How to destroy your company in less than 24 hours...
aka: The perils of inexperienced boards and why Microsoft may end up controlling OpenAI
By now, many of you have seen the news about the board of OpenAI firing CEO Sam Altman: https://www.geekwire.com/2023/sam-altman-leaving-openai-after-board-says-he-was-not-consistently-candid-in-his-communications/ and exiting him from the company immediately.
Needless to say, this was a shock to many (including Sam!), coming so soon after an incredibly successful series of product launches from OpenAI.
As of this writing, there are no definitive accounts of what happened behind closed doors. By the standards of normal CEO transitions, it's rare to see such a strong personal attack on the outgoing CEO. What does seem to be clear is that the board's actions were done hastily, and key partners and investors like Microsoft were informed only minutes before the announcement.
As one might expect, social media is currently abuzz with speculation. The most likely theories revolve around an internal coup led by one of the other OpenAI cofounders (Ilya Sutskever) (https://x.com/karaswisher/status/1725702501435941294?s=20). More imaginative and fanciful ideas suggest that Sam had discovered the OpenAI's AI had become self-aware, and the AI had to act...
I will not add to the speculation here--hopefully, a more complete story will emerge in time. Boards must hire and fire CEOs, and company leadership changes are part of the ordinary course of business. The interim CEO, Mira Murati, is exceptionally talented and may prove to be an even better CEO than Sam.
What matters is how boards handle leadership changes. OpenAI is now a masterclass example of how NOT to make the change. By acting precipitously and without an orderly and planned transition, the OpenAI board has overnight shifted the company from the leading AI company to the leading AI technology company with unstable and inexperienced board leadership. If the rumors are true about the change being driven primarily by internal politics, that is even worse for the board. Experienced board members would have seen right through such machinations and kept a steady hand on the tiller. Even if a CEO change was warranted, it could have been done more orderly and thoughtfully.
If I were still at Microsoft, I'd be furious. Microsoft has invested billions in OpenAI, and more importantly, OpenAI technologies are at the heart of a remarkably recharged and innovative Microsoft. Just this past week, Microsoft held their major 'Ignite' conference, showcasing over a hundred new advances in their adoption of AI.
It's too much to say that Microsoft bet the farm on OpenAI, but it is safe to say they've made a substantial strategic bet on the technology--and so far, without any publicly announced internal replacements. As I wrote about almost a year ago, this puts Microsoft strategically dependent on OpenAI: https://www.thoughtfulbits.me/p/microsoft-will-have-to-buy-openai
Now, Microsoft has a dilemma. They can continue the bet on OpenAI "as-is". After all, yesterday's announcement was just a staffing change, and nothing changed about the technology overnight.
However, in doing that, they risk betting on a sinking ship. What happens when Sam (and the other cofounder who left with Sam, Greg Brockman) starts a competing AI company? Already, several other high-profile OpenAI technical talent have resigned: https://x.com/hellokillian/status/1725797467315486902?s=20
If you look at the pace of progress in AI from both open source and other companies (like X.ai), it's clear that while OpenAI is leading at the moment, it's not the only game in town. With time, money, and, importantly, technical talent, another project could catch up and surpass OpenAI.
If enough of the current OpenAI talent jump ship and rally behind a new Sam/Greg AI company, what happens to OpenAI then and the bet that Microsoft placed?
Clearly, Microsoft will need to act. The simplest (and thus arguably most likely) scenario would be for Microsoft to guarantee another multi-billion dollar investment, but with the following conditions:
The current board is replaced entirely (and Microsoft now gets a seat on the board)
Sam and Greg are brought back, at least for a period of time.
Of course, there are many variations of this approach (including just an outright acquisition of OpenAI). However, any solution must address a key principle: restoring confidence in OpenAI's leadership and future direction. At this point, I do not know how that could be done with the current board. Ironically, in firing Sam, the current board has likely signed their own death warrants in that they will, in turn, get fired.
Microsoft needs to act fast, though. Sam and Greg are 100%-for-sure fielding calls now from big-money investors and other AI contenders. If it weren't for the bad blood between Sam and Elon Musk, having Elon Musk's X.ai company acquire OpenAI would be an intriguing choice—ditto for Google, who has struggled to catch up despite actually inventing the key technologies behind OpenAI.
Stay tuned, everyone! The coming days and weeks will be interesting.
Setting aside the people drama of the moment, the real news happened last week with the launch of Custom GPTs: https://openai.com/blog/introducing-gpts. We just saw the "iPhone AppStore" moment in the journey of AI progress, and I predict this advance will impact nearly every business on the planet.
More on this in the next post!
ThoughtfulBits: Ideas that Matter is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber.