OpenAI CEO Sam Altman estimates that virtual workers could enter the workforce as early as 2025, forcing many to change their operations.
“I believe that by 2025, we could see AI agents, or virtual employees, enter the workforce for the first time and dramatically change the productivity of companies,” OpenAI CEO Sam Altman wrote in a blog post on January 6. “Putting great tools into the hands of people in incremental steps will lead to significant, increasingly widely distributed results.”
Agentic AI is a system designed to perform complex tasks independently, continuously learning from input and making decisions without human supervision. According to the Guardian, OpenAI is planning to launch an Agentic AI codenamed Operator this month. Bloomberg reported that Operator will perform actions on behalf of users, such as writing code and booking travel tickets.
In the post, Altman also said that OpenAI already knows how to build superintelligence AGI. “We are now confident that we know how to build AGI to the best of our knowledge,” he wrote. “OpenAI loves our current products, but we are here for the bright future. With AGI, we can do anything differently.”
The OpenAI head cited the example of superintelligent tools that could dramatically accelerate scientific discovery and innovation beyond humanity’s current capabilities, thereby “dramatically increasing abundance and prosperity.”
He described the time he was fired more than a year ago as “the fog of war.” “Being fired after just one phone call felt strange and inexplicable, like a dream gone wrong,” he wrote. “I was in a hotel room in Las Vegas, unable to explain or come up with a satisfactory answer to the question: Why did this happen?”
According to Altman, the whole event was a “massive failure of management by well-intentioned people,” including himself. “Looking back, I wish I had done things differently. I want to be a better, more thoughtful leader than I was a year ago,” he said.
