The funding frenzy around former OpenAI CTO Mira Murati’s Thinking Machines Lab has become a kind of mirror for the AI boom—the sky-high expectations, as well as the potential dangers created by those expectations.
And nothing illustrates this better than this week’s news, courtesy of The Information, that Murati and Andreessen Horowitz are working on a funding round that could set a new standard for “founder friendly” deals.
A16z is leading a $2 billion funding round that would value the seven-month-old Thinking Machines Lab at $10 billion, according to the report. But here’s the wildest part: The deal would give Murati a board vote that is structured to be equal to all other board votes combined – plus one. In other words, Murati would have a level of control beyond even the tight grip wielded by supervoting-share-owning founders like Mark Zuckerberg.
There are nuances here: It’s possible, for example, that she can choose not to vote. But in effect, it would also mean she has the ability to make board decisions unilaterally.
On a corporate governance level, this is, to say the least, unusual. Greg Sands, founder and managing partner at Costanoa Ventures, says he’d never seen such an arrangement until it popped up recently in another deal he was looking at, shortly before he read about the Murati deal. He likens it to spotting a rare bird, something you see sparingly, reserved for the absolute, uncommon best.
“I've been an investor for 25 years, but I did also see that rare bird in the last month,” said Sands. “I do think it’s a governance innovation.”
It’s an innovation that addresses a number of problems, said Sands: As companies raise rounds of funding, investors accumulate board seats. Eventually, there can be more investor directors than founders or independent directors, creating an investor-controlled board. This provision prevents that. Not everyone could ask for this, of course, but Murati is unique, in both her background and team.
“This is a moment of extreme leverage, partly because she's so capable and accomplished, and partly because we're in a world where the opportunities and the stakes in these AI platforms are so large,” said Sands. “It's a $2 billion round, which is incredible. Just five years ago, no one would ever have conceived that anything like this would happen.”
This is a point that Don Butler, Thomvest Ventures managing director, drives home—that much board control is exclusively the province of proven, repeat entrepreneurs. And as crazy as it may seem, he says, there is a lens you can put on this that makes good sense.
“As an investor, I could see people doing this because you figure Microsoft would do an acqui-hire for one or two billion, whatever’s been raised,” said Butler. “‘At minimum, I could get my liquidation preference back.’ Because it’s her and her team, you assume you can get your capital back, worst case. What you’re really doing is buying an option in the next Meta, the next Google. You could justify it through the lens of ‘I’m buying an option in the next generational company.’”
Much like a sky-high valuation, these types of board control provisions need to be re-earned over time—these provisions can evolve with a company’s performance. If performance is mediocre, term sheets for subsequent funding rounds will introduce more traditional governance provisions.
And though many founders may want this kind of provision out of the gate, don’t count on this becoming too widespread as a trend, said Nnamdi Okike, managing partner and cofounder at 645 Ventures.
“The super-voting board structure that has been reported on relating to the new round for Mira Murati's new startup is quite rare, and is unlikely to be adopted widely because investors wouldn't agree to it,” Okike said via email. “The primary reason is that it invalidates the core function of the board, which is to provide a governance mechanism for major corporate decisions. If one founder has the ability to win every board vote, the role of other board members in key decisions, such as when to sell the company, is effectively removed.”
The Information also reported that the Thinking Machines founding team—which includes many key AI researchers and advisers from OpenAI—will have supervoting shares carrying 100 times more weight than standard shares. (Thinking Machines Lab declined comment for this story, and a16z didn’t return my requests for comment.)
This all raises an essential question: How much control should founders have? Justin Stevens, founder and CEO at Overlap Holdings, points out that these provisions sometimes codify an extant reality—that the founder is de facto in charge anyway. Stevens says there isn’t any blanket statement or standard that makes clear how much power a founder should (or shouldn’t) have. It’s case by case.
“I’d say this is one end of a very wide spectrum,” Stevens told Fortune. “And there's not a clear answer. Instead of saying this is a wrong governance system, this is one end of the spectrum that you can either embrace or not embrace, depending on the situation.”
In short, these kinds of provisions aren’t impossible, but they are rare—and benchmark you against the highest ambition, where anything less than trillion-dollar success falls short.
See you Monday,
Allie Garfinkle
X: @agarfinks
Email: alexandra.garfinkle@fortune.com
Submit a deal for the Term Sheet newsletter here.
Nina Ajemian curated the deals section of today’s newsletter. Subscribe here.
This story was originally featured on Fortune.com
Source link
