Enterprise AI 2025: OpenAI Report Shows 8x Growth
The OpenAI State of Enterprise AI Report 2025 provides the first comprehensive data on AI usage in enterprises. The results show: companies that systematically deploy AI achieve measurable productivity gains. At the same time, the gap between leaders and laggards is widening. Here you will learn what the numbers mean for your business.
The Current State: Enterprise AI Leaves the Pilot Phase
After three years of intensive development, Enterprise AI has reached a turning point. What started as an experiment is becoming core infrastructure. OpenAI now serves over 7 million ChatGPT Workplace seats, and usage intensity continues to rise. The question is no longer whether companies use AI, but how deeply they integrate AI into their processes.
— Ronnie Chatterji, Chief Economist, OpenAI
The data is based on anonymised, aggregated usage data from over 1 million business customers and a survey of 9,000 employees from nearly 100 companies. This makes the report one of the most comprehensive insights into actual AI usage in enterprises.
Four Key Findings from the OpenAI Report
The report identifies four central developments that define the current state of Enterprise AI. These findings are based on real usage data and show where AI is already delivering measurable value.
The Four Main Findings
- Scaling and Integration: Usage is growing not only in breadth but also in depth. 20% of all Enterprise messages already run through Custom GPTs or Projects.
- Measurable Productivity: 75% of surveyed employees report improved speed or quality of their work. Average time savings are 40-60 minutes per day.
- Global Growth: Adoption is accelerating worldwide. Germany is among the most active markets by message volume.
- Growing Gap: Frontier Workers send 6x more messages than the median. The differences between leaders and laggards are increasing.
Particularly notable is the development of Custom GPTs and Projects: weekly users of these configurable interfaces have increased 19-fold year-to-date. Companies like BBVA regularly use over 4,000 GPTs, showing that AI-driven workflows are increasingly being implemented as permanent tools embedded in daily operations.
Implications for the European Market
According to OpenAI, Germany ranks among the most active markets for ChatGPT business usage by message volume, with the UK also among the largest Enterprise markets outside the US. European businesses face particular requirements around data protection, compliance and worker participation that must be considered when introducing AI.
Regulatory Framework
Compliance Requirements for European Businesses
- GDPR: Processing personal data through AI systems requires clear legal bases and transparency towards data subjects.
- EU AI Act: From 2025, new requirements apply to AI systems, particularly for high-risk applications in HR, finance and customer service.
- Works Council Rights: In many EU countries, the introduction of AI tools is subject to worker participation and consultation requirements.
- NIS2 Directive: Cybersecurity requirements for AI systems, especially when processing sensitive business data.
Opportunities for European Businesses
European SMEs can build competitive advantages against larger competitors through AI adoption. The time savings of 40-60 minutes per day have a proportionally greater impact on smaller teams.
75% of users report being able to complete tasks they previously could not perform. This enables closing skills gaps without hiring additional staff.
European businesses are known for high quality standards. AI can help maintain these standards under increasing cost pressure by automating routine tasks.
Those who establish GDPR-compliant and EU AI Act-compliant AI processes early can use this as a differentiator against international competitors.
— Dr. Elena Bergmann, AI Strategy Consultant
Challenges in the European Context
The introduction of Enterprise AI in Europe comes with specific hurdles that go beyond technical aspects. Cultural factors, worker participation rights and high data protection requirements demand an adapted approach.
Success Factors for European Businesses
- Early involvement of works councils in planning AI projects to avoid resistance and build acceptance.
- Data Protection Impact Assessment before deploying AI tools that process personal data.
- Clear governance structures for handling AI-generated content and decisions.
- Training programmes for employees that cover both technical usage and critical evaluation of AI outputs.
The OpenAI data shows that companies with systematic change management and clear executive support achieve significantly better results. This applies particularly to European businesses, where additional coordination processes are required.
How Companies Are Using AI in Practice
The report shows in detail which use cases are most common in enterprises and where the greatest productivity gains are achieved. Usage patterns vary significantly by industry and function.
The most common API use case. Companies integrate AI-powered assistants directly into their products and internal systems to provide users with contextual help.
Automation of multi-step processes through AI agents. These can independently execute complex tasks, from data analysis to document creation.
Codex and similar tools are used for code generation, refactoring, testing and debugging. Weekly active users have doubled in six weeks.
Particularly in the finance sector an entry point, as support is a scalable cost centre with proven ROI. AI agents already resolve 53% of calls completely.
The data shows an interesting trend: coding-related messages outside of Engineering, IT and Research have increased by an average of 36% over the past six months. This means non-technical teams are increasingly able to take on technical tasks.
Measurable Results from the Report
The OpenAI Report provides concrete figures on the productivity gains companies achieve through AI deployment. This data comes from the survey of 9,000 employees and aggregated usage data.
75% of users can complete tasks with AI that they previously could not perform: programming support, data analysis, technical tool development and Custom GPT design.
Users who save more than 10 hours per week use 8x more credits than users with no time savings. They use multiple models and deploy AI for various task types.
85% of marketing teams report faster campaign execution, 75% of HR professionals improved employee engagement. Benefits are not limited to technical roles.
A BCG study shows: AI leaders achieved 1.7x revenue growth, 3.6x higher shareholder returns and 1.6x EBIT margin over three years compared to laggards.
Practical Examples from the Report
The OpenAI Report contains detailed case studies of companies that have successfully integrated AI into their processes. These examples show concretely how AI leads to measurable business outcomes.
Intercom uses the OpenAI Realtime API for Fin Voice. Latency decreased by 48%, 53% of calls are fully resolved by AI, and calls escalated to humans are completed 40% faster.
Mylow answers nearly 1 million questions per month. When customers engage with Mylow, the conversion rate more than doubles. Customer satisfaction increased by 200 basis points.
Career Scout helps job seekers find relevant jobs 7x faster. The hiring probability increases by 38%, and 84% of users rate the service as valuable.
A Legal AI Chatbot automates over 9,000 queries annually about signatory authority. This equals 3 full-time positions and delivers 26% of the Legal Services division's annual savings KPI.
— Moderna Case Study, OpenAI Report 2025
The Growing Gap: Leaders vs. Laggards
One of the most important findings of the report is the increasing divergence between companies and employees who use AI intensively and those who fall behind. This gap has concrete implications for productivity and competitiveness.
Frontier Workers (95th percentile) send 6x more messages than the median. For data analysis tasks, the difference is even 16x. For coding, the factor is 17x.
Frontier Firms generate 2x more messages per seat than the median and 7x more messages to GPTs. They systematically invest in infrastructure and operating models.
19% of monthly active users have never used data analysis, 14% never reasoning, 12% never search. Among daily users, these figures drop to 1-3%.
Users who work on about 7 different task types with AI report 5x more time savings than users with only 4 task types. Broad usage amplifies the effect.
The data clearly shows: the differences are not due to tool availability, but to usage intensity and breadth. Companies have the opportunity to adopt the patterns of Frontier Workers and Frontier Firms through systematic adoption.
What Successful Companies Do Differently
The report identifies five practices that leading companies consistently implement. These patterns distinguish organisations that use AI as core infrastructure from those that remain at superficial usage.
1. Deep System Integration Through Context
Leading companies activate connectors that give AI secure access to company data. This enables context-aware responses and automated actions. About a quarter of companies have not yet taken this step.
2. Workflow Standardisation and Reuse
They actively promote the creation, sharing and discovery of reusable solutions for common tasks. GPTs often drive this work, while the most sophisticated organisations embed API-powered assistants directly into core systems.
3. Executive Leadership and Sponsorship
They set clear mandates, secure resources, align teams and create space for experimentation. All of this enables deployment at scale.
Additional Success Factors
- Data Readiness and Evaluations: They codify institutional knowledge into machine-readable routines, build APIs for key data pipelines and run continuous evaluations to track model performance on real-world outcomes.
- Deliberate Change Management: They build structures that accelerate organisational learning, combining centralised governance and training with distributed enablement through embedded AI champions.
- Continuous Adaptation: OpenAI releases a new feature or capability roughly every three days. The primary constraints for organisations are no longer model performance or tooling, but organisational readiness.
Strategic Implications for 2025 and Beyond
The report outlines how Enterprise AI will develop in the next phase. The shift is from asking for outputs to delegating complex, multi-step workflows. For companies, this means a fundamental change in how work is organised.
Companies will use AI not only for productivity gains but discover new ways to serve customers and deliver value. Faster iteration, deeper personalisation and new experiences become possible.
Coding and analytical tasks are increasingly appearing outside traditional specialist roles. This expands what non-technical teams can achieve and changes job profiles.
Despite broad adoption, industry patterns remain different. Technology, Professional Services, Finance, Healthcare and Manufacturing each have their own focus areas and use cases.
Organisations that bring AI capabilities into market-facing workflows will use AI not just as a productivity tool, but as a durable engine for revenue growth and competitive advantage.
— OpenAI State of Enterprise AI 2025 Report
Conclusion: What You Can Do Now
The OpenAI State of Enterprise AI Report 2025 shows clearly: the question is no longer whether Enterprise AI works, but how deeply you integrate it into your organisation. The data demonstrates measurable productivity gains, but also a growing gap between leaders and laggards.
The Key Takeaways
- Usage intensity decides: Not the availability of tools, but the depth of integration determines success. Frontier Workers and Frontier Firms show what is possible.
- Broad usage amplifies the effect: Those who use AI for more task types achieve disproportionately more time savings. Seven task types bring 5x more returns than four.
- Organisational readiness is the bottleneck: The models are capable enough. What is missing is governance, change management and systematic integration.
- The time to act is now: Enterprise AI is still in an early phase. Companies that adopt the patterns of leaders now can catch up.
For European businesses, this means: the combination of high quality standards, strong compliance culture and systematic approach can be a competitive advantage when AI is introduced correctly. The data shows that the effort pays off: 40-60 minutes of time savings per day, new capabilities for employees and measurable business outcomes are achievable.