A Declaration of Support Signed by the Vast Majority of 770 Employees

On Friday, November 17, 2023, Sam Altman was ousted from his position as CEO.

OpenAI employees were stunned. 'What on earth just happened?' Confusion and anger spilled into hallways, meeting rooms, and Slack messages. The man who had built the company to this point was being forced out with no explanation.

The weekend passed, and Monday morning arrived. The board appointed someone named Emmett Shear as the new interim CEO , a figure known for co-founding Twitch. The board may well have thought, 'It's over. We won.'

The employees thought otherwise.

That afternoon, a document began circulating inside OpenAI. The title was simple: 'We stand with Sam.' The contents were stronger still. 'If the board does not reinstate Sam Altman and Greg Brockman, we will all resign and join Microsoft.'

At first, a few dozen people signed. An hour later, the count was past 100. By evening, it had crossed 500. The next morning, signatures broke through 700.

The total headcount was 770. That meant nearly 95 percent of the company had signed.

Among them were researchers, engineers, and designers. HR, finance, and legal staff signed too. Even Mira Murati , the interim CEO the board had installed after ousting Sam , added her name. Was she betraying the board that had made her CEO? No. She loved the company, and believed it couldn't survive without Sam.

What made it even more striking was that the employees didn't stop at signing.

Employees began posting the same phrase on X (formerly Twitter): 'OpenAI is nothing without its people.' One by one, then in twos, then from hundreds of accounts, the same sentence appeared. It had the rhythm of a protest chant.

Sam Altman retweeted every single one of those messages from his own account. Quietly, without a word, he simply pressed the like button.

Why did so many people rally behind Sam Altman?

Was it money? Partly, yes. OpenAI employees didn't receive stock, but they held something called PPUs , a special right that would pay out handsomely if the company did well. With Sam gone, the company looked like it might collapse, and those PPUs would turn into worthless paper.

But it wasn't only that.

Sam Altman gave his employees a dream. 'We can build AGI. We can change the course of human history.' That might sound like bluster, but for the people who actually built ChatGPT, it was real. They genuinely believed they were changing the world.

Sam trusted his employees. He gave them enormous responsibility, let them make their own decisions, and pushed them to move fast. When someone made a mistake, he said, 'That's fine , try again.' Employees felt Sam respected them as human beings.

So when Sam was ousted, his employees felt betrayed.

'The board made this decision without listening to us at all.' 'This is our company , are they saying our voices don't matter?' 'Can we build AGI without Sam? Can this company even survive without him?'

The employees' signatures were not a simple protest.

It was an ultimatum.

Microsoft had already announced it was bringing on Sam and Greg. CEO Satya Nadella stepped forward personally and said, 'Sam Altman will lead our new AI research division. Any OpenAI employee who wants to follow him is welcome.'

Those weren't empty words. Microsoft had invested $13 billion (roughly 17 trillion Korean won) in OpenAI. It had deep pockets, plenty of positions, and all the infrastructure needed. Taking in 700 employees was effortless for a company of that size.

The board must have been terrified.

'Are 700 people actually going to walk out?' 'That would leave only 70 people in the building.' 'You want 70 people to build GPT-5?' It was impossible.

OpenAI was built from people, not buildings. The people who knew how to train GPT models, the people who knew how to run the servers, the people who managed the customers , if they all left, OpenAI would become a ghost company with nothing but a name.

The investors were furious too.

Prominent firms like Sequoia Capital and Thrive Capital called the board directly. 'What have you done?' 'Are you planning to destroy an $86 billion company overnight?' 'Bring Sam back immediately.'

Salesforce CEO Marc Benioff also stepped in. 'OpenAI talent , come to us. We'll pay more and sort out your visa situations.' Résumés from OpenAI employees started flooding into Google DeepMind as well.

OpenAI was on the verge of complete collapse.

And all of it happened in just five days.

Sam was fired on Friday. Chaos followed over the weekend. Employees signed the letter on Monday. The board surrendered on Tuesday. It was the fastest CEO reinstatement in Silicon Valley history.

Where did that loyalty come from?

Sam Altman had a certain kind of charisma. He wasn't a commanding speaker, and he didn't have a big voice. But when he talked with someone, he looked them in the eyes and genuinely listened. He didn't just say 'your opinion matters' , he actually incorporated employees' ideas into the product.

ChatGPT itself had come from an idea put forward by employees.

At first, no one had thought about building a chatbot. Then a few employees said, 'People seem to enjoy having conversations in the GPT-3 Playground.' Sam said, 'Then turn it into a product,' and within a few weeks, ChatGPT was out in the world.

Sam was not afraid of failure.

'Build fast, ship fast, learn fast.' That was his philosophy. Don't wait for perfection , put it out into the world and watch how people respond. Inside that kind of culture, employees were free to experiment and take risks.

So losing Sam wasn't just a matter of losing one CEO.

It meant the company's entire culture, vision, and way of working would collapse. The board thought that removing one person , Sam , would settle things. But the employees believed that Sam was OpenAI.

The 700 signatures saved the company.

To be precise, they protected the company's soul. If the employees had stayed quiet, Sam would have gone to Microsoft. OpenAI would have kept its name but lost everyone who mattered, and it might have faded into history.

But the employees stood up.

They fought for the company they had built, the leader they believed in, and the future they had dreamed of. And they won.

This episode showed what a 21st-century company can look like. In the old days, the board made every decision. But now it became clear where the real power in a company lies. In a technology company, the most important asset is not the buildings, not the patents, and not the money.

It's the people.

And the dream they are building together.

The Board's Surrender and the CEO's Return

By Monday night, the situation looked hopeless.

The board was surrounded on all sides. Seven hundred employees had declared they would leave; Microsoft had said it would take them all; investors were furious and preparing lawsuits. Even Mira Murati, the interim CEO the board itself had installed, had signed the letter.

And yet the board did not give up.

More precisely, they couldn't give up. They had been the ones to fire Sam, and their pride would not let them turn around and say, 'Sorry, we were wrong.' So they looked for another way.

'Let's find a CEO who isn't Sam!'

The board spent the entire weekend searching for a new CEO candidate. Among the people they reached out to, one name stood out above the rest: Dario Amodei, the CEO of an AI company called Anthropic.

What kind of company is Anthropic?

It was founded by people who left OpenAI. They thought Sam Altman's rapid commercialization approach was dangerous, so they walked out. Their argument was: 'We'll build AI more safely and more ethically.'

From the board's point of view, Dario was a perfect candidate. He understood AI technology, cared about safety, and seemed unlikely to move as fast as Sam. There were even reports that the board floated the idea of merging the two companies.

Dario said no.

'Sorry, but we're going our own way.' He had no intention of walking into OpenAI's chaos.

The board's Plan B had also failed.

On Sunday evening, the board announced a man named Emmett Shear, the founder of Twitch, as the new CEO. The board probably thought: 'Good. It's over. Sam isn't coming back.'

The employees grew even angrier.

'So they're completely ignoring what we want?' 'Who on earth is Emmett Shear?' 'Can the guy who built Twitch even understand ChatGPT?' Angry messages exploded across social media.

Microsoft CEO Satya Nadella spoke publicly.

'We will work well with CEO Emmett Shear. At the same time, we welcome Sam Altman returning to OpenAI. The choice rests with the OpenAI board and its employees.'

What he was really saying was: 'We don't care either way. Whether Sam comes back or not, we win. If he doesn't come back, 700 employees come to us. If he does, our investment is safe.'

Tuesday morning arrived.

Sam Altman appeared at OpenAI headquarters, wearing a visitor's badge. He walked into the company he had founded as a guest. He posted a photo on X: 'First and last time visiting OpenAI as a visitor.'

That single photo spread around the world.

People laughed, cried, and raged. 'The founder walks into his own company as a visitor?' 'The board has truly lost its mind.' Public opinion was entirely on Sam's side.

Negotiations began.

Adam D'Angelo (CEO of Quora) appeared as the board's representative. Sam Altman, Greg Brockman, investors, and even Emmett Shear took seats at the negotiating table.

Sam's demands were clear.

"Overhaul the board entirely. If the people who pushed me out are still there, I will never come back."

The board was cornered.

Accept, and they would suffer a total defeat. Refuse, and the company would collapse. They had to choose one or the other.

At 10 p.m. on Tuesday, a post appeared on OpenAI's official account.

"We have reached an agreement in principle for Sam Altman to return as CEO. The new board will consist of Bret Taylor (Chair), Larry Summers, and Adam D'Angelo."

It was a return after just five days.

Fired on Friday, back on Tuesday. It was the shortest firing in Silicon Valley history. Steve Jobs returned to Apple after 11 years. Sam came back in five days.

The makeup of the new board was interesting.

Bret Taylor was the former co-CEO of Salesforce, a massive technology company. Smart, experienced, and seen as neutral , he became chair.

Larry Summers was a Harvard professor and former U.S. Secretary of the Treasury. An economist with government experience and a prestigious reputation. That kind of person became a board member.

Adam D'Angelo was the only one who stayed, even though he had voted to remove Sam. Why was he kept? It was likely a negotiated outcome. There may have been a deal along the lines of: "You played a mediating role, so we'll give you one seat."

Ilya Sutskever was out. Helen Toner and Tasha McCauley were gone too. Everyone who had voted to remove Sam lost their seat.

Sam Altman posted a brief message on X.

"I love OpenAI. Everything I did over the past few days was to protect this team and its mission. When I agreed to join Microsoft on Sunday evening, I thought that was the best path for me and the team. With the support of the new board and Satya, I am now returning to OpenAI."

It was not a declaration of victory. It was a message of reconciliation.

Employees cheered. "OpenAI is nothing without its people" filled social media again , this time as a celebration.

Not everything had a happy ending.

Wounds remained. Trust had been broken. Ilya Sutskever had been Sam's closest colleague, and yet he had betrayed him. Would those wounds heal easily?

Cracks appeared inside the company too. The conflict between two camps , "Safety comes first" versus "We need to move fast" , had surfaced openly. Sam's return did not make that argument disappear.

On the outside, people began to see OpenAI differently.

"Can this company be trusted?" "What if the CEO gets fired again?" "Is it safe to invest?" Cracks had formed in its reputation.

Competitors saw an opening. Companies like Google, Anthropic, and Mistral began marketing themselves as stable alternatives. "We're not chaotic like OpenAI. Choose us."

Microsoft came out as the biggest winner.

Satya Nadella made a series of brilliant moves. He immediately offered Sam a job, said Microsoft would take in the entire staff, and at the same time kept lines open with OpenAI's board. He built a structure where Microsoft would win no matter which side prevailed.

Microsoft's stock actually rose on the news of Sam's return. Investors concluded that Microsoft had taken full control of OpenAI.

One thing had changed.

Microsoft joined the board as a non-voting observer. No voting rights, but a seat in every meeting with full visibility. The message was clear: "We will never again find out five minutes before it happens."

OpenAI was no longer a fully independent company.

On November 29, the official announcement came.

"CEO Sam Altman, President Greg Brockman, and CTO Mira Murati have officially returned. OpenAI will continue to pursue its mission of developing safe and beneficial AGI."

The company appeared to have returned to normal.

But internally, much had changed. Sam Altman's power was stronger than ever. There was no longer a board capable of stopping him. The new directors had come to support Sam.

The voices that had prioritized safety grew quieter. Ilya was gone, and Toner and McCauley had left too. OpenAI had now shifted decisively toward "faster, and faster still."

The employees felt relieved, but also a little uneasy.

"We won. Sam is back." That felt good. But at the same time, a question lingered: "Are we really on the right path?"

The five-day drama was over.

The king had returned. But the kingdom was not the same as before. There were wounds, there was suspicion, and things had changed.

This episode left a lesson not just for OpenAI, but for tech companies around the world. "A board can fire a CEO, but employees can bring him back." "The real owners of a company are not those with power on paper, but the people doing the work."

Sam Altman came back stronger. He had confirmed that his team cared enough to fight for him. He had also confirmed that he had built a team capable enough to keep the company running without him.

But he also came away humbled.

"The cleanup is harder than the crisis itself." He said this later. Healing the wounds after his return, rebuilding trust, pulling the team back together , all of that was far harder than simply coming back as CEO.

The five-day war was over. The real fight was just beginning.

Ilya Sutskever's Apology

Ilya Sutskever.

You cannot understand OpenAI without knowing this name.

He was a co-founder of OpenAI, alongside Sam Altman. He served as chief scientist, was one of the people who built GPT from the ground up, and was a legendary researcher in the field of AI. He had studied under Geoffrey Hinton , often called the godfather of AI , and was a central figure in the deep learning revolution.

If Sam Altman was a business genius, Ilya was a technical genius.

Together they were a perfect pairing. When Sam said, "Let's build this," Ilya would answer, "I know how." While Sam was out persuading investors, Ilya was in the lab training models.

OpenAI's employees held Ilya in deep respect.

"Without Ilya, there would be no GPT." "He's one of the few people who truly understands AI." "If Sam is the face of the company, Ilya is the brain." These were the kinds of things people said inside the company.

And yet, this was the same Ilya who had voted to push Sam out.

On that Friday, when the board fired Sam, Ilya was there. He voted yes. He had ousted the friend, the colleague, the partner with whom he had built the company.

Why?

Ilya took AI safety more seriously than almost anyone. He believed that AGI , artificial general intelligence , could save humanity, but could also destroy it. That was why he thought the work had to proceed slowly, carefully, and safely.

Sam Altman was different.

Sam pushed for "build fast, ship fast, change the world." When they launched ChatGPT, when they built GPT-4, speed always came first. "We have to move faster than the competition." "We have to own the market before Google catches up."

Ilya was uneasy.

"Are we moving too fast?" "Did we do the safety checks properly?" "If we keep rushing like this, something will go wrong." His worries kept growing.

From the summer of 2023, cracks began to form between them.

Sam held developer conferences, announced new features, raised investment, and expanded rapidly. Ilya kept saying, "Wait , this wasn't what we originally planned."

They argued repeatedly.

"Safety comes first" versus "speed is everything." "Let's go slowly" versus "you have to move fast to win." "Let's hold to the nonprofit mission" versus "the mission only survives if the company does."

The conflict surfaced in the boardroom too.

Ilya spoke with the other board members , Helen Toner and Tasha McCauley. "Sam is going too far. We need to do something." They agreed. "You're right. OpenAI is drifting from its original purpose."

On Friday, November 17, they acted.

Ilya followed his conscience. "I have to save the company. If Sam keeps going like this, it's dangerous." He genuinely believed that.

When Sam was fired, the company fell into chaos. Greg Brockman resigned. Employees were furious. Investors protested. Microsoft stepped in.

Ilya was stunned.

"I never expected this." He thought the employees would back him up. "We saved the company. Everyone will be grateful." But it was the exact opposite. The employees called him a traitor.

When Monday's letter of 700 signatures appeared, Ilya's name was on it.

He had turned around at the last moment. The very person who had agreed to push Sam out was now signing a letter demanding Sam's return.

And he posted on X.

"I deeply regret my involvement in the board's actions. I never intended to harm OpenAI. I love everything we built together, and I will do everything I can to reunite the company."

That post was a bombshell.

The central figure behind Sam's ouster had apologized publicly. The board's moral authority collapsed entirely. "If even Ilya has regrets, this decision couldn't have been right."

Sam Altman replied with a heart emoji.

❤️

He said nothing. Just sent a single heart. Was it a sign of forgiveness? Or did it mean "we'll talk later"?

The employees found themselves in a tangle of complicated feelings.

Should they hate Ilya? He was a traitor. Or should they try to understand him? He loved the company, cared about safety, and acted with good intentions. He just went about it the wrong way.

After Sam returned, Ilya stayed at the company.

At first it seemed fine. Ilya kept doing research, Sam ran the company, and everyone said "we're okay now."

But inside, things were different.

Ilya was no longer a close colleague of Sam's. They sat together in meetings, but the easy jokes of earlier days were gone. Trust had been broken. Once broken, it doesn't come back easily.

In May 2024, Ilya left OpenAI.

Officially, the announcement said he was leaving "to start a new project." But everyone knew the real reason. They couldn't be together anymore.

Ilya started his own AI company.

It was called Safe Superintelligence (SSI). Even the name says something. "Safe superintelligence." He was going to pursue the dream he couldn't realize at OpenAI.

Two geniuses had set out together to change the world. One a business genius, the other a technical genius. A perfect combination. But they disagreed on how to get there. One said "fast," the other said "slow."

In the end, they parted ways.

Ilya's apology was the most human moment in the whole drama.

He made a mistake, owned it, and asked for forgiveness. Nobody is perfect. Even geniuses make mistakes. What matters is the courage to admit them.

Sam Altman was generous too. After returning, he didn't push Ilya out immediately. He gave him a chance to keep working together. Ilya left eventually, yes, but Sam didn't force him out.

This episode raises a question.

When building AI, which matters more: speed or safety? Should you move fast and change the world, or move slowly and make sure it's safe?

There's no clean answer. Ilya wasn't wrong, and Sam wasn't wrong. Both were right in their own way, and both wanted to help humanity on their own terms.

The tragedy is that they couldn't go together.

The Incident Carries Layered Meaning

After five days of drama, people struggled to make sense of what had happened.

It wasn't a simple CEO firing.

It was a textbook case of how a twenty-first-century company actually works. Inside it were power, vision, loyalty, culture, and a new era's idea of leadership.

The most important lesson is this.

"Who is the real owner of a company?"

The answer used to be simple. Shareholders are the owners. The people who put in the money own the company, elect the board, and appoint the CEO. If the board doesn't like the CEO, they fire him. End of story.

The OpenAI incident shattered that formula.

The board fired the CEO. Legally, it was a perfectly valid move. The bylaws said so, and the procedure was followed. But the company nearly collapsed.

Why?

Because the employees refused. Seven hundred people declared, "We follow Sam, not the board." The employees, who held the real power, beat the board, which held the legal authority.

This was a revolution.

No blood was shed, unlike the French Revolution, but the power structure flipped completely. The top-down command system collapsed, and the bottom-up expression of will won.

How did this become possible?

Because OpenAI was a special kind of company. It wasn't a factory. Its assets were human minds. The people who knew how to build GPT, who could write the code, who could train the models. Without those people, the company was just an empty building.

Microsoft understood that.

"We don't need the building. Just bring the people." That single line from Satya Nadella brought down the OpenAI board. It proved that human capital, not physical assets, is where the real value lies.

What is corporate bureaucracy?

It's a system that prizes rules, procedures, and formalities. "Board resolutions matter." "That's what the bylaws say." "We're legally in the right." That kind of reasoning.

That's not wrong. Companies need rules. But what happens when following the rules causes a company to lose sight of its essence? The OpenAI board fell straight into that trap.

They were formally correct.

"The CEO wasn't being straight with the board." "We have to protect the safety of the company." "The bylaws give us the authority." All of it was true. In court, the board might have won.

But in practice, they were completely wrong.

They couldn't read the employees' feelings. They didn't understand the company culture. They didn't know why Sam Altman mattered. They saw only the power on paper and missed where the real force actually lived.

What is a vision?

It's a powerful dream about the future. "We're going to build AGI. We're going to change the course of human history. This isn't just a company , it's a mission." That is exactly what Sam Altman gave his employees.

The employees weren't at OpenAI for the paycheck.

Money mattered, of course. OpenAI's PPUs held enormous value. But that wasn't the whole story. They could have earned more at Google or Meta.

They stayed at OpenAI because of the dream.

"We built ChatGPT. The world changed. What do we build next?" That excitement, that possibility, that challenge. That's what moved them.

Sam Altman was the symbol of that dream.

The employees believed that without him, there was no dream. That's why 700 people signed. Vision mattered more than money, more than stability, more than career.

The board didn't know this.

They thought, "The company runs fine without Sam." They saw the CEO as replaceable. They were wrong. Sam wasn't just a CEO , he was the embodiment of the vision.

What this incident showed is that in a twenty-first-century company, vision is stronger than power.

The dream in people's hearts carries more force than the rules on paper. The collective will of employees wields more influence than the voting rights of a board.

Many people drew comparisons to Steve Jobs.

In 1985, Steve Jobs was also pushed out of Apple. The board fired him. He returned eleven years later and turned Apple into the world's most valuable company.

Sam Altman came back in five days.

Why so fast? Because the times had changed. In 1985, employees had no way to organize collectively on behalf of a CEO. In 2023, Slack, X, and email let them mobilize in a matter of hours.

Technology changed the power structure.

Thanks to social media, the employees' voices spread instantly around the world. "OpenAI is nothing without its people." That phrase reached millions within hours. The board could not withstand the pressure of public opinion.

The investors had changed too.

Investors of an earlier era would have sided with the board. "Rules are rules. If the CEO caused problems, removing him is the right call." But the investors of 2023 were different. They were pragmatists. "If the company collapses, our money is gone. Bring Sam back."

That is modern capitalism.

Results matter more than rules. Substance matters more than form. Survival matters more than principle.

The OpenAI crisis is a blueprint for the future.

Similar events will happen again. A founder with a powerful vision clashes with a bureaucratic board, employees take the founder's side, and in the end the founder wins. That is the pattern.

Is that a good thing?

It is a complicated question. On one hand, yes. A leader with vision can change the world without bureaucracy getting in the way. Like Sam Altman.

On the other hand, it is dangerous. A leader with the wrong vision becomes just as unstoppable. What if Sam had genuinely made a dangerous decision? What if the board couldn't stop him? Who would be held accountable?

That was exactly what Ilya Sutskever feared.

"Sam is moving too fast. Someone needs to hit the brakes." His concern wasn't wrong. AI can be genuinely dangerous. But the method he chose was the wrong one.

Balance is needed.

Vision is needed, and so is accountability. Speed is needed, and so is safety. A leader's freedom is needed, and so is board oversight.

OpenAI failed to find that balance.

After Sam's return, the company tilted entirely toward "speed." Every voice that had worried about safety was gone. Now, who is left to stop Sam?

The new board members came to support Sam. Bret Taylor, Larry Summers. Fine people, but not the kind who would ever fight him.

That is the price of winning.

Sam won, but in winning he took on a far heavier burden. There is no one left to stop him. If he makes a mistake, that mistake becomes OpenAI's mistake, and could become the mistake of the entire AI industry.

That is the true meaning of this episode.

"In twenty-first century companies, people are everything." Not buildings, not patents, not money. People. The vision that binds them together. The culture that believes in that vision and follows it.

The board held only power on paper.

The employees held real power. When the two collided, real power won.

That is the victory of a powerful vision over corporate bureaucracy.

History books will one day read: "In November 2023, 700 OpenAI employees stood up for their CEO. They toppled a board that held only formal authority and defended the vision they believed in. It was a turning point in twenty-first century corporate history."

Sam Altman is not merely a CEO.

He represents leadership for a new era. A leader who guides through vision, not commands. A leader respected through trust, not authority. A leader who moves people through a dream, not rules.

OpenAI is not merely a company.

It is a revolution. A collective charge toward the future. That is why the employees fought for it with everything they had.

Founders around the world study this episode. "How do I make my employees fight for me?" Boards draw their own lesson. "You can't touch a CEO carelessly."

Investors come away changed. "People matter more than rules." Employees gain confidence. "We have power too."

That is the greatest legacy the ousting and return of Sam Altman left on twenty-first century corporate history.

The moment vision defeated power. The moment people defeated rules. The moment the future defeated the past.

And the moment one person's dream moved 700 hearts and ultimately changed the fate of a vast company.