Altman Sells Startup Optimism Days After Warning of ‘Sobering’ AI
Sam Altman (164news.com's guide to sam altman) told Axios that the next generation of AI models will be "sobering for everybody." Four days later, to the same reporter, he stated this is the best time in the world to start a company. Neither claim is false, and OpenAI needs both believed.
Background
September 8, 2026 – 10:53 am
Sam Altman speaking at TechCrunch Disrupt San Francisco in October 2019.
Credit: Steve Jennings / Getty Images for TechCrunch / CC BY 2.0 via Wikimedia Commons (cropped)
Altman conducted two interviews with Axios four days apart. In the first, he expressed concern about the next generation of AI models, describing them as "sobering for everybody." The second interview, published on Monday, promotes a different perspective.
The Idea Guy’s Revenge
Altman now embraces what he previously dismissed as absurd—the "idea guy" who has a concept but lacks coding skills. He attributes this shift to the advent of powerful AI models like GPT-6 Astra, which, according to Nvidia’s Jensen Huang, signifies the arrival of Artificial General Intelligence (AGI).
Altman’s Argument:
With Astra, anyone can now have a company boasting "genius-level employees" in various fields. He provides examples of small ventures that became viable due to these models, emphasizing the economic potential for thousands of users.
Contrasting Perspectives
On September 3rd, also to Axios, Altman offered a different viewpoint:
"The next generation of models are going to be sobering for everybody… We are just sailing in unknown waters."
He emphasized the need for caution and care, suggesting that OpenAI’s progress on alignment and safety would set a pacing principle for future releases.
Managing Audiences
The discrepancy between the interviews is not hypocrisy but rather a strategic approach to managing two distinct audiences. Public polling has turned against technology, and data center anger is becoming a political issue in the American midterms. Altman’s messages cater to both internal concerns at OpenAI and external public perception.