Two years after ChatGPT's launch, the search for profitable applications of generative AI has found a clear winner: software development. While many AI use cases remain speculative, coding tools have demonstrated immediate value, attracting unprecedented investment and sky-high valuations.
Cursor, developed by San Francisco-based Anysphere, has emerged as the sector's standout success. The AI-powered code editor raised $900 million in May 2025, led by Thrive Capital, Andreessen Horowitz, and Accel, at a $9 billion valuation. This represents a remarkable leap from its $2.6 billion valuation just five months earlier. With only 60 employees, Cursor achieved $100 million in recurring revenue by January 2025, less than two years after launch, and is reportedly now generating around $300 million in annual recurring revenue.
Meanwhile, OpenAI has agreed to acquire Windsurf (formerly Codeium) for approximately $3 billion, marking the ChatGPT maker's largest acquisition to date. Windsurf, founded in 2021, launched its code generation product in November 2024 and quickly reached $50 million in annualized revenue. The tool has gained popularity for its ability to translate plain English commands into functional code, sometimes called "vibe coding," allowing people with no programming knowledge to write software.
These startups are thriving despite competition from established players. Microsoft's GitHub Copilot, launched in 2021 and considered the dominant player in code generation, grew to over $500 million in revenue last year with more than 15 million users. Google, Microsoft, and OpenAI all announced new code-generation products in May 2025, while Anthropic is also developing similar tools.
Corporate adoption of AI coding tools is accelerating rapidly. Google's CEO revealed that "well over 30%" of Google's code is now AI-generated, while Microsoft's CEO stated that approximately 20-30% of their code comes from AI. Amazon claims to have saved "the equivalent of 4,500 developer-years" through AI coding assistance.
However, challenges remain. Most AI coding startups operate with negative gross margins, as they rely on foundation models from companies like OpenAI and Anthropic, making their costs per query substantial. As AI revolutionizes the industry, many jobs—particularly entry-level coding positions involving repetitive tasks—may be eliminated, though experts emphasize that human intelligence remains essential to the software development lifecycle.