Don’t be taken by the hype: MIT economist says AI likely to impact just 5% of jobs

Nobel Laureate and MIT economist Daron Acemoglu believes that the excitement around artificial intelligence may be overblown, arguing that current AI tools are likely to impact only about 5% of jobs and contribute around 1% to global GDP.

“It’s hugely uncertain and these are just guesses. I think it’s very difficult to know because it’s a very rapidly changing technology,” Acemoglu said. “But the basis of my prediction, uncertain though it may be, still remains.”

Speaking in a detailed interview with MIT, Acemoglu said AI has not yet demonstrated transformative capabilities in core production areas. “The industry has not produced applications that are critical for the production process or for generating new goods and services that are gonna be hugely valuable,” he said. In contrast, “from the very early days of the internet… it was clear how the internet was gonna change everything.”

He questioned the belief that general-purpose AI (AGI) will soon overhaul cognitive work. “If you’re a CEO, a CFO, an entertainer, a professor, a construction worker or a blue collar worker, I think those things are beyond what AI can perform,” he said.

The economist added that the technologies currently excel in predictable environments that require limited judgment, such as IT security, basic accounting, or routine software engineering. These conditions, he said, are not reflective of most real-world jobs.

“So when you do that calculation, you end up with about 20% or so of the economy that is either at the crosshairs of AI to be automated or could be majorly boosted by AI input… it’s not gonna be profitable to do them,” he said. “That’s how I arrived at the 5% number.”

He also drew a distinction between “easy to learn” and “hard to learn” tasks. “No task that we perform in reality is just recounting already established knowledge or playing a parlor game,” Acemoglu said. “They involve interactions… based on tacit knowledge or matching contextual understanding with the specific task at hand.”

Current models, he explained, can only mimic human decision-makers using existing data. “But if you do that, you’re not gonna get much better than the human decision makers,” he said.

Acemoglu also warned against over-optimism about AGI. “I don’t expect any occupation that we have today to have been eliminated in five or 10 years’ time,” he said. “If you are an AGI believer… you must have in your mind a list of occupations that will completely disappear.”

He added that the current approach to AI development is not “pro-human” and urged a shift toward systems that collaborate with workers rather than replace them. “The best possible way is much more pro-human,” Acemoglu said. “It requires a bigger celebration of the places where AI is better than humans, and the places where humans are better than AI.”

Looking ahead, he urged business leaders to focus not on cost-cutting but on creating new products and services. “Don’t be taken by the hype. I think the hype is an enemy of business success. Instead, think where my most important resource, which is your human resource, can be better deployed,” he said.

The International Monetary Fund (IMF) has warned that AI could affect nearly 40% of jobs globally, with advanced economies likely to face higher exposure due to the greater prevalence of white-collar roles. In a report released in January 2024, the IMF noted that while AI has the potential to boost productivity and economic growth, it also poses significant risks of disruption, particularly in sectors where tasks are routine or cognitive in nature.