Cursor CEO explains how his AI coding assistant will compete with OpenAI and Anthropic

Cursor reaches $1 billion milestone with no IPO plans

Anysphere, the company behind the AI coding assistant Cursor, has reached $1 billion in annualized revenue as of November, yet CEO Michael Truell made clear at Fortune's AI Brainstorm conference that an IPO isn't on the horizon1

1

2

2

. The AI coding firm recently raised $2.3 billion at a $29.3 billion valuation last month, providing substantial resources to focus on feature development rather than preparing for public markets1

1

. This aggressive growth trajectory positions Anysphere among the fastest-rising startups in the AI sector, attracting attention from investors and competitors alike.

How Cursor plans to compete with OpenAI and Anthropic

When asked how Anysphere would compete with OpenAI and Anthropic—the very LLM providers Cursor depends on—Cursor CEO Michael Truell dismissed concerns with a striking analogy1

1

. He likened their coding products to "a concept car" while positioning Cursor as a production automobile ready for real-world use2

2

. Truell explained that Cursor takes "the best intelligence that the market has to offer from many different providers" and combines it with proprietary large language models designed for specific products1

1

. In November, Anysphere confirmed its home-grown models "now generate more code than almost any other LLMs in the world," highlighting the scale and efficiency of their internal development2

2

. This strategy allows Cursor to function as a comprehensive coding tool that integrates multiple intelligence sources with optimized user experience design.

Usage-based pricing model sparks controversy

The relationship between Cursor and its model-provider competitors became a subject of speculation earlier this year when OpenAI reportedly explored acquiring Anysphere, an offer the company declined1

1

. Investors told TechCrunch that AI coding editors were losing money due to high API costs charged by model makers1

1

. Rather than sell, Anysphere pivoted to a usage-based pricing model in July, directly passing along API fees to users instead of maintaining an all-inclusive subscription fee1

1

2

2

. The change caused an uproar among some customers who faced unexpectedly high bills1

1

. Truell defended the shift by noting how user behavior evolved: "When we started Cursor, you would turn to Cursor for a quick JavaScript question and now you're turning to it to do hours of work for you"1

1

2

2

.

Building spend controls for engineering teams

To address enterprise concerns about cost management, Anysphere is developing cloud-computing-like tools that allow companies to monitor usage and implement spend controls for engineering teams1

1

. Truell revealed that "we have a whole team internally dedicated to enterprise engineering and building things like spend controls and billing groups and visibility"1

1

2

2

. These features aim to give businesses better financial oversight as AI tools become more resource-intensive, addressing a critical pain point for organizations scaling their AI adoption.

Expanding beyond code generation into the software development lifecycle

Looking ahead, Truell outlined two major focus areas for the coming year. First, Cursor aims to handle complex agentic functions that can complete end-to-end tasks autonomously1

1

. He specifically mentioned bug fixes that might be "easy to describe but take weeks of someone's time, thousands of times running the code" as tasks Cursor should handle entirely1

1

. Second, the company is "thinking about teams as the atomic unit that we serve," suggesting a shift toward serving entire development organizations rather than individual coders1

1

. Cursor is also expanding into other parts of the software development lifecycle beyond writing code, with its code review product already being used by some customers to analyze every pull-request, whether written by AI or humans1

1

. Meanwhile, competition intensifies as Amazon just released a coding tool that can run for days on end, and major players including Anthropic, OpenAI, Microsoft, and AWS launched a new consortium under the Linux Foundation to develop open source agentic interoperability standards, contributing projects like Anthropic's Model Context Protocol1

1

.