Companies that are early adopters of artificial intelligence are demonstrating a rapidly compounding advantage, now utilizing 3.5 times as much intelligence per worker as typical firms, an increase from a ratio of 2 to 1 just a year ago. This disparity is not simply about wider AI access; research indicates message volume explains only 36% of this advantage, with most of the difference stemming from more complex AI use. These leading firms are sending 16 times as many Codex messages per worker as their counterparts, suggesting a shift toward delegated workflows and agentic tools. According to a new B2B Signals report, “Typical firms are using AI to answer questions; frontier firms are using it to help execute complex work.” This suggests the focus is moving from simply having AI tools to strategically embedding them within operations.
Frontier Firms Demonstrate 3.5x Intelligence Per Worker
Frontier firms are now leveraging 3.5 times more intelligence per worker than their typical counterparts, a dramatic increase from a 2x advantage observed last year, signaling a compounding effect of artificial intelligence integration within leading businesses. This isn’t merely about wider access to AI tools; the disparity stems from how these organizations are utilizing the technology, embedding it more deeply into core workflows. This suggests a fundamental change in operational strategy, where AI is not just a support function but an integral component of complex processes. Cisco, for example, utilized Codex to reduce software build times by approximately 20%, saving over 1,500 engineering hours monthly and increasing defect resolution throughput by 10 to 15 times.
As Cisco’s team explained, the biggest gains came when they treated Codex as “part of the team.” The most significant advantage is observed in the adoption of advanced, agentic tools, particularly Codex, where frontier firms send 16 times as many messages per worker as typical firms. Similar patterns emerge with ChatGPT Agent, Apps in ChatGPT, Deep Research, and GPTs, indicating a proactive embrace of tools that facilitate coding, task delegation, contextual application, and complex research. Travelers Insurance exemplifies this approach with its AI Claim Assistant, projected to handle approximately 100,000 first notice of loss calls within its first year, demonstrating how AI can be integrated directly into products and internal systems. Ultimately, the focus is shifting from simply measuring AI access to understanding where it’s deepening workflows and reshaping team operations.
Codex & Agentic Workflows Drive Leading AI Advantage
The initial surge in artificial intelligence adoption is now giving way to a more nuanced competition; organizations previously differentiated by access to AI are now diverging based on how they utilize it. This isn’t simply a matter of increased activity, however; message volume alone accounts for only 36% of this disparity, indicating that the true differentiator lies in the complexity and depth of AI integration. A key indicator of this advanced usage is the adoption of agentic workflows, particularly with tools like Codex. This suggests a shift from AI as a simple assistant to AI as a collaborative partner capable of handling more substantive work.
the biggest gains came when they treated Codex as “part of the team.”
Tokens Measure Depth of AI Use Beyond Message Volume
OpenAI’s recent analysis of enterprise AI adoption reveals a growing divergence in how organizations are leveraging the technology, with a focus shifting from simple access to demonstrable impact. Rather than simply counting the number of employees using AI tools, the company is now tracking “intelligence demanded” through the measurement of tokens, units of text processed by AI models, to gauge the complexity of tasks assigned. Workers at these leading organizations are prompting AI to undertake more complex assignments, providing detailed context, and generating more substantial outputs.
