Sam Altman confirms OpenAI’s ambition to build artificial general intelligence
Sam Altman, CEO of OpenAI, outlined the company’s vision and progress toward artificial general intelligence (AGI) - though he remained cagey on its arrival.
He did so during a fireside chat at the AI Startup School in San Francisco, a recording of which emerged over the weekend.
Altman, a lifelong tech geek who began his career by hacking Stanford’s mainframe and launching his first company before age 20, made clear that AGI remains OpenAI’s primary goal.
AGI refers to machines capable of learning, reasoning, and performing a wide range of tasks at a human level. “We’re going for AGI,” Altman said at the start of the discussion, underscoring the urgency and focus of the organisation.
He highlighted OpenAI’s latest model, GPT-4o, which exhibits improved reasoning and understanding across complex information.
More significantly, Altman described GPT-5 as a forthcoming “multimodal supermodel” able to process and integrate text, images, and other data types simultaneously, moving closer to human-like general intelligence.
Altman emphasised that OpenAI’s approach is not simply to replicate existing AI applications but to identify and develop solutions that address current gaps. This targeted innovation aims to meet real-world needs rather than building AI for its own sake.
The conversation also addressed the role of AI-driven robotics. Altman envisions a future where robots undertake a broad spectrum of tasks at scale, from industrial settings to everyday life.
He cautioned, however, that this development must be carefully managed to address safety and ethical concerns.
Speaking candidly about leadership pressures, Altman acknowledged the limits of individual effort and stressed the value of collaborating with trusted colleagues on meaningful projects.
“At the end of the day, it’s about finding people you like working with and doing something that matters,” he noted.
Altman concluded by sharing his optimism about AI’s potential to revolutionise scientific research.
He pointed to early evidence that AI tools are doubling productivity for biologists and other researchers, and predicted that future AI systems will autonomously execute complex projects, accelerating discoveries beyond current human capacity.
Altman’s remarks offer a window into the rapidly evolving landscape of AI development, one where AGI is no longer a distant dream but an active pursuit.
His balanced perspective highlights both the immense opportunities and responsibilities that lie ahead as AI continues to advance.
