Silicon Valley GirlOpenAI Co-Founder: Start Building With AI Before You Feel Ready | Greg Brockman
At a glance
WHAT IT’S REALLY ABOUT
Proactive AI agents will unlock a near-term entrepreneurship boom
- Greg Brockman argues that today’s AI models are vastly underutilized because users struggle to discover a few high-value workflows that make them “power users.”
- He describes a near-term shift from chat interfaces to proactive, context-connected agents that research, draft, prioritize, and execute tasks with human approval.
- Brockman explains that he no longer writes software directly, instead supervising AI that produces most of the implementation while he provides direction and feedback.
- He advises founders to build defensible AI products through domain expertise, stakeholder-aware workflow design, and trust/guardrails rather than brittle prompt-based wrappers.
- He predicts a major entrepreneurship boom within 1–2 years as AI collapses the cost of expertise and enables small teams to operate with leverage previously reserved for large organizations.
IDEAS WORTH REMEMBERING
5 ideasAI adoption hinges on finding 2–3 repeatable use cases, not learning every feature.
Brockman notes that many users never get past “scratching the surface,” and that retention jumps when someone discovers just three distinct, personally relevant workflows where AI consistently helps. The adoption challenge is less model capability and more helping people find their “magic” use cases quickly.
The next interface shift is proactive agents that suggest and act, with approval loops.
He describes assistants that notice neglected emails, research the right response, draft it, and ask for approval—shifting the user experience from command-driven to suggestion-driven. This reduces the “blank page” problem of figuring out what to ask for.
Coding becomes orchestration: humans set direction; AI implements and iterates.
Brockman says he no longer codes directly; he “directs an AI” to write software and focuses on feedback, guidance, and iteration. The human role becomes specifying intent, reviewing outputs, and steering toward the desired outcome.
To avoid obsolescence, build on domain insight and relationships—not model workarounds.
He argues that as app-building mechanics get commoditized, durable products come from deep user/problem understanding, workflow design, trust/guardrails, and stakeholder complexity (e.g., education or healthcare). Businesses that “wrap prompts” around current model limits are more likely to be obsoleted by better models.
AI lowers the cost of expertise and raises what small teams can realistically attempt.
For small businesses, he frames ChatGPT as “a world expert business consultant” that makes starting easier and enables tasks that previously wouldn’t get done at all (websites, storytelling/marketing assets, operational docs). He emphasizes raising the “ceiling of ambition” by imagining what you’d do with the world’s best programmer on staff.
WORDS WORTH SAVING
5 quotesLike, I don't write software anymore. I direct an AI for it to write software. I provide lots of feedback and guidance.
— Greg Brockman
You don't want an interface that has buttons and sliders and model pickers. Like, none of that. That's not the AI we were promised.
— Greg Brockman
Well, I think the answer to being ambitious enough will always be no- because there's always more we can do, right?
— Greg Brockman
And so I think this kind of empowerment, you know, being able to help people be able to be more independent, be able to l- live their life the way that they want, all of that, it's happening. It's happening right now at massive scale.
— Greg Brockman
This white page problem is, like, the hardest problem in AI in some ways, and utilizing AI is this question of where do you get started, what do you use? I think my answer is always start small. Like, just try something. Type something random, push enter, see what happens, right?
— Greg Brockman
High quality AI-generated summary created from speaker-labeled transcript.
