Discovery Digest Volume III: WTF is Happening with OpenAI?
A list Interesting Posts, Articles, & Tools the P2P Team Discovered This Week
Part of the job of a Product Manager is to keep their ears to the ground. You must be aware of innovations, changes, and new tools in the technological landscape to avoid falling.
Every week I find myself bookmarking tweets, saving articles, and exploring new methods to do my job.
Discovery Digest has become the new P2P series to spread wealth. In it, we'll share the best or most interesting news, developments, and tools that have captivated our attention.
This past weekend, there was only one story so weird, so captivating, it gets its own dedicated issue of Discovery Digest.
That Story? What is happening with the Board of Directors, Sam Altman, and OpenAI?
LAST CHANCE FOR 20% OFF ALL P2P Offerings
As this article took me WAY longer than I expected, I decided to extend this year’s Black Friday / Cyber Monday sale
Subscribe now for 20% off P2P Premium newsletter - the best price it’s ever going to be.
Additionally, use the Promo code BFCM24 at checkout for 20% off any of the following P2P services:
Got a product resume you need to up-level? I’ll review your resume, then talk tactics on a call.
Are you struggling with a product problem today? I’ll hop on a call and help you put forth the game plan to solve your issues.
Got a big interview you want to nail? I’ll act as a hiring manager and conduct a mock interview, with a follow-up after to review your performance and improve tactics and strategies.
Now, back to the article.
Primer: Highlevel Background On Open AI, and the Conflict at Hand
OpenAI is the parent company most famous for creating ChatGPT - the quickest product ever to gain 100 million active users.
The secondary market was valuing OpenAI at up to $90 billion
They grew so fast, that they almost had to partner with Microsoft to deal with the demands of such a wildly successful product. Microsoft invested $10 billion into the company almost a year ago today.
It's why the world was totally, utterly shocked when OpenAI Unceremoniously fired their Silicon Valley golden boy CEO Sam Altman.
With Sam’s dismally, complete and utter chaos ensued.
Why was Sam fired?
This has been a wildly confusing topic, pretty much from the start, as the board of OpenAI has been the diametric opposite of transparent throughout this entire saga.
To understand the why behind the initial firing, it’s best to understand who this board is.
The following tweet from notable Silicon Valley investor Chamath Palihapitiya outlines the unique dynamics of the Board structure well
To summarize the tweet:
Early Innovations and Transition to "Capped-Profit" Model: Founded as a non-profit in 2015, OpenAI made significant strides in AI research and transitioned to a capped-profit model in 2019 in order to raise capital, as AI development gets expensive. This transition was while remaining controlled by its non-profit entity.
Developments in AI and ChatGPT: OpenAI introduced groundbreaking AI models, including GPT-3 and DALL-E, leading to the release of ChatGPT which gained rapid global attention and further advancements with GPT-4.
Leadership Changes and Controversy: OpenAI removed co-founder Sam Altman as CEO due to a lack of consistent candor in his communications with the board, amidst speculation about disputes over the company's profit versus nonprofit motives and the pace of commercialization.
The for-Profit / Non-Profit The last bullet point - the lack of communication - is where this story totally and utterly goes to sh!t.
What we know off the top, from OpenAI’s initial press release, and subsequent media messaging was that:
There was a struggle between the original non-profit goals and the profit-seeking subsidiary.
There was a battle between the “Deceleration, move slow” people, and the “Acceleration, move fast, and get to market” group. These 2 factions would but heads throughout the chaos.
Sam Altman didn’t do anything overtly illegal/shady/ flagrant that would warrant the typical abrupt firing.
The whole thing Saga reeked of a power struggle. The non-profit board had no fiduciary responsibility to maximize profit, their charter is to Safely Develop AGI (Artificial General Intelligence). Meanwhile, in order to fund the incredible resource demands inputs required
That element was critical to the dynamic going on because the Microsoft / OpenAI agreement revolves around the establishment of this fact. Directly from the agreement:
The board determines when we’ve attained AGI. Again, by AGI we mean a highly autonomous system that outperforms humans at most economically valuable work. Such a system is excluded from IP licenses and other commercial terms with Microsoft, which only apply to pre-AGI technology
This led to man thinking - could this whole fiasco be related to the achievement of AGI?
Whatever was happening, there was a disconnect on what should be the north star to be pursued. This can be demonstrated in several behaviors from board members.
Discord in the OpenAI board
The member composition of the OpenAI board prior to the events did not represent that of a $90 billion company. It included the following individuals, each of whom at their own Saga and part to play.
Greg Brockman (OpenAI Chairman & President)
Ilya Sutskever (OpenAI Chief Scientist)
Sam Altman (OpenAI CEO)
Adam D'Angelo (CEO of Quora and an independent director)
Tasha McCauley (Technology entrepreneur and an independent director, wife of actor Joseph Gordon-Levitt)
Helen Toner (Georgetown Center for Security and Emerging Technology, serving as an independent director)
The Coup was led by Adam, Tasha, Helen, and Ilya. Shortly after Sam was fired, Greg quit, and chaos ensued.
Who is Team Coup
Helen Tonner:
Helen was an academic on the board and the source of many conspiracy theories.
Some suspect she’s a Chinese Spy, as she worked and studied in China
Some suspect she is a US Federal Agent plant, as she currently works for Georgetown.
What we know for certain is, that she butheads with Sam Altman directed when she seemingly criticized OpenAI in a public forum, which praised another company Anthropic. More on Anthropic shortly…
Adam D'Angelo
Current CEO of Quora, he has what some consider a conflict of interest via Poe - an AI Chatbot powered by Quora.
The theory goes, that Adam forced Sam out due to Adam setting course the ability to create custom ChatGPT agents, effectively killing Poe.
Ilya Sutskever
Ilya is the most fascinating board member
Originally a member of the Coup, he’s widely considered brilliant. Here’s what Elon had to say on the matter.
So, when he turned on Sam, people thought some serious malfeasance went down.
Ilya, however, got brought into the Employee rebellion - something which I hadn’t seen before.
After the announcement, in true I am Spartacus Moment, we saw the following tweet from Sam.
What was this? An open rebellion from the staff.
Everyone who shared a heart was indicating they were ready to jump ship with their captain, Sam Altman.
It eventually turned into a letter where 95% of the company agreed that Sam gets re-instated, and the board resigns, or they quit.
Most surprising on the list, the original Brutus himself - Ilya who, apparently after getting chewed out by follow board member Greg Brockman’s wife - came back to team Sam, pleading Ilya see the error of his way.
It worked, as Ilya would flip sides and join the rebellion against the board.
Satya, Emmet Shear, and how an employee rebellion played out
Where I view this story getting most interesting was the power politicking around the talent shuffle.
The main player I’ve yet to, due to their conspicuous lack of presence on the OpenAI board is Microsoft CEO Satya Nadella. Pretty shocking they would give $10 billion dollars to an organization and have no direct oversight of how that money gets managed. One would think Satya could - justifiably be pissed.
But oh no - Satya was completely ready for this.
The reason why no board presence? The good ole’ American government. Microsoft has been in the regulatory anti-monopoly grinder before. Taking a board seat would have brought greater scrutiny to the heat they don’t want. (Such as what’s happening with Google as we Speak)
To avoid such, they took no board seat. But what about the clowns on the board threatening to destroy the company?
He did not care. Satya made sure to lock down Sam Altman and Greg Brockman - something he probably had ready months prior in case a situation like this occurred.
Shortly after the coup - Satya announced the following on X.
Satya crafted a heads I win, tails-you-lose scenario. If OpenAI died, he just acquired its heads, and subsequent staff to follow for free. If they returned to OpenAI, he would prove he has Sam and Greg’s back.
The tweet below sums up my views accurately.
The End Game
The stage began to be set.
Emmet Shear, former CEO and Co-Foudner of Twitch was named interim CEO. Remember Anthropic from earlier? He may have been a consultant, or may own shares (I’ve dug around - this is very unclear)
https://twitter.com/heyBarsee/status/1726868367531765816
It also appeared at one point, the OpenAI team proposed a merger with Anthropic.
Most notably - he is Team Slow Down.
We have clear factions of Team Slow down, AI might doom us all, and Team Move Fast: Let’s scale this bad boy to the moon
In support of Sam (Team Move Fast) roughly 95% of the company signed a letter demanding the board resign, reinstate Sam as CEO, or they quit and (explicitly) join Sam at Microsoft.
Emmet, looking at this whole situation, quickly realized that this whole situation was a complete and utter mess.
So at this point, the OpenAI board has the following set against them.
One of their board members in favor of change flipped
Their temp CEO is questioning their motivations
The entire company is threatening to quit and join Microsoft
They have no goodwill in the public square (X)
They were backed into a corner so, as one might expect, caved.
Sam Altman was restored as CEO with Helen and Tasha being jennisoned.
Interstingly, Adam D'Angelo, CEO of Quora, remained - meaning it’s likely that Helen was the leader of the Coup.
The Takeaway:
This was an exercise of power.
The board thought they had the power to overrule Sam because they had the votes.
In reality, they did not.
Sam had the People, and Microsoft in his pocket, even without the board. That meant he had the talent and the money backing him. Without those 2, OpenAI is nothing but words on legal documents.
The rebellious OpenAI board members took a shot at the king - but we’re woefully unprepared to win the battle that would ensue. They thought they could just fire Sam and “All would be well”.
They were out of their league. Sam is no rookie to the high-stakes tech battles. I can only imagine how many scalps Satya had to take to reach the top spot at Microsoft.
It doesn’t matter if you have the power on paper.
The power to destroy a thing is the absolute control over it
-Dune
Sam had the power to destroy OpenAI
So… he won.