ChatGPT usage and adoption patterns at work

Launched just two and a half years ago, ChatGPT is used by workers across every industry, in every job function, and at companies of every size. Today, over a quarter of U.S. workers—and 45% of those with postgraduate degrees—report using ChatGPT for work.

Enterprise tech has always followed a familiar pattern: big upfront costs, long rollouts, and slow adoption before the payoff. ChatGPT broke that mold when people ported it from their personal lives into their jobs. They didn’t need months of training or complicated onboarding; they just started using it to get meaningful work done.

Already, we see clear signals. Everyone from scientists to marketers to operators is folding ChatGPT into everyday work. From debugging code to brainstorming campaigns, it’s becoming the first step in core workflows.

This report shares new data from our own analysis, combined with peer-reviewed sources, about who’s using ChatGPT at work, how people are putting it to use, and the ways it’s taking root inside organizations.

This report combines findings from independent third party industry-wide studies with analysis done by OpenAI on usage of ChatGPT and ChatGPT Enterprise. All analyses done by OpenAI in this report were performed on anonymized or aggregated usage data. OpenAI did not review any user or customer content (including model input or output), and did not analyze any identifiable data. All analysis of usage trends was conducted using automated content classifiers. Where the report references specific ChatGPT prompts, those ChatGPT prompts are fully synthetic examples, and not actual user or customer prompts.

When ChatGPT was released in November 2022, it mostly targeted a small group of AI researchers and enthusiasts. But within months, it had 100 million weekly active users, and today has over 700 million weekly active users, making it one of the world’s most visited websites. 

Widespread personal use rapidly spread to the workplace. As the statistics show, consumer adoption is very likely advancing AI at work. 

This is a path we've often seen before: software that gains traction with consumers makes its way into the workplace, often driven most heavily by younger employees. ChatGPT is following that same pattern, reflected in its rapid growth in weekly active users, high penetration with workers under 30, and frequent-often daily-use.

In just a few years, AI in the workplace has gone from niche to mainstream. The numbers tell the story:

AI adoption isn’t unfolding evenly across the economy. Workers in some industries have moved quickly to embed ChatGPT into their operations, while others are proceeding more slowly. 

By looking at which sectors are embracing the tool fastest, we can see both the near-term opportunities and the areas where adoption may take longer to gain traction. 

Certain industries are adopting ChatGPT at higher-than-expected rates. IT and finance lead the way, which makes sense given the tool’s strengths in coding, analysis, and information-heavy work. Manufacturing adoption points to a broader digital transformation: factories using AI for process automation, predictive maintenance, and supply chain optimization. Early investments in industrial AI may be paving the way for widespread ChatGPT use among engineers, analysts, and operations managers.

Other industries lag behind. Retail, construction, transportation, wholesale trade, and agriculture all show significantly lower adoption. In most cases, this tracks with their smaller share of knowledge workers, where the need for AI tools is less immediate.  

Healthcare is a special case. Despite being one of the largest and most data-intensive sectors, adoption has been slower. Strict privacy and compliance rules and risk-averse organizational cultures may be factors. Still, we’re starting to see growth in targeted areas like clinical documentation and administrative workflows, suggesting healthcare could soon become a hotbed of AI adoption.

Adoption patterns vary across departments, but a few themes stand out. In the first three months, four categories dominate usage: writing, research, programming, and analysis. Together, they account for the majority of messages sent. This variety highlights the flexibility of ChatGPT; teams turn to it to draft communications, gather and synthesize information, write code, and interpret data.

Technical teams are among the heaviest users, with analytics, engineering, and IT roles making up a large percentage of early usage. Programming is the top task, especially for engineering roles, but users also request a substantial amount of research and documentation help. This suggests ChatGPT is being used nearly as much for planning as for coding. 

IT teams lean most heavily on research and troubleshooting, often using ChatGPT as an information resource before moving into automation.

Example prompt for coding

Note: the above synthetic prompt is an example written specifically for this report solely for illustrative purposes

People in go-to-market roles, including marketing, communications, sales, and customer experience, are also major adopters. These functions rely on ChatGPT primarily for writing, research, creative ideation, and media generation. 

Across functions, the early usage pattern is consistent: AI is augmenting expertise, not replacing it. Engineers are iterating on prompts to debug code and generate unit tests. Analysts are using chain-of-thought prompting to clean and interpret datasets. Customer support teams are drafting thoughtful, brand-aligned responses. The common thread is that ChatGPT is extending the reach of specialized skills and becoming a partner in core workflows.

Source: Aggregated ChatGPT Enterprise department data collected during onboarding; automated
content classifiers

Interestingly, coding is spreading beyond engineering. Designers may be leaning on programming for front-end prototyping and snippet help-and they use ChatGPT for coding at a much higher rate than finance and sales. Project managers combine writing, media generation, coding, and data analysis—acting as the glue across teams. But product, operations, marketing, finance, and HR all use ChatGPT for coding to some extent. 

We see this trend validated in a study done by Boston University and BCG, which examined the impact of ChatGPT on the technical competency of BCG consultants. The study found that consultants armed with and trained on ChatGPT score 49, 20, and 18 percentage points higher than those in the control group on the three technical tasks, and performed close to the level of real BCG data scientists on two of the three tasks.

Good writing is no longer a specialist function reserved for content teams. With ChatGPT, anyone can turn notes into clean copy and iterate quickly. Meetings, memos, and customer messages become clearer and more inclusive because everyone can express their ideas well, not just trained communicators. AI is becoming the front door for routine communication and coordination, compressing drafting, tone calibration, and versioning into a single pass.

Design teams stand out for their use of media generation, relying on it 2–4x more than other groups. The heavy use in these functions for core work tasks highlights an emerging role for ChatGPT beyond text. 

All go-to-market teams use ChatGPT most for writing, research and media generation tasks, but in different ways. Here are some sample prompts indicating the type of queries we're seeing:

Note: the above synthetic prompts are examples written specifically for this report solely for illustrative purposes

Early data shows a consistent trend: most departments rely on the core tools in ChatGPT, including search, data analysis, file uploads, retrieval, and canvas. Adoption of more advanced features—such as reasoning models, deep research, projects, and custom instructions—are higher among power users, including R&D teams. The result for many employees is that ChatGPT is woven into daily workflows mainly through accessible, low-friction tasks rather than specialized use cases.

Technical functions stand out as the exception. Analytics, engineering, IT, and research roles are much heavier users of advanced capabilities. Their work often demands multi-step reasoning, large-scale data synthesis, or complex problem-solving. Engineers prompt for code generation or debugging; analysts use deep research to interpret datasets; and IT professionals query knowledge bases to resolve tickets and troubleshoot systems. Higher-powered tools naturally align with technical tasks that are structured, data-heavy, and decision-driven.

Different technical teams also lean into distinct features. IT teams are more likely to use retrieval and search, treating ChatGPT as a knowledge companion for quick answers to configuration or policy questions. Engineering teams show stronger use of GPTs, programming tools, and data analysis, reflecting their more code-centric workflows. This divergence underscores that adoption depends not only on technical fluency but also on the type of work and context within each department.

Two opportunities emerge from this data. First, advanced features remain underused, even where they could deliver broad impact. Barriers may include discoverability, awareness of use cases, or the setup required to use them.

Second, early champions in analytics, IT, legal, and engineering are already pushing into more complex workflows. As enablement programs expand and product improvements lower the barrier to entry, adoption will likely shift from core daily tasks toward deeper reasoning and collaborative workflows that reshape decision-making across the enterprise. 

ChatGPT is already making workers more productive in measurable ways. Internal benchmarks show meaningful increases in productivity, driven by employees who use it to write and communicate faster, research more effectively, and reduce the effort required for repetitive tasks. Most companies are still in the early stages of adoption, but we’re beginning to see organizations embed ChatGPT at the departmental level to make entire processes more efficient. 

Unlike traditional enterprise software, which spreads through top-down rollouts after long decision cycles and training programs, ChatGPT entered the workplace from the bottom up. Employees and small teams brought it in on their own, experimented with workflows, and demonstrated value before companies formalized procurement. This grassroots pattern has made it the fastest adopted enterprise technology in recent history. 

That dynamic is now shifting. New capabilities, from autonomous agents to advanced coding support to decision-assist tools, are expanding the role of ChatGPT beyond personal productivity. It’s becoming a platform for entire workflows. Executives use it to shape strategy, engineers to design and debug systems, and customer support agents to evaluate complex solutions. Increasingly, ChatGPT functions as an operating system for daily work: a shared layer where decisions are made, problems are solved, and output scales.

Work has always evolved alongside technology. Not long ago, much of it centered on finding answers, drafting emails, and repeating solved problems. Increasingly, it’s shifting toward synthesis, creativity, and speed: work that’s improved by natural, intuitive interactions with AI.

In the years ahead, AI will embed itself into nearly every workflow. As this happens, employees will spend less time performing tasks and more time supervising and shaping AI output. The crossfunctional reach of ChatGPT means individuals will be able to take on tasks once spread across multiple departments. A product manager, for example, might use it to analyze customer feedback, test and refine a new feature, and draft the legal and marketing content needed to bring it to market.

Collaboration is moving from siloed documents and messages into shared, real-time workspaces where teams solve problems together. Features like memory are making the product more contextaware, giving employees a partner that remembers preferences, projects, and workflows unique 
 to them. And the ability to bring structured and unstructured data directly into ChatGPT is broadening its role as the central interface for enterprise knowledge, and GPT‑5 is accelerating this shift.

Crucially, early evidence suggests this shift not only makes workers more productive, but actually makes their work more enjoyable. It does this by shrinking time-consuming and lower-value tasks and enabling them to refocus time on meaningful, core work. In a six-month randomized field experiment across thousands of knowledge workers, access to AI cut weekly email time by 31%. Another study looked at software developers, finding AI coding tools enabled them to spend more time coding, more time on exploratory work, and less time on project-management. Together, these findings suggest that tools like ChatGPT may reduce busywork, freeing up time for more strategic, satisfying, and ultimately higher-value work.

The scale of this change echoes past technological revolutions. Electricity reshaped factory work, the internet redefined commerce and communication, and AI is now setting the stage for the next leap. The enterprises that adapt quickly and thoughtfully will capture the earliest and largest gains: faster decision cycles, productivity breakthroughs, and new opportunities across every function.