Brain Drain at OpenAI: How the Exodus of Key Talent Opens Doors for Rivals in the AI Arena

To go back to this article, go to My Profile and then look at your saved stories.

Rivals Find Opportunity Amid OpenAI's Talent Departure

Despite OpenAI receiving a substantial investment of $6.6 billion last week, the recent upheaval involving the departure of key personnel, including Chief Technology Officer Mira Murati, Chief Research Officer Bob McCrew, and Vice President of Research Barret Zoph, did not seem to deter investors.

However, these three recent exits are part of a continuing trend of significant technical expertise leaving the organization. In recent times, OpenAI has seen the departure of numerous researchers essential to the creation and development of the algorithms, methods, and systems that have established its position as a global frontrunner in artificial intelligence and a well-recognized entity. Additionally, conversations with former OpenAI staff members revealed to WIRED that the company's gradual pivot toward commercial endeavors remains a contentious issue.

"One ex-employee, now with a competing AI firm, mentions that individuals with a penchant for research are being nudged towards product-focused roles at OpenAI, based on conversations with friends still at the company. This individual has also noted that lately, several acquaintances from OpenAI have reached out to explore job opportunities. Furthermore, an analysis conducted by Lightcast for WIRED, which specializes in monitoring job listings to gauge employment trends, suggests a notable shift in OpenAI's recruitment focus. While in 2021, nearly a quarter of its job advertisements were aimed at general research positions, by 2024, this figure had dropped sharply to just 4.4 percent."

The exodus of talent from OpenAI might significantly impact its strategic trajectory and long-term achievements. Specialists and previous staff members acknowledge that the organization possesses a solid roster of skilled individuals. However, they also note that rising competition is making it increasingly difficult to preserve its competitive advantage.

The most recent high-profile exit, announced on Thursday, involves Tim Brooks, the leader of OpenAI's Sora AI video creation initiative. Brooks shared on X that he is set to move to Google DeepMind, one of OpenAI's top competitors.

"A former OpenAI employee, who has since transitioned to an academic career, believes the losses might begin to alter the situation. They have requested anonymity to avoid jeopardizing their cooperative ties with the AI sector."

Currently, according to this individual, a significant number of students continue to prioritize OpenAI as their preferred employer. The company is perceived to be ahead of its competitors by several months, and potential hires are usually prepared to overlook any perceived internal conflicts to join the organization. However, candidates are also attracted to the opportunity of collaborating with specific researchers or groups, and their preferences may shift as more prominent researchers depart for competing AI firms or start their ventures.

An examination of OpenAI's key studies highlights the significant turnover in its workforce. Analysis of the authorship behind an initial iteration of OpenAI's groundbreaking GPT language technology indicates that less than 50% of the 31 contributors are still with the organization, based on information gathered from LinkedIn and various public online profiles. A notable portion of the group that pioneered GPT exited OpenAI in 2021 to establish Anthropic, which has emerged as a formidable competitor. Approximately one-third of the individuals acknowledged in a technical blog post about ChatGPT have departed since its publication.

A crucial inquiry centers around whether OpenAI possesses a sufficiently robust team of AI experts capable of maintaining its lead in significant AI breakthroughs, a necessity for validating its rapidly increasing market value.

According to a New York Times report referencing a document shown to potential investors, OpenAI is setting its sights on achieving annual revenues of $100 billion by the year 2029. The document outlines projections that ChatGPT is expected to generate $2.7 billion in revenue for the current year, a significant increase from $700 million in 2023, alongside an additional $1 billion from alternative revenue streams. Meanwhile, the competitive landscape is heating up, with tech behemoth Google and other companies investing heavily in rival AI technologies. Furthermore, the tech titan Meta is offering AI models and tools at no cost, which frequently rival the quality of those offered by commercial entities.

OpenAI faces a significant challenge in replacing Ilya Sutskever, one of its cofounders and the key technical mind behind many of its initial projects. Sutskever left in May to start his own AI venture, named Safe Superintelligence.

Sutskever led the way in advancing deep learning and was among the first to advocate for the idea that enhancing the computational power behind model creation would lead to more advanced AI—an insight that turned out to be incredibly accurate. Additionally, he was pivotal in recognizing the potential of a neural network structure called the transformer, created by Google's team, as a significant element in enhancing AI's ability to understand and generate language.

Sutskever played a role in the temporary removal of CEO Sam Altman from his position in November 2023, but he soon expressed regret and advocated for Altman's return shortly thereafter. This incident appeared to spark a debate within the organization regarding Altman's leadership, the company's business direction, and its willingness to take risks. According to an article in The Wall Street Journal, Murati frequently found himself amidst these internal disputes, especially concerning the rapid launch of the company’s voice interface.

Following Sutskevar, Jan Leike also exited the organization, where he had been pivotal in leading a group focused on the future implications of AI technology. Then, in August, John Schulman, one of the original founders of OpenAI and a well-regarded research scientist, disclosed his decision to leave. Schulman had contributed to the development of methods for enhancing large language models through human feedback, employing reinforcement learning techniques. This strategy was instrumental in the creation of an effective and communicative chatbot, similar to ChatGPT. Additionally, another prominent figure, Andrej Kaparthy, who had returned to the company in 2022 after serving as the director of AI at Tesla, departed in February 2024. Kaparthy left to establish Eureka Labs, a startup focused on AI in education.

A former OpenAI staff member, who requested anonymity due to restrictions from their present employer, highlighted the importance of the research contributions made by those who have recently left the company. Despite these significant departures, they pointed out that OpenAI continues to possess a pool of exceptionally gifted young researchers. Furthermore, the organization is in a strong financial position to attract new talent with competitive salaries, bolstered by its increasing market value.

OpenAI has recently secured another round of funding as it transitions into a profit-driven entity. Among its initial founders was Elon Musk, who contributed $45 million in initial funding. Following Musk's departure, Altman transformed OpenAI into a limited profit-oriented company governed by a board of directors that operates independently of investor influence. The attempt at a takeover last November highlighted the inherent challenges of this governance model amidst the rapid expansion and significant investment in the tech sector.

A previous employee of OpenAI, who also requested anonymity, verified that the emphasis on product development was still a source of discord, and mentioned that they had been informed by present employees to anticipate further unrest within the organization.

Former employees describe the internal situation at the company as multifaceted. "For some, the atmosphere is grim, while others do not feel the same way," remarked one individual.

In response to inquiries regarding the matter, OpenAI directed WIRED to a communication from Altman to employees following the exits of Murati, McGrew, and Zoph, which he also posted on X. In this message, Altman identifies a number of researchers who will ascend to higher positions, amongst them Mark Chen, who will be elevated to senior vice president of research, and Jakub Pachocki, who will take on the role of chief scientist at the company.

"Shifts in leadership are a common occurrence in businesses, particularly those that expand rapidly and have high demands," Altman stated. "Clearly, I won't claim that it's typical for such a change to happen so suddenly, but we are not your average company."

"Believing that only a handful of people possess a unique talent, and that it has just left the premises is a concern," states Simon Johnson, a faculty member at the Sloan School of Management at MIT.

Nevertheless, Johnson believes that what OpenAI considers a loss could actually benefit the wider technology sector.

"He believes that there is a significant need for increased competition in the AI sector, a variety of business models, and a higher influx of intelligent individuals offering innovative solutions. Consequently, he expresses his ease with the departure of individuals from the field."

After OpenAI's leadership dismissed CEO Sam Altman in November, employees voiced their dissent and considered resigning, with numerous individuals rallying around the message "OpenAI is nothing without its people" on X. It appears the company is keen on exploring the extent to which it can substitute some of its crucial team members.

Recommended for You …

Delivered to your email: A selection of the finest and most unusual tales from the archives of WIRED

Elon Musk poses a threat to national security

Conversation: Meredith Whittaker Aims to Disprove Capitalist Ideals

What's the solution for dealing with an issue like Polestar?

Invitation: Be part of The Major Q&A session happening on the 3rd of December in San

Additional Content from WIRED

Evaluations and Manuals

© 2024 Condé Nast. All rights reserved. Purchases made through our website may generate revenue for WIRED due to affiliate agreements with retail partners. Content from this site cannot be copied, shared, broadcast, stored, or used in any form without explicit approval from Condé Nast. Advertising Choices

Choose a global website