There’s plenty of concern around how AI will eliminate jobs, but the Chairman of OpenAI’s board believes that that is the wrong way to look at the situation.
Bret Taylor — who also co-created Google Maps and served as co-CEO of Salesforce — argues that the right lens for understanding AI’s impact on work isn’t the individual employee, but the business process. “The atomic unit of productivity in AI is a process, not a person,” he said on the Cheeky Pint podcast. It’s a reframe with real consequences for how companies should think about deploying AI.
Taylor uses the example of a personal assistant to draw the distinction. “If you have an assistant, he or she might help you prepare for a podcast, might help you prepare for a meeting. He or she might also get you a cup of coffee. AI will be really good at the first two, but quite poor at the last one. So, short of robotics, no matter of AGI will get you a cup of coffee.”
That boundary — the digital versus the physical — is central to his argument. “It’s wrong to think about AI as replacing people. In addition to being inhumane, it’s just nonsensical, because AI operates in the world of digital technologies.”
To make the process-level argument concrete, Taylor walks through a mundane but instructive example: onboarding a new supplier.
“Think about all the departments and people involved in that. There’s a legal department to do a contract. There’s a finance department and procurement to negotiate the relationship. You probably have IT that’s involved to onboard them into your core systems. And then there’s usually a business that’s responsible. It’s fairly mundane, happens all the time.”
He then poses a CEO-level question: “Let’s just say you tracked what the median amount of time it takes to onboard a new supplier was, and it was 17 days, just for argument’s sake. I bet you could say, as a CEO of a company, I want to use AI to optimize that process and make it 17 hours, or one day. And you could go through — if you had a product manager on that and optimized every part of it — I bet you could achieve that. But the hard part isn’t a person’s job. It’s actually all the systems and people in between.”
The implication is that AI’s leverage point is latency and coordination across systems, not the replacement of any individual role. A 17-day supplier onboarding isn’t slow because any single person is inefficient — it’s slow because of handoffs, waiting periods, and fragmented systems. That’s precisely where AI can compress time.
This is a useful corrective to the dominant anxiety around AI and employment. The fear, broadly, is of direct substitution — AI takes the job, the person is redundant. Taylor is describing something structurally different: AI compresses multi-step, multi-department workflows, making processes faster without necessarily making any specific role disappear. The value creation is in the coordination layer, not the human layer.
That said, the on-the-ground reality is more complicated. Amazon has announced plans to shrink its corporate workforce over the coming years due to AI-driven efficiencies, and has already followed through with 14,000 corporate layoffs. Meanwhile, companies like Salesforce and Klarna have paused software engineer hiring entirely. The jobs aren’t always disappearing mid-process — sometimes they disappear because the process itself has shrunk. Taylor’s framing may describe the mechanism accurately while underplaying that outcome.
Still, the process-first lens is valuable for businesses thinking about where to actually deploy AI. Rather than auditing headcount, the more productive question may be: where are our slowest, most friction-heavy workflows — and how much of that friction lives in digital handoffs that AI could eliminate? Supplier onboarding is one example. Procurement approvals, compliance reviews, and customer onboarding are others. The 17-days-to-17-hours thesis is compelling precisely because it doesn’t require anyone to be fired — it just requires the process to be redesigned around what AI is actually good at.