OpenAI CEO Sam Altman says the term “artificial general intelligence” is losing its relevance as advancements in AI make it increasingly hard to define. Traditionally, AGI has referred to AI systems capable of performing any intellectual task a human can do. For years, OpenAI has said its mission is to develop AGI that is both safe and beneficial for humanity.
Speaking on CNBC’s “Squawk Box” last week, Altman said he no longer finds the label particularly useful. He noted that definitions of AGI vary widely between individuals and companies, with some framing it as AI capable of handling a significant amount of the world’s work — a standard he sees as flawed since the nature of work keeps evolving. Instead, Altman argued, the focus should be on the steady, exponential growth of AI capabilities.
A Broader Debate on Definitions
Altman is not alone in questioning the value of the term. Nick Patience, vice president and AI practice lead at The Futurum Group, said that while AGI is an inspiring “North Star,” it often fuels hype rather than clarity. He described it as a vague, sci-fi idea that risks overshadowing real advances in specialized AI.
The term has nevertheless played a role in attracting investment. OpenAI, along with other AI firms, has raised billions on the promise of reaching AGI, most recently achieving a valuation of $300 billion and reportedly preparing a share sale that could push it to $500 billion.
OpenAI recently launched GPT-5 for all ChatGPT users, describing it as faster, smarter, and more useful, especially for tasks such as writing, coding, and healthcare-related assistance. However, the release drew mixed reviews online, with some calling it a modest upgrade rather than a major leap forward. Wendy Hall, professor of computer science at the University of Southampton, said the update was “incremental, not revolutionary,” and called for AI companies to be measured against agreed-upon performance metrics to cut through marketing hype.
Shifting the Conversation
Altman has acknowledged that GPT-5 does not meet his personal definition of AGI, as it cannot yet continuously learn on its own. He has encouraged a shift away from a binary “is it AGI or not” framework toward discussing measurable levels of progress. Looking ahead, Altman still expects AI to deliver breakthroughs in areas like mathematics and science within the next two years. Patience, meanwhile, sees AGI as a distraction — a narrative useful for fundraising but less helpful for meaningful discussions about what AI can actually do.
Author’s Opinion
Framing AI progress around AGI risks turning a nuanced, technical field into a buzzword competition. While the dream of human-level AI can inspire research, it also encourages unrealistic expectations and fuels skepticism when new models fail to live up to the hype. A better approach would be to focus on specific, measurable capabilities and their real-world applications, rather than chasing an ever-shifting goalpost.
Featured image credit: Rappler
For more stories like it, click the +Follow button at the top of this page to follow us.