Patel: AI's primary bottleneck is the inability to learn on the job
Dwarkesh Patel · DWARKESH PATEL on the Biggest Problem of AI · Jul 8, 2025 · at 6:34
Dwarkesh Patel explains why prompt engineering and current LLMs cannot replace human knowledge workers who continuously learn from workflows over time.
0:00 / 0:45exact quote · 45.9s
▶ Watch the full episode on YouTube → 720p mp4 · rendered on demand · StarZero watermark
“The key problem I see Is this, the models can't do on-the-job training. So if you think about a human employee you might have some. And these human employees, the good thing about them is that, you know, you train them for six months or a year, and over time they're getting better and better. They're learning about all the context and intricacies of your workflow, what you like. They, they'll fail, but they'll learn from their failures, or they'll interrogate them in this, like, very organic, deliberative way. They'll pick up small efficiencies, efficiencies and improvements as they practice a task. This just doesn't happen with an LLM. Every session you're getting this amnesiac mind that's very smart, but it has, it's lost all awareness of how you like things done, how your business works and so forth.”
quote is from the automated transcript, cleaned for reading: filler sounds and stutters are removed, nothing is rephrased. names can be misheard (the analysis reads context, assessments check outside sources). how →
More from Dwarkesh Patel
Patel: Humanity Has an 80% to 90% Chance of Surviving AI
“In the 80% of worlds or 90% of worlds where we don't get paper clipped you will get to say you worked on something really cool at a time that was really important in the history of humanity.”
Patel: Prompt correction is an order of magnitude worse than human learning
“I think the, this is at least an order of magnitude less efficient and less less capable than the way humans learn.”
Patel: Tacit experiential knowledge cannot be captured in a system prompt
“This, like, tacit knowledge that you build up through practice is not this, like, written instruction manual that you can just write out as a system prompt.”
Patel: AI automation is bottlenecked by task depth, not task breadth
“I think the bigger problem is not just the width or the width of the pool, how many different tasks you have to RL on, but it's a depth in the sense that a job doesn't involve doing a thousand different five minute tasks individually. It's the fact that you're…”
Patel: Aggregated on-the-job AI learning would functionally produce superintelligence
“Even if there's no software progress out of that point, that algorithms aren't improving, just that ability to learn on the job from everything in the economy would functionally produce what looks like a super intelligence, right? No human will be, will have m…”