Reinforcement learning is terrible — but everything else is worse.

Karpathy’s sharpest takes yet on AGI, RL, and the future of learning.

Andrej Karpathy’s vision of AGI isn’t a bang — it’s a gradient descent through human history.

Karpathy on AGI & Superintelligence.

* AGI won’t be a sudden singularity — it will blend into centuries of steady progress (~2% GDP growth).

* Superintelligence is uncertain and likely gradual, not an instant “explosion.”

* The next decade will be about making agents reliable, multimodal, and memory-capable — not yet human-level minds.

On Reinforcement Learning (RL)

* RL is “*errible — but everything else is worse.”

* Its reward signals are sparse, noisy, and misaligned with real-world learning.

* Humans don’t just replay successes — they reflect, critique, and reason about what worked.

* The future: hybrid systems that combine process-based feedback, reflection, and better credit assignment.

On Self-Driving & Long Timelines.

* Self-driving took so long because the real world is messy and safety-critical.

* True autonomy needs robust perception, edge-case coverage, and simulation — not just clever code.

* AGI development will face similar complexity and bottlenecks.

On Education & Learning.

* Real learning happens when you build from scratch and struggle with the details.

* Future education will shift to AI-assisted apprenticeship and reflection, not rote memorization.

* The key skill: learning how to learn — cultivating adaptability and meta-reasoning.

01:57:08 — Future of education.

The Andrej Karpathy episode. During this interview, Andrej explains why reinforcement learning is terrible (but everything else is much worse), why AGI will just blend into the previous ~2.5 centuries of 2% GDP growth, why self driving took so long to crack, and what he sees as the future of education. It was a pleasure chatting with him.

Transcript: https://dwarkesh.substack.com/p/andre…

• Labelbox helps you get data that is more detailed, more accurate, and higher signal than you could get by default, no matter your domain or training paradigm. Reach out today at https://labelbox.com/dwarkesh
• Mercury helps you run your business better. It’s the banking platform we use for the podcast — we love that we can see our accounts, cash flows, AR, and AP all in one place. Apply online in minutes at https://mercury.com
• Google’s Veo 3.1 update is a notable improvement to an already great model. Veo 3.1’s generations are more coherent and the audio is even higher-quality. If you have a Google AI Pro or Ultra plan, you can try it in Gemini today by visiting https://gemini.google

To sponsor a future episode, visit https://dwarkesh.com/advertise 00:00:00 – AGI is still a decade away 00:30:33 – LLM cognitive deficits 00:40:53 – RL is terrible 00:50:26 – How do humans learn? 01:07:13 – AGI will blend into 2% GDP growth 01:18:24 – ASI 01:33:38 – Evolution of intelligence & culture 01:43:43 — Why self driving took so long 01:57:08 — Future of education.
Apple Podcasts: https://podcasts.apple.com/us/podcast…
Spotify: https://open.spotify.com/episode/3iIY…

Labelbox helps you get data that is more detailed, more accurate, and higher signal than you could get by default, no matter your domain or training paradigm. Reach out today at https://labelbox.com/dwarkesh.
Mercury helps you run your business better. It’s the banking platform we use for the podcast — we love that we can see our accounts, cash flows, AR, and AP all in one place. Apply online in minutes at https://mercury.com.
Google’s Veo 3.1 update is a notable improvement to an already great model. Veo 3.1’s generations are more coherent and the audio is even higher-quality. If you have a Google AI Pro or Ultra plan, you can try it in Gemini today by visiting https://gemini.google.

To sponsor a future episode, visit https://dwarkesh.com/advertise.

00:00:00 – AGI is still a decade away.
00:30:33 – LLM cognitive deficits.
00:40:53 – RL is terrible.
00:50:26 – How do humans learn?
01:07:13 – AGI will blend into 2% GDP growth.
01:18:24 – ASI
01:33:38 – Evolution of intelligence & culture.
01:43:43 — Why self driving took so long.
01:57:08 — Future of education.