The Circumstances Around Sam

Just weeks ago, OpenAI was on an upward trajectory, largely thanks to ChatGPT's groundbreaking success, which positioned it as a leading force in AI. In early November, the company unveiled new features for developers, likely solidifying its status as the go-to generative AI provider for startups and large businesses alike.

Remarkably, OpenAI has only 770 employees, a minuscule figure compared to its chief competitor, Alphabet, which boasts over 118,000. Despite its size, OpenAI was on track to reach an $80 billion valuation through private shares—an astonishing achievement of over $100 million in market value per employee.

At the heart of OpenAI's ascent was Altman, a visionary yet unconventional leader. His playful antics, such as joking about achieving Artificial General Intelligence on Reddit, captured attention while he also oversaw the swift rise of one of the fastest-growing applications and the introduction of significant enterprise features.

The board's decision to remove Altman was perplexing, as no one anticipated such a move. As reported by Casey Newton in his Platformer newsletter, the board has not provided a coherent explanation, resorting instead to vague references to "communication issues" and AI safety, alongside requests for Altman's return. The exact reasons for his dismissal may remain a mystery, but it’s clear that OpenAI’s unique structure contributed to this outcome.

A Unique Organizational Structure

Initially founded as a non-profit, OpenAI's mission statement from 2015 reads like a tech utopia: "OpenAI is a non-profit artificial intelligence research company. Our goal is to advance digital intelligence in the way that is most likely to benefit humanity as a whole, unconstrained by a need to generate financial return." However, the landscape has changed significantly since then, especially with the launch of a lucrative API for enterprise clients and a $13 billion funding effort.

In 2015, the idea of developing Artificial General Intelligence with a few billion dollars in funding may have seemed plausible, but the emergence of Large Language Models (LLMs) changed the game. These models require massive amounts of data for training, resulting in exorbitant costs—training GPT-4 alone is estimated to have cost between $60 million and $100 million, with daily operational costs hitting $700,000.

Such expenses are unmanageable for a conventional non-profit. Consequently, OpenAI transitioned into a hybrid model, maintaining elements of its non-profit roots while gaining the flexibility to attract outside investment. This peculiar structure has placed Altman in a unique position—unlike typical CEOs in fast-growing startups, he holds no equity in OpenAI. While he serves on the board, he ultimately answers to its unusual hybrid governance, as evidenced by last week's dramatic events.

A Leadership Void

Silicon Valley is rife with tales of founders pursuing singular visions against board opposition. Founders often retain enough shares to influence decisions, as seen with Facebook’s rapid growth and its strategic acquisition of Instagram, or Google’s bold move to create Android, both driven by their founders’ determination.

In contrast, Altman's lack of control over OpenAI hampered his ability to steer the company effectively. While the reasons for his firing remain unclear, it’s evident that this lack of authority was a significant vulnerability for the organization.

Future Directions

What lies ahead? Following his ousting, Altman accepted a prominent role at Microsoft, OpenAI’s largest investor. Reports indicate that over 700 OpenAI employees signed an open letter expressing their intention to join him. Shortly thereafter, news emerged suggesting Altman may return to OpenAI under yet-to-be-disclosed terms.

Even if Altman successfully reclaims his position, the governance issues have been exposed. OpenAI is unlikely to sustain its previous rapid growth, especially with a vigilant board scrutinizing every decision he makes. At best, the company may continue forward while its employees discreetly update their resumes and fend off recruiters eager to offer them lucrative opportunities.

Should many OpenAI staff transition to Microsoft, it’s uncertain how many will adapt to the corporate culture long-term. Historically, Silicon Valley talent often circulates between firms, joins competitors, or establishes their own ventures.

The Impact of Talent Dispersal

This potential dispersal of talent could significantly influence the broader AI landscape. Few individuals possess a deep understanding of today’s advanced LLMs, and many of those experts are likely at OpenAI. If they move to other companies—both within and outside the AI field—the strong prohibitions on non-compete agreements in California will allow them to carry their expertise elsewhere.

Instead of a concentrated pool of 770 experts within one organization, we could see a proliferation of companies, each benefiting from the insights of former OpenAI personnel. This could foster a more diverse ecosystem of AI innovation, effectively replacing the monolithic structure that once existed.

Historical precedents abound, with the decline of Fairchild Semiconductor leading to the emergence of industry giants like Intel and AMD. The dispersion of knowledge from Fairchild was instrumental in establishing the sector that now defines Silicon Valley, embodying the concept of "creative destruction."

Ironically, should OpenAI's expertise spread across the economy, it might bring the organization closer to fulfilling its original mission of enhancing digital intelligence for the benefit of humanity. Revolutions—whether in politics or technology—rarely remain contained. If OpenAI's disintegration catalyzes improvements in AI talent and knowledge in sectors like healthcare, governance, transportation, or green energy, the outcomes could far surpass the development of advanced chatbots, which appeared to be OpenAI’s primary focus before Altman's exit.

The board's decision to undermine their own organization may well be remembered as a monumental governance blunder. It undoubtedly harms OpenAI, its investors, and the startups reliant on its services. However, by scattering its talent and expertise, OpenAI could unwittingly inspire the very AI revolution it was established to champion.

Regardless of OpenAI's fate, ChatGPT is likely here to stay. After testing thousands of ChatGPT prompts over the past year, I've compiled my go-to favorites into a free guide, *7 Enormously Useful ChatGPT Prompts For Creators.* Download your copy today!

Update 11/22 — Updated regarding Altman’s potential return to OpenAI.