Early adopters are in a much better position to win – “just do it,” says Sam Altman Today is AI trainee but soon it will be an experienced worker who is able to tackle complex problems The next step is AGI, but does that mean that accelerated scientific discoveries, self -improvement or something else?
Openai CEO Sam Altman has offered his honest insight into the evolving AI landscape comparing previous, current and future trends to pointing companies in the right direction.
“I think, just do it,” said Altman in an interview with Snowflake CEO Sridhar Ramaswamy at the company’s Summit 2025 event, highlighting that companies waiting for the next AI breakthrough risk to be left behind.
“When things change rapidly, they have companies that have the fastest adoption speed … win,” he added.
The AGI definition changes
As reflected on how his guidance has changed in the past year, Altman noticed a major shift in AIS capabilities just in this short period.
“We hit a real bending point for the use of these models,” he said, but “An interesting question is what we say differently next year.”
Looking ahead, Altman said he sees that companies adopt AI not only to automate routine processes, but to tackle greater challenges.
By 2026 he predicts that companies will “throw a ton computer” on “hugely important problem[s]”And” The models will be able to find out things that teams of people on their own cannot do. “
Altman compared today’s AI to “an intern who can work for a few hours,” but foresees a near future where AI looks like an experienced or senior worker. Openai -the founder predicts that “in some limited cases [and] In some small ways, “we could see agents who are able to discover new knowledge fairly soon, and the use cases are only set to rise with additional developments.
When asked about artificial general intelligence (AGI), Altman explained that people are very good as adapting their expectations and challenging static definitions of the expression. He threw his mind back five years and said that most people may have considered today’s chatgpt as AGI, but our expectations today are promoted.
“What matters is the progress you see year-over-year in the last five years should continue for at least the next five, probably far beyond that,” he added, suggesting that we are not near the top.
His preferred purpose of AGI would be to increase, perhaps “quadrupled” scientific discovery, but he also acknowledges that others see AGI as a system capable of self -improvement, potentially smooth consciousness.
The ending was asked a hypothetical question – what would he do if he had 1,000x more calculation?
“I want to ask it to work super hard on AI research, find out how to build much better models, and then ask the much better model what to do with all the calculation,” he said.
Altman offers us a glimpse of the not too distant future from Openai that “the models over the next year, two years, will be quite breathtaking … There is a lot of progress coming.”
You also like
