Sam Altman now says AGI, or human-level AI, is 'not a super useful term’ — and he's not alone

Sam Altman, the CEO of OpenAI, recently expressed that the term 'artificial general intelligence' (AGI) is becoming increasingly irrelevant as the field of AI evolves rapidly. AGI is defined as a type of artificial intelligence capable of performing any intellectual task that a human can manage. For years, OpenAI has aimed to develop AGI that is safe and beneficial for society. In a recent appearance on CNBC's 'Squawk Box,' Altman stated, 'I think it's not a super useful term,' when asked if the latest release of the GPT-5 model brings us closer to AGI. Historically, Altman has suggested that we might see the emergence of AGI in the not-so-distant future. However, he pointed out that the term’s ambiguity is a significant issue, as different organizations and individuals apply varied definitions. For instance, one interpretation describes AGI as an AI that can handle 'a significant amount of the work in the world,' which is problematic given the constantly shifting nature of work itself. Altman emphasized that the focus should be on the exponential growth in AI capabilities rather than the terminology. Altman’s concerns about AGI are echoed by industry experts. Nick Patience, an AI practice leader at The Futurum Group, remarked that while AGI serves as a compelling goal, it often clouds the tangible advancements being made in specialized AI applications. He noted that AGI captures public interest and drives funding but can also generate hype that obscures real achievements. OpenAI and other startups have garnered billions in investments, hinging on the promise of creating a form of AI that could be labeled as AGI. OpenAI has seen its valuation soar to $300 billion, with reports indicating they are preparing a secondary share sale at a staggering $500 billion. Their recent launch of GPT-5, an advanced language model, has been met with mixed reviews. Although OpenAI touts the model as smarter and more efficient, some critics argue it offers only incremental improvements rather than the revolutionary leap many expected. Wendy Hall, a computer science professor at the University of Southampton, criticized the incremental nature of GPT-5, stating, 'By all accounts it's incremental, not revolutionary.' She urged that AI companies should disclose their performance metrics when introducing new products, calling the current landscape 'the Wild West for snake oil salesmen.' Despite these critiques, Altman acknowledged that OpenAI's latest model does not fully align with his vision of AGI, as it lacks the ability to learn autonomously. While OpenAI continues to pursue AGI as its ultimate objective, Altman suggested focusing on specific progress levels instead of a binary classification of whether a system qualifies as AGI. He remarked, 'We try now to use these different levels ... rather than the binary of, 'is it AGI or is it not?' as we get closer to our goals.' Looking ahead, Altman predicts that significant breakthroughs in various fields, including mathematics and scientific research, could emerge within the next two years. He concluded, 'There's so much exciting real-world stuff happening, I feel AGI is a bit of a distraction, promoted by those that need to keep raising astonishing amounts of funding.' Patience reiterated the sentiment, emphasizing that it is more beneficial to discuss specific AI capabilities rather than the vague notion of 'general' intelligence.