San Francisco: OpenAI CEO Sam Altman has shed light on why some of the world’s leading artificial intelligence researchers were willing to take what initially appeared to be an “absurd” bet on artificial general intelligence (AGI) and join a relatively young research organisation.
OpenAI was founded around the belief that AGI was possible and that developing it could become one of the most consequential technological projects in history. Altman has previously described the company’s early ambition as extraordinarily high, with the goal of building AGI and ensuring that its benefits were broadly shared.
The explanation offers a glimpse into the thinking that helped OpenAI attract top AI talent at a time when established technology companies had far greater resources and, in many cases, could offer researchers more lucrative compensation.
Why OpenAI’s AGI ambition looked risky
When OpenAI began, AGI remained a highly uncertain proposition.
Unlike specialised AI systems designed for particular tasks, AGI generally refers to an AI system capable of performing a broad range of cognitive tasks at or beyond human levels. There is still no universally accepted test for determining when AGI has actually been achieved.
For researchers joining OpenAI in its early years, therefore, the decision involved betting on a technology whose timeline and feasibility were both uncertain.
The company nevertheless attracted prominent researchers because of its mission and the possibility of working on a problem that could fundamentally change computing and science.
Mission attracted top AI researchers
OpenAI’s early appeal was not simply about salaries or conventional career progression.
The organisation positioned itself as a research laboratory focused on developing highly capable AI while attempting to ensure that the technology benefited humanity.
That mission helped convince researchers to leave larger companies and established laboratories.
Historical accounts of OpenAI’s early recruitment indicate that some researchers were prepared to reject substantially higher offers from competitors because they were attracted by the opportunity to work with a strong group of researchers and pursue the company’s ambitious mission.
The researchers were betting on a technological breakthrough
The central wager was that improvements in machine learning would not remain limited to narrow applications.
Instead, OpenAI researchers believed that increasingly capable models could eventually generalise across different tasks, learn new abilities and potentially approach human-level performance across a wide range of domains.
That idea became increasingly credible as neural networks, large-scale computing and massive training datasets produced rapid improvements in AI capabilities.
The success of systems such as GPT eventually transformed AGI from a distant research concept into one of the central objectives of the technology industry.
ChatGPT changed the AGI conversation
The launch of ChatGPT in November 2022 dramatically changed public awareness of what large language models could do.
Millions of people began interacting directly with an AI system capable of generating text, answering questions, writing code and performing a range of other tasks.
The rapid adoption also accelerated investment in AI research and infrastructure.
At the same time, it intensified debate over whether increasingly capable language models represented a path towards AGI or remained sophisticated but fundamentally specialised systems.
OpenAI’s original bet became more credible
Altman’s comments come against the backdrop of OpenAI’s evolution from a small research organisation into one of the world’s most prominent AI companies.
In his own reflections, Altman said OpenAI was started because its founders believed AGI was possible and could become one of the most impactful technologies in human history.
That original belief has continued to shape OpenAI’s strategy, even as the company has expanded into consumer products, enterprise AI and large-scale infrastructure.
The organisation’s focus on increasingly capable models remains closely connected to its long-term AGI objective.
Researchers also recognised the risks
The AGI bet was never simply about technological optimism.
OpenAI’s researchers have also been concerned about what could happen if AI systems became significantly more capable.
In 2023, several OpenAI researchers raised concerns with the company’s board about an AI breakthrough and its potential dangers. Their concerns became part of the events surrounding Altman’s temporary removal as CEO that year.
The episode demonstrated the tension at the heart of AGI research: scientists want to develop increasingly powerful systems while also ensuring that those systems remain safe and controllable.
Why AGI remains controversial
The idea of AGI itself remains contested.
Some researchers expect systems with broadly human-level capabilities to emerge relatively soon, while others believe current approaches may not be sufficient to produce genuine general intelligence.
A recent analysis of thousands of predictions found that many researchers and technology observers now place major AI milestones in the late 2020s or early 2030s, although such forecasts remain highly uncertain.
This uncertainty explains why OpenAI’s original decision looked so ambitious.
The company was effectively investing years of research and billions of dollars around an outcome whose timing could not be predicted with confidence.
The AGI race has intensified
OpenAI is no longer pursuing AGI in isolation.
Google DeepMind, Anthropic, Meta and other major technology companies are investing heavily in frontier AI research.
The competition has increased the amount of computing power, data and capital being directed towards increasingly capable AI systems.
It has also created concerns that companies could feel pressure to prioritise speed over safety in the race to achieve technological breakthroughs.
From research mission to global AI race
OpenAI’s transformation illustrates how dramatically the AI landscape has changed.
What began as an ambitious research project built around an uncertain hypothesis has become a central player in a global race involving technology companies, governments and investors.
The company’s early researchers were effectively betting that general intelligence could be engineered.
Today, that bet is being pursued at a vastly larger scale.
What Altman’s explanation means
Altman’s account helps explain why OpenAI was able to recruit researchers despite competing against companies with considerably greater resources.
For many scientists, the attraction was the opportunity to work on a problem that could define an entire technological era.
The prospect of being among the researchers who helped create a genuinely general-purpose intelligence was itself a powerful incentive.
That combination of mission, talent and technological optimism became one of the foundations of OpenAI’s rise.
The biggest question remains unanswered
Despite rapid advances in AI, the fundamental question of whether and when AGI will actually arrive remains unanswered.
Current systems can perform an extraordinary range of tasks, but broad capability does not automatically establish that an AI system possesses human-like general intelligence.
The definition of AGI itself remains disputed, making claims about having reached it difficult to evaluate objectively.
For OpenAI, however, the underlying ambition remains unchanged: develop increasingly capable AI while attempting to make its benefits broadly available.
Conclusion
Sam Altman’s explanation of OpenAI’s early AGI bet highlights the mission-driven thinking that helped the company attract some of the world’s leading AI researchers.
The researchers were taking a major technological gamble: that advances in machine learning could eventually produce systems with broad, human-level capabilities. OpenAI’s subsequent growth and the rapid progress of generative AI have made that once-radical idea far more influential, even though AGI itself remains an unresolved scientific and technological goal.
The original bet has also created a second challenge. As AI systems become more capable, the industry must balance the pursuit of breakthroughs with concerns over safety, control and the wider impact of advanced AI.