Virtual employees could join workforces this year and transform the way businesses operate, according to OpenAI CEO.
The first AI agents may begin working for organizations this year, Sam Altman wrote, as AI companies push for uses that generate returns on substantial investments in the technology.
Microsoft, the company’s biggest backer behind ChatGPT, has already announced the introduction of artificial intelligence agents (tools that can perform tasks autonomously) with blue-chip consultancy McKinsey among the early adopters.
“We believe that, in 2025, we may see the first AI agents ‘join the workforce’ and materially change the production of companies,” Altman wrote in a blog post published on monday.
OpenAI is reportedly planning to launch an AI agent codenamed “Operator” this month, after Microsoft announced its Copilot Studio product and rival Anthropic launched the Claude 3.5 Sonnet AI model, which can perform tasks on the computer. how to move the mouse cursor and write. text.
McKinsey, for example, is creating an agent to process new client inquiries by performing tasks such as scheduling follow-up meetings. The consultant has predicted that by 2030Activities that account for up to 30% of hours worked across the US economy could be automated.
Bloomberg reported that the Operator will use a computer to perform actions on behalf of a user, such as writing code or booking travel.
Last year, Microsoft’s AI chief Mustafa Suleyman indicated that the company is moving toward agents that can make purchasing decisions, saying he had seen “awesome demos” in which the agent trades independently, although there have also been “car crash moments” in development. However, an agent with these capabilities will emerge “in quarters, not years,” Suleyman said.
Before making the agent prediction, Altman also wrote on his blog that OpenAI knows how to build artificial general intelligence (AGI), a theoretical term he has referred to in the past as “AI systems that are generally smarter than humans.” ”.
“We are now confident that we know how to build AGI as we have traditionally understood it,” he wrote, adding that OpenAI was now directing its ambitions toward “superintelligence.”
“We love our current products, but we are here for a glorious future. “With superintelligence we can do anything else,” he wrote.
“Super-smart tools could greatly accelerate scientific discovery and innovation far beyond what we are capable of on our own, and in turn vastly increase abundance and prosperity.”
Altman also participated in a Q&A with Bloomberg published this weekend in which he predicted that Elon Musk will continue his feud with OpenAI this year, but will stop short of using his relationship with Donald Trump to harm the company.
after newsletter promotion
Altman said he expected the world’s richest person to maintain his legal battle with OpenAI, although he downplayed the possibility of being challenged to a cage fight with Musk, who has asked Meta’s Mark Zuckerberg for a mixed martial arts fight in 2023.
“I think he will do all kinds of bad things. “I think he will continue to sue us and withdraw lawsuits and file new lawsuits and whatever else,” Altman. he told Bloomberg.
“He hasn’t challenged me to a cage fight yet, but it turned out that I don’t think he was that serious with Zuck either… he says a lot of things, he starts them, he undoes them, he gets sued, he sues, he fights with the government, he is investigated by the government . “That’s just Elon being Elon.”
Musk dropped an initial lawsuit against OpenAI in June last year, but returned two months later with a new complaint that has been expanded to include Microsoftthe largest sponsor of OpenAI. The lawsuit accuses OpenAI of seeking profits over safety and “actively attempting to eliminate competitors.”
Musk and Altman have a conflicted history. The two co-founded OpenAI in 2015 before Musk left the company over an internal power struggle several years later. OpenAI was founded with the goal of creating “safe and beneficial” AGI.
Altman added that he did not expect Musk to use his influence within the incoming Trump administration to hinder competitors like OpenAI. Musk launched a new AI business, xAI, in 2023.
“Will he abuse his political power as co-chairman, or whatever he’s called now, to pick on a business competitor? I don’t think I’ll do that. I really don’t. It may turn out to be proven wrong,” he said.