Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
The system that will win enterprise AI
As we start Q2, I want us to start where we always should: with our customers. I’ve spent time with leaders at our largest organizations, our most impactful startups, and major enterprise companies. The message is clear. People are passionate about what we’re building, and want deeper insight into our roadmap so they can plan with confidence and stay ahead of the market.
Enterprise AI is entering a more mature stage. Initial capabilities are still important, but they are no longer sufficient. Customers want relevance: how well AI is integrated into workflows, knowledge, controls and daily operations, and how effectively it is deployed, trusted and improved over time. They want a system they can trust and build on.
We’re building this system: best-in-class business models, a platform for agents, deep integration with business context, and the ability to deploy and improve at scale. Clients validate this trend in the clearest way possible. Multi-year, multi-product and nine-figure deals are on the rise, and existing clients are expanding as we consolidate our capabilities across more of their organizations.
I’m very proud of the way this team is turning out. We earn trust through the depth, quality and care we bring to the work. The opportunity before us is enormous, and the biggest obstacle facing us now is not demand. It’s ability. That’s why talent remains a top priority in Q2. We will continue to hire intentionally, maintain a high standard, and continue to build a team that matches the excellence our clients expect from us and we expect from each other.
We have everything we need to expand our progress from here. We have the account. We have the products. We have customer withdrawals. This is the moment to step up and state clearly and confidently that OpenAI is the platform that organizations should trust to build, deploy, and scale.
Here are five customer-driven priorities that I want us to focus on.
1. Win the model layer to work
Companies buy business results. They pay for templates that help employees write faster, analyze better, program more productively, support customers more effectively, and make higher-quality decisions. They pay for higher revenue per employee, faster cycle times, lower support costs, and better execution.
Spud is an important step in the intelligence foundation for the next generation of business. Early feedback from our customers is very positive. Spud is not only our smartest model yet, but it also delivers everything that matters for high-value professional work: stronger thinking, better understanding of intents and dependencies, better follow-through, and more reliable output in production.
Better model performance moves the rest of the stack up. Spud will make all our major products that much better. It expands the workflows we can have and gives customers another reason to unite around us. This is our iterative deployment strategy in practice: pushing the boundaries, deploying them in real products, learning from real usage, and integrating these lessons into better systems on the road to superior application.
Our computing advantage positions us to make continuous leaps in capabilities. Customers already feel it in terms of the real product: higher code limits, lower latency, and more reliable execution of complex workflows. Every step forward in computing allows us to train stronger models, meet more demand, and lower the cost per unit of intelligence. This is permanent commercial leverage.
2. Earn the agent platform tier
The market has moved from claims to agents. This transformation represents a huge opportunity for us.
Customers want systems that can think, tool, work across workflows, and perform reliably within real business environments. This means coordination, control, observability, security, integration and governance.
Frontier allows us to have the platform layer. We need to position Frontier as the default platform for enterprise customers — the core intelligence layer that organizations use to build, deploy, manage, and scale systems.
This is where our advantage can be doubled. Boundary relationships intelligence directly model client performance. As our models improve, the platform becomes more valuable. As the platform is consolidated, switching costs rise. As customers manage more workflows through the system, OpenAI becomes harder to replace and more central to how work gets done.
This is how we move from product vendor to operating infrastructure.
3. Expand your market through Amazon
Our partnership with Microsoft has been key to our success. But it has also limited our ability to meet companies where they are – for many, this is Bedrock.
Since we announced the partnership at the end of February, the demand from our customers for this offering has been frankly amazing. We are firing on all cylinders to establish it as a scaled distribution channel.
Amazon Stateful Runtime Environment is important because it expands the reach and upgrades the product surface at the same time. By enabling memory, context, and continuity across interactions, we move beyond access to stateless models toward systems that can operate reliably over time and across complex business processes.
This will expand our market in three ways: 1. It reduces adoption friction for AWS native customers. 2: It strengthens our position with regulated, security-sensitive buyers by working within their existing AWS environment and governance model. 3. It also integrates our platform from model access to production runtime for long-running multi-step agents.
4. Sell the complete set of original AI
Customers want a platform, not specific solutions. Here’s what we have: ChatGPT for Work is the front door to knowledge work. Codex is a system for developing software and agents. An Application Programming Interface (API) is the intelligence engine embedded within customers’ products and workflows. Frontier is the agent platform. Amazon Runtime extends our reach to premium production and execution.
This breadth is a huge strategic advantage because customers don’t all start in the same place. Some start with employees. Some start with the developers. Some start with internal systems. Some start with outside products. Our mission is to meet them where they come in and then expand them across the full range.
This is the flywheel around which we must build: better models lead to more usage, more usage leads to deeper integration, deeper integration drives adoption of multiple products, and adoption of multiple products makes us harder to replace.
We must stop thinking as a company with separate product lines. We must think as a platform company with multiple entry points and one integrated enterprise offering.
5. Private publishing
The biggest bottleneck in enterprise AI is no longer whether the technology works or not. It is both
Companies can deploy it successfully and at scale.
DeployCo gives us the opportunity to turn product demand into repeatable enterprise transformation. It will serve as a deployment engine that helps companies prove value faster, reduce risk, and scale adoption across the enterprise.
This can become a force multiplier for everything else we build. It helps customers move faster. It sharpens our feedback loops. It highlights repeatable publishing patterns. It improves product, sales and customer success simultaneously. Together with our partners at Frontier Alliance, it gives us a serious path to scale implementation across the market.
Companies that win with enterprise AI won’t just have the best models. They will be best able to deploy these models in real workflows, within real organizations, with real, measurable value. We have to be the best in the world at this.
A note about the competitive landscape
The market is as competitive as you’ve ever seen. I think that’s a good thing in the end. This means that the opportunity is huge and important. However, there is no doubt that they can be noisy, volatile and distracted at times. Competition inspires us and will make us all better and most importantly our customers will feel the benefit. To this point, as you haven’t heard me say many times, the number one focus should be spending time with our customers. When we spend time with our clients, listen to their problems and ambitions, and focus on how we can invest in them and help them, everything else calms down and comes into focus.
With all that said, here are some things worth keeping in mind, especially regarding anthropology.
● Their story is built on fear, limitations, and the idea that a small group of elites should control artificial intelligence. Our positive message will win out over time: build strong systems, put the right safeguards in place, expand reach, and help people do more.
● It shows their strategic mistake of not getting enough compute into the product. Customers feel this through throttling, poor availability, and a less reliable experience. We saw the exponential calculation curve earlier, acted on it faster, and now have a real structural advantage.
● Their focus on programming gave them an early opportunity. But you don’t want to be a one-product company in the platform wars. As AI spreads beyond developers to every team, workflow, and industry, this narrowness could become a real barrier.
● The advertised operating rate has been inflated. They use accounting treatment that makes revenue appear larger than it really is, including total revenue share with Amazon and Google. Our analysis shows that this overstates its run rate by about $8 billion (at the current price of $30). We reported on Microsoft’s revenue sharing network, which is more consistent with the standards we would adhere to as a public company.
Let’s go build
Finally, one of the best things about the work we do is the people we do it with. I am very proud of this company and our team. It is an honor to work with all of you and to be alive in this moment at the Center for the Future. It allows everyone to stay focused, work as a team, and work at the highest level of excellence and in the same direction.
The market is ours to win, so let’s execute accordingly.