OpenAI Wins Enterprise Deal Amid Google Competition

OpenAI Reports Significant Growth in Enterprise AI Adoption

New data released by OpenAI on Monday reveals a substantial increase in the utilization of its AI tools within the enterprise sector over the past year. Message volume on ChatGPT has experienced an eightfold rise since November 2024, and employees are reporting time savings of up to one hour per day.

Competitive Pressures and Strategic Positioning

This announcement arrives shortly after CEO Sam Altman distributed an internal memo, categorized as “code red,” addressing the competitive challenges posed by Google. The timing highlights OpenAI’s efforts to solidify its position as a leading provider of AI solutions for businesses, despite facing increasing pressures.

Currently, approximately 36% of U.S. companies utilize ChatGPT Enterprise, compared to 14.3% who have adopted Anthropic’s offerings, according to the Ramp AI Index. However, the majority of OpenAI’s revenue continues to originate from consumer subscriptions, a segment now potentially threatened by Google’s Gemini.

Infrastructure Investment and Economic Impact

OpenAI has pledged $1.4 trillion towards infrastructure development in the coming years, making expansion within the enterprise market crucial to its business model. The company’s chief economist, Ronnie Chatterji, emphasized the importance of firm-level adoption for realizing the full economic benefits of transformative technologies.

“While consumer adoption is important from an economic growth perspective, historically transformative technologies, such as the steam engine, truly demonstrate their impact when businesses integrate and scale their use,” Chatterji stated during a briefing.

Increased Complexity in AI Usage

The latest findings from OpenAI indicate that adoption among larger organizations is not only expanding but also becoming more deeply embedded within existing workflows. Organizations leveraging OpenAI’s API are now consuming 320 times more “reasoning tokens” than they were a year ago.

This surge suggests that companies are increasingly employing AI for more intricate problem-solving tasks. Alternatively, it could indicate extensive experimentation with the technology, potentially leading to high token consumption without guaranteed long-term value.

Sustainability Concerns and Resource Allocation

The increased consumption of reasoning tokens, which directly correlates with higher energy usage, may present sustainability challenges and increased costs for businesses. TechCrunch has requested information from OpenAI regarding enterprise budget allocation for AI and the long-term viability of this growth trajectory.

Custom GPTs and Workflow Automation

Beyond overall usage statistics, OpenAI is observing shifts in how companies are implementing its tools. The utilization of custom GPTs – used by companies to encapsulate institutional knowledge or automate processes – has increased nineteenfold this year.

These custom GPTs now account for 20% of all messages within enterprise environments. OpenAI cited BBVA, a digital bank, as an example, noting its regular use of over 4,000 custom GPTs.

Brad Lightcap, OpenAI’s chief operating officer, commented, “This demonstrates the extent to which users are able to adapt this powerful technology to their specific needs.”

Time Savings and Capability Enhancement

According to OpenAI, these integrations are resulting in significant time savings for employees. Participants in surveys reported saving between 40 and 60 minutes daily through the use of OpenAI’s enterprise products.

However, this figure may not account for the time invested in learning the systems, crafting effective prompts, or correcting inaccuracies in AI-generated output.

Expanding Skillsets and Potential Security Risks

The report also revealed that enterprise workers are increasingly using AI tools to broaden their skillsets. Three-quarters of those surveyed indicated that AI enables them to perform tasks, including technical ones, that were previously beyond their capabilities.

OpenAI noted a 36% increase in coding-related messages originating from teams outside of traditional engineering, IT, and research departments. While OpenAI emphasizes the democratization of skills, increased “vibe coding” could potentially introduce security vulnerabilities and other flaws.

Lightcap addressed this concern by highlighting OpenAI’s recent release of Aardvark, an agentic security researcher currently in private beta, as a potential solution for identifying bugs, vulnerabilities, and exploits.

Adoption of Advanced Features and Mindset Shifts

OpenAI’s report also indicated that even frequent ChatGPT Enterprise users are not fully utilizing the most advanced features available, such as data analysis, reasoning, and search. Lightcap suggested that complete adoption requires a fundamental shift in mindset and deeper integration with enterprise data and processes.

He explained that the adoption of these advanced features will take time as companies adapt their workflows to fully leverage the technology’s potential.

The AI Adoption Divide

Lightcap and Chatterji also highlighted a “growing divide in AI adoption,” with “frontier” workers utilizing tools more frequently and realizing greater time savings compared to “laggards.”

“Some firms view these systems simply as software – a purchase to be made and distributed to teams. Others are embracing AI more fundamentally, almost as an operating system, essentially re-platforming their operations,” Lightcap explained.

OpenAI’s leadership views this disparity as an opportunity for laggards to accelerate their adoption. However, for workers whose tasks are being replicated by AI systems, “catching up” may feel more like a race against time.