OpenAI Says Agentic AI Is Most Of EU Business AI—But What’s The Base?

🔍 Read the full analysis: OpenAI Says Agentic AI Is Most Of EU Business AI—But What’s The Base? on ThorstenMeyerAI.com

Before you orderOffer from Amazon

Get smart everyday buys delivered free with Prime

• Fast, free delivery on millions of items
• Prime Video, Amazon Music and more included
• Member-only deals all year

Start your free Prime trial →Free trial for eligible customers · Cancel anytime

As an affiliate, we earn on qualifying purchases.Hide

OpenAI said agentic tools generated 55% of output tokens in its ChatGPT Enterprise and ChatGPT Business accounts in the 27 European Union member states by Sept. 30, according to a report published Oct. 14. The figure measures text produced by OpenAI products, not the share of EU businesses or work tasks using agents, and offers a view of the company’s customers rather than the whole European economy.

In Europe’s Agentic Opportunity: Moving from Breadth to Depth with AI Agents, OpenAI Economic Research counts output from Codex, its coding agent, and ChatGPT Work, an agent mode launched in July that can work with files, browsers and company tools. Codex represented 34% of output tokens in the covered EU accounts and ChatGPT Work 21% by Sept. 30. Standard ChatGPT accounted for the remaining 45%.

OpenAI also reported that weekly active users of Codex or ChatGPT Work at businesses with 200 or fewer users rose 8.1 times between the first week of May and the week ending Oct. 5. The company said every EU member state recorded at least fivefold growth. Belgium, Italy and Luxembourg had the largest reported increases, at 12.1, 10.8 and 10 times, respectively. The report does not give absolute user counts for this measure.

The report compares firms by output tokens per active user. OpenAI says EU companies in the top 10% by that measure generated about four times as many tokens per user as typical EU firms in September. The top-performing firms outside the EU generated more than twice as many as the EU frontier group. OpenAI estimates EU companies make up about 10% of frontier firms worldwide, compared with about 65% from North America.

What the Token Share Can Show

The figures point to growing use of agent features within OpenAI’s EU business customer base, while also highlighting a gap between the company’s highest-usage firms in Europe and those outside the region. If the pattern extends beyond OpenAI customers, it could matter for how companies invest in software, staff skills and workplace processes. The report alone cannot establish that broader trend.

The 55% headline is a measure of where OpenAI’s generated tokens are being used, not a direct measure of adoption or results. Agent workflows can read files, plan, generate code or documents, and check or revise work. Those multi-step interactions may produce substantially more tokens than a short chat exchange. OpenAI’s global Enterprise Signals page itself describes tokens as an imperfect measure of business value.

That distinction matters for business leaders and policymakers. A rising token share may reflect deeper use of agents, the launch of a new tool, or the greater amount of computation an agent task requires. It does not establish how much time or money firms saved, how many tasks were completed successfully, or whether the technology improved productivity.

A New Agent Product Shapes the Trend

The report combines a company-wide policy message with data from OpenAI’s own commercial products. OpenAI argues Europe should move from broad AI adoption to deeper use of advanced tools, build workers’ skills and confidence, and make those capabilities more accessible and affordable. The findings are based on accounts using ChatGPT Enterprise and ChatGPT Business, not a survey of all European firms.

Timing affects the growth figures. Codex accounted for roughly one-fifth of EU output tokens in May, according to the report’s chart; it reached 34% by Sept. 30. ChatGPT Work launched in July and accounted for 21% by the end of September. OpenAI says a major part of the small-business user growth followed that launch. The May-to-October comparison therefore spans a period before the new product existed.

Reported agentic output shares varied widely by country, from 85% in Lithuania and 81% in Cyprus to 22% in Slovenia and 26% in Austria. Germany was at 51%, ranking 17th among the 27 countries. OpenAI cautions that some smaller countries have relatively high agentic use among larger firms. Without country-level account and user counts, these figures should be read as descriptions of OpenAI’s customer mix, not national rankings of AI readiness.

What the Report Does Not Measure

The report does not establish what share of all EU businesses use agents, or what proportion of workplace tasks involve them. Its scope is OpenAI’s Enterprise and Business accounts in the EU. It also does not provide the absolute number of small-business users behind the reported 8.1-times increase, making it hard to judge the size of the starting base.

Country figures come without published account or user sample sizes in the source material. Their variation may reflect differences in customer mix, company size or product use as well as differences in adoption. The data do not show whether an agent completed a task successfully, how much work it replaced or assisted, or whether it generated measurable business value.

OpenAI’s report is also a vendor’s account of its own products. The available figures do not independently verify the trends across other AI providers or the wider European economy. It remains unclear how much of the increase reflects sustained organizational changes versus the introduction of ChatGPT Work and its early uptake.

Evidence Needed Beyond OpenAI Accounts

The next step for readers assessing the claims is to compare them with independent research and data from other providers, as those become available. More detailed reporting from OpenAI—especially absolute user counts, country-level sample sizes and evidence about completed tasks or outcomes—would help clarify how representative the reported growth is.

OpenAI has framed deeper business use, workforce skills and affordability as Europe’s opportunity, but the report does not set out a measured timetable for achieving those goals. It also does not announce a specific next data release. For now, the confirmed development is a substantial agent share of output in OpenAI’s EU business accounts; its reach and economic effects beyond those accounts remain open questions.

Key Questions

What does OpenAI mean by “55% of output”?

It means Codex and ChatGPT Work generated 55% of the output tokens in the covered ChatGPT Enterprise and Business accounts in the EU by Sept. 30, 2026. It is not a claim that agents handled 55% of users’ tasks or that 55% of EU firms use them.

Which products did OpenAI count as agentic?

The report counts Codex, OpenAI’s coding agent, and ChatGPT Work, an agent mode launched in July 2026 that can work with files, browsers and company tools.

Did small-business agent use rise 8.1 times across Europe?

OpenAI reported an 8.1-times increase in weekly active users of Codex or ChatGPT Work at businesses with 200 or fewer users, comparing the first week of May with the week ending Oct. 5. The report does not give the absolute number of users, and the period includes the July launch of ChatGPT Work.

Does the report prove that European businesses are less productive with AI?

No. OpenAI compared output tokens per active user among groups of firms, but tokens are not a direct measure of productivity or business value. The report does not establish the financial or operational results of the firms it tracks.

Source: ThorstenMeyerAI.com