The Core of the Warning
In recent interviews and public appearances, Sam Altman has articulated a significant fear: that artificial intelligence could become controlled by a handful of companies, individuals, or models. This concentration, he argues, would strip the vast majority
of people of any meaningful say in how this transformative technology shapes the world. Speaking on a recent podcast, Altman stated, "The right approach is to say we want people deeply in control of the future." His warning is not just about economic monopolies, but about a future where the direction of society is steered by the choices of a select few, rather than by broad social consensus. He envisions a world where AI and society should "co-evolve," a process that is impossible if power is not widely distributed.
What is Concentrated AI Power?
The concentration Altman speaks of refers to a scenario where the most advanced AI—often called frontier models—is developed and held by a very small number of entities. Building these systems requires immense resources: billions of dollars for computational power, massive and proprietary datasets for training, and elite teams of specialised researchers. This high barrier to entry naturally leads to a consolidation of power among a few tech giants and well-funded labs. The threat is that these few players would not only dominate the market but also embed their own values and objectives into the AI systems that could one day manage everything from public information to critical infrastructure. This creates what Altman calls the risk of being locked into one company's or one person's "moral worldview."
The Paradox of Safety and Liberty
Interestingly, Altman points to a paradox where the very fear of AI's existential risks could lead to this undesirable concentration of power. He worries that in a panic to prevent a scenario where AI escapes human control, society might opt for heavy-handed regulation that only a few large companies can comply with. This creates a trade-off: sacrificing liberty for a perceived sense of safety. "There are a lot of people who are so nervous about the magnitude of those risks... and feel a need to protect the world," Altman explained, cautioning against an approach that says, "'we should trade off a lot of liberty for safety.'" These comments are seen by many as a critique of rivals like Anthropic's CEO, Dario Amodei, who has advocated for stronger government regulation due to the potential dangers of advanced AI.
OpenAI's Complicated Position
Altman's warning is complicated by the fact that his own company, OpenAI, is a primary example of the power concentration he describes. As the creator of hugely influential models like GPT-4 and its successors, OpenAI is one of the very few entities at the frontier of AI development. While the company has not made its most powerful models open-source, it has released some smaller open-weight models. Altman's stance is that democratising the benefits of AI is crucial. He has stated that OpenAI's goal is to be a platform that empowers others and has even suggested that a global regulatory body, similar to the International Atomic Energy Agency, could help manage AI's development. This highlights the fine line the company is trying to walk between leading AI development and preventing the dystopian outcome it fears.
The Path to Empowering Individuals
Despite his concerns, Altman remains optimistic about AI's potential to empower individuals if developed correctly. He has said that the AI industry has, so far, done a poor job of explaining the technology's benefits. He foresees AI driving "the greatest boom in people starting smaller businesses than we have ever seen." The key, in his view, is to ensure that AI tools are widely accessible, fostering a wave of entrepreneurship and creativity rather than consolidating power. The debate is shifting from purely technical safety to questions of power structures and governance. The central challenge is how to balance the need for safety and oversight with the goal of keeping the technology's power distributed among the many, not concentrated in the hands of the few.
