Generative AI is moving into everyday work faster than earlier waves of digital technology did, according to a new U.S. survey. Tools like ChatGPT are already being used across a wide range of tasks, from writing and programming to customer support and administrative work.
The finding is striking because the comparison is with personal computers and the Internet, two technologies that reshaped work over time. The survey suggests that generative AI has reached meaningful usage levels in two years, while PCs and the Internet took more than five years to reach similar rates.
A faster adoption curve
The study, titled The Rapid Adoption of Generative AI, is based on two survey waves. Researchers first ran a pilot survey in June 2024 with 2,551 responses, then followed with a full survey in August 2024 with 5,014 responses.
The data came from the Real-Time Population Survey (RPS), described as a nationwide labor market survey of U.S. adults aged 18 to 64. That matters because the results focus on how working-age adults are using the technology, not just how the tech industry talks about it.
The headline result is simple: generative AI adoption has advanced more quickly than the adoption of PCs and the Internet. The source also notes an important caveat. PCs and the Internet came with higher costs and required more technical infrastructure, so the comparison is useful but not perfectly direct.
That caveat does not erase the broader signal. Generative AI tools can spread through existing devices and online services, which helps explain why people can begin experimenting with them without the same kind of hardware or infrastructure barrier that shaped earlier technology waves.
Workplace use is already broad
The workplace numbers show how quickly generative AI has entered professional routines. The survey found that 28% of employees use generative AI for work. Nearly a quarter use it weekly, and about one in nine use it daily.
For context, the source compares that with the early history of workplace PC adoption. After the 1981 launch of the IBM PC, it took three years for workplace PC use to reach 25%.
Generative AI is also not confined to one narrow group of workers. The survey found high use in computer and managerial roles, where nearly half of workers reported using generative AI. But the technology is also present across the broader workforce, with one in five workers overall reporting GenAI use.
The reported work uses include:
- Writing
- Data interpretation
- Administrative work
- Programming
- Customer support
That mix helps explain why adoption can spread across industries and skill levels. A tool that can assist with text, code, analysis, and routine office work has more entry points than a tool built for a single specialist workflow.
Productivity gains remain limited for now
Fast adoption does not automatically mean a large economy-wide productivity jump. The researchers estimate that generative AI currently supports between 0.5 and 3.5 percent of total work hours.
Using productivity gains from small-scale experiments, they calculate that this could raise labor productivity by 0.1 to 0.9 percent. That is a measurable effect, but still modest compared with the scale of attention around generative AI.
The source also notes that a Microsoft study suggests significantly higher productivity gains. At the same time, it cautions that the result is likely to depend heavily on the task at hand.
This is an important distinction. Generative AI may be very useful for some work while having less impact on other tasks. The survey points to broad adoption, but the productivity story is more uneven and still developing.
What the survey really shows
The strongest takeaway is not that generative AI has already transformed all work. It is that the technology has entered work routines at unusual speed, and it is being used by more than just early technical adopters.
Compared with PCs and the Internet, generative AI appears to have crossed early usage thresholds faster. But the source is careful about why the comparison has limits: older technologies required more spending, more setup, and more infrastructure.
For readers tracking the future of work, the useful signal is the gap between adoption and proven productivity impact. Generative AI use is spreading quickly. The larger question is how much of that use turns into durable productivity gains, and which tasks benefit most.
For now, the survey supports a balanced view. Generative AI adoption is moving faster than earlier transformative technologies, workplace use is already widespread, and the measurable productivity effect remains modest based on the estimates provided.