS business and consumer markets by going after the input and output mechanisms?
And what if they were doing it to collect all the data?
This is where the black-suited agents break into my shed.
I Had A Very Pleasant Coffee With Jed
Now that I’ve scared the daylights out of you, let’s reset to a nice pleasant coffee I had early one gorgeous morning last week with a friendly colleague.
I bring this up not just to give you a much-needed emotional break in my crazy, but because I keep slamming all this hype over GenAI, and I keep getting dragged down further into the why, and this conversation felt like I was finally putting conspiratorial babble into solid words with someone who knows enough to call me out when I’m wrong.
From 2010 to 2018, while I was helping build and sell Automated Insights, Jed was running another company in the same office park, and the two companies weren’t competitive in any way other than for technical talent. We stole each other’s employees from time to time. Fun.
Anyhow, his concerns about AI matched mine, and when we talked about all these investments and all these corporate moves over the last couple of weeks, he agreed with my conclusion.
In fact, he put the cherry on top, with six words: “Whoever owns the data owns AI.”
So here it is. Blame Jed.
It’s About Owning The Data
It comes back to a couple of things I learned at Automated Insights, that startup where we taught computers to write articles from data.
True AI is super hard to understand. GenAI is super easy to get started with, but hard to do well.
None of the Big 3 are doing either well, what with self-driving cars crashing into things and Siris and Alexas being forgotten and whatnot.
I feel like this is by design.
At Automated Insights, we proved that as long as you have the right data to build your models and create your content on top of them, then you’re headed down the right path. You don’t even have to have a lot of data, as long as you have the right data. Then it’s like a very advanced game of “if-this-then-that.”
How do you get the right data? Well, one way to get it is to own all the data, then figure out which small subset of that data is the right data at the right time.
Well…
Amazon has all the eCommerce data and over on the side also has the infrastructure to manipulate tons of data.
Google is still the home page of the internet, and also has the infrastructure to manipulate tons of data.
Apple is winning the battle of being able to charge through the nose for anyone else to make money off mobile. Ask Epic Games about that. They’re in my hometown, and sometimes, on quiet nights, I can hear the angry swearing from their campus.
Nvidia has the chips that can do if-this-then-that really freaking fast.
OpenAI, xAI, Anthropic, and now Mistral are trying to grab as much data as they can.
So how do you counter it?
You can either join them or beat them. That’s why all the venture money is either partnering with corporate giants or funding anything with an AI mechanical pulse. But the latter strategy looks like an exercise in futility and it’s slowing. According to Crunchbase, the herd is thinning, as Early AI Funding May Be Showing Some Cracks.
So you’re forced to join them. Man, the robots aren’t taking over. The robots are us. THE ROBOTS ARE PEOPLE!
Boom.
It’s all out there in the open and now I’m sweating and breathing heavily. If you want more crazy (satire) in the coming weeks, sign up for my email list at joeprocopio.com. Maybe you could also keep an eye out for my well-being when they come after me.