AI coding tools have rapidly advanced in recent months, says OpenAI’s president.
At a Sequoia Capital talk uploaded Thursday, Greg Brockman said that AI has become more than just a supporting tool for software engineers.
“If you look even over the course of December, we went from these agentic coding tools writing 20% of your code to writing 80% of your code,” Brockman said. “Which means they go from being kind of a sideshow to being the main thing that you’re doing.”
Brockman, who cofounded the AI company in 2015, said he would advise founders to “lean in” and embrace AI tools because they’re making rapid strides.
For example, he said, Codex — OpenAI’s code-generation platform — has recently evolved from a tool primarily for software engineers to one that can support “anyone who’s doing work with a computer.”
Still, Brockman said OpenAI ensures that a human is responsible for all code that is merged.
“That thoughtfulness of not just saying ‘oh just blindly use’ this or ‘we don’t want to use this at all.’ I think neither extreme is quite right,” he said about AI-generated code.
AI’s rapidly growing role in writing code
Brockman is part of a group of tech leaders touting AI’s ability to generate a large chunk of a company’s code.
Last month, Google CEO Sundar Pichai said that 75% of new code created inside the company is now generated by AI and reviewed by human engineers.
That figure has risen rapidly in recent years, from 25% in 2024 to 50% last year.
Meta is looking to make headway in this area as well.
In March, Business Insider reported that Meta expected 65% of engineers in its creation organization — responsible for building and maintaining core creative experiences — to write more than 75% of their committed code using AI.
Anthropic CEO Dario Amodei has also predicted how much code AI will generate.
“I think we will be there in three to six months, where AI is writing 90% of the code. And then, in 12 months, we may be in a world where AI is writing essentially all of the code,” Amodei said at a conference last year.
In a blog post earlier this year, Amodei wrote: “Because AI is now writing much of the code at Anthropic, it is already substantially accelerating the rate of our progress in building the next generation of AI systems.”
