ChatGPT prolongs work hours, a study finds

After the introduction of ChatGPT, workers most exposed to generative artificial intelligence (genAI) worked longer hours and had less leisure time without screens.

A recent study published in Social Science Research Network sheds light on how the introduction of ChatGPT, a generative AI chatbot, caught workers off guard and led to working longer hours.

The researchers used data from the American Time Use Survey, conducted annually, including 24,000 respondents on average. They use detailed 24-hour diaries at 15-minute intervals, allowing them to log in time spent on work and leisure.

The study’s data span from 2004 to 2023 and capture changes in work hours after the introduction of ChatGPT in late 2022.

The study found that workers in fields with the highest exposure to generative AI worked about 3.15 hours more per week since the chatbot was introduced.

Remote workers and those in roles with higher exposure to AI monitoring technologies, particularly those aiming at direction and evaluation, experienced greater increases in work hours. However, self-employed individuals didn’t experience the same effect.

The study looked into the period between 2022 and 2023, so it cannot be ruled out that other chatbots or later versions of ChatGPT would yield different results.

Decrease in screen-free leisure time

Spending more time on work resulted in reduced leisure time, about 3.20 hours per week, according to the study. Non-screen-based activities like relaxing, socializing, and traveling were affected the most.

“This shift indicates that workers in AI-exposed occupations not only work more but also reallocate their leisure time toward screen-based activities, which are generally more passive and less restorative,” the authors wrote.

Have thoughts about this topic? Others do, too. Join them in the discussion.

Firms benefited from the AI exposure, as they had higher returns on assets and saw increased labor productivity, the study suggests.

However, employees don’t necessarily reap the same benefits. Based on Glassdoor reviews, they reported lower satisfaction, particularly regarding work-life balance, even though AI exposure was associated with higher compensation.

AI’s impact on productivity is questionable

Despite the widespread use of the technology – 21% of US workers say at least some of their work is done with AI – there’s no clear consensus on how it benefits both companies and employees.

An overwhelming majority (95%) of organizations are getting zero return from integrated AI pilots, according to a viral report published in July. However, the report suggests that AI enhances individual productivity rather than financial performance.

A paper from the Federal Reserve Bank found that 77% of generative AI users said the technology saved them at least two hours per week, with more frequent users reporting greater time savings.

When the research considered all workers, including those who don’t use genAI, the technology was found to save only 1.4% of total hours.

A 2025 study published in Nature suggests that using genAI boosts immediate task performance, but can lead to reduced employee motivation long-term.

Discussions about genAI’s ability to improve productivity are often accompanied by employees’ concerns over their future, with junior roles being most likely to be replaced by AI.

According to a Goldman Sachs report, broad AI adoption could displace 6-7% of the US workforce. However, the impact will likely be transitory, meaning the technology will create new job opportunities.

Unlock more exclusive Cybernews content on YouTube.