The Last Invention – Episode Summary

Podcast: The Last Invention
Episode: Wetwear
Date: April 9, 2026
Host: Longview (Andy Mills, Gregory Warner)
Special Guests: Han Wang Chong (CEO, Cortical Labs), Minas Liarokappas (CEO, Acumino; Director, New Dexterity Research Group)

Overview & Main Theme

This episode titled "Wetwear" explores the emerging frontier of "biological computers"—computing devices powered not by silicon chips, but by living brain tissue. Through interviews with pioneers in the field, including Han Wang Chong and roboticist Minas Liarokappas, the hosts unravel the strange, ethically fraught, and potentially world-changing journey towards merging biology and computing. The episode raises fundamental questions about intelligence, machine learning, embodiment, and what it could mean to literally bring 'life' to machines both intellectually and physically.

Key Discussion Points & Insights

1. A Brief History of the Computer and its Evolving Role

• Computers once seemed a 'mysterious,' niche tool—now, they're an essential part of modern society ([00:13-01:41]).
• The next evolutionary leap is "even more transformational"—computers powered by living brain tissue ([01:41]).

2. Han Wang Chong’s Wetware Computer: Origins & Design

• Han Wang Chong: Inspired by Demis Hassabis (DeepMind) to return to "our origins in neuroscience," he visits the University of Melbourne and discovers neurons growing on chips for drug research ([03:12-04:29]).
• Lab uses fetal mouse brain tissue on glass chips to observe neuron activity and responses to drugs ([04:29-05:16]).
• The leap: If neurons process information, can we teach them to 'compute' and learn?
  • Quote: “Why hasn’t anyone tried to get these neurons in a dish to try to do some sort of computer intelligence?” – Han Wang Chong ([06:13])
• Early experiments—like the "leechilator" using leech neurons to add numbers—couldn't teach the tissue to learn ([06:38-08:15]).
• Han builds a wetware computer the size of an Xbox, using human neurons, connected to wires that allow two-way communication ([08:33-09:59]).

3. Training Brain Cells to Play Pong

• Goal: Teach live brain tissue to learn via reward and punishment, starting with the game Pong ([10:29-12:03]).
• Biological computers don’t work with binary code; instead, neurons 'want' predictability, i.e., they minimize prediction error—“They seek to minimize surprise.” ([12:55-13:01])
• Analogy: Neurons prefer "a consistent beat" just like dancers in a club, hate being surprised ([13:30-14:04]).
• Reward: Give firing neurons a predictable sine wave (like a treat).
  Punishment: Blast them with white noise/static ([14:32-15:29]).
  • Quote: "It's bacon. Bacon treats for neurons." – Gregory Warner ([14:59])
• Remarkably, with this system, neurons "mastered Pong in five minutes" ([15:59-16:11]).

4. Ethics & Experience—Can Neurons Suffer?

• Question posed: Is this cruel? Are the brain cells suffering? ([16:35-16:55])
• Han doesn’t know for sure, but with only ~200,000 neurons, there’s no self-awareness—yet repeated punishment (“the silent treatment”) can force neurons into a coma-like state; at that point, “he just has to toss the whole thing in the garbage and start again.” ([18:10-19:43])
• Quote: "It would be like if you were in solitary confinement and there were no walls, no lights, no feeling, no bed that you were sitting on..." – Gregory Warner ([18:36])

5. Commercialization & Applications

• Han starts selling the wetware computer, which now can play Doom (a much harder game), not just Pong ([19:46-20:04]).
• Current customers ([21:31-22:35]):
  • Medical researchers: Use for advanced drug testing on live, computing cells (e.g., epilepsy drugs).
  • Crypto-gamers & quantum biology enthusiasts: Vague references, not detailed here.
  • Roboticists: The most consequential—aiming to use biological computers in embodied robots.

6. Robotics & The Embodied Intelligence Dream

• Classic AI paradox: Computers excel at math and chess, but struggle with 'easy' human tasks like walking or folding laundry ([26:56-28:55]).
• Roboticists (Minas Liarokappas):
  • Emphasize the limitations of current robots—they lack flexibility and can’t generalize tasks ([29:19-29:53]).
  • Quote: “I believe that in the future biological computing will facilitate [human-like dexterity].” – Minas Liarokappas ([29:53])
  • Idea: Biological material—not just in the 'brain' of robots, but perhaps in their 'skin' or 'hands'—could allow robots to sense and interact with the world more like humans ([30:42-32:14]).
  • Quote: “Maybe it’s the skin.” – Minas Liarokappas ([31:13])
• Theorizes that embodied knowledge—AI that physically interacts with the world—is vital for achieving AGI (artificial general intelligence) ([33:14-35:36]).
  • Quote: “You cannot model the complexity of the human world with text, video, or images.” – Minas Liarokappas ([35:13])
  • “Definitely. 100%. Yeah. You need to interact with the world. We feel the world around us.” – Minas Liarokappas ([35:36])

7. Wetware and Human Augmentation

• The same substrate could be used to build next-generation brain-machine interfaces—possibly giving us new skills, new knowledge, or even integrating 'superintelligence' into human minds ([37:34-38:38]).
• Quote: “Maybe we create a robot out of our wetware... AI that maybe hits AGI, maybe goes all the way to superintelligence, and then in turn, we take that super intelligence... and we stick it back in us?” – Andy Mills ([38:11])

Notable Quotes & Memorable Moments

• The 'Goopiest' Story Yet:

  • “Definitely the grossest, the most goopy for sure. It's a goop filled story.” – Andy Mills ([03:03])

• Wetware as Animal Training:

  • “To work with a biological computer, it's a lot more like training a puppy or like teaching...a baby, because you're using reward and punishment.” – Gregory Warner ([11:30])

• Neurons Seek Predictability:

  • “What they want is predictability.” – Gregory Warner ([12:55])
  • “It's bacon. Bacon treats for neurons. Like, they love it because it is predictable.” – Gregory Warner ([14:59])

• Computers That Smell:

  • “Does it smell like a pet shop or something like that?” – Andy Mills ([10:06])
  • “It smells like a hospital. Sort of like that faintly sweet, but also sterile. Also like kind of plasticky metal smell.” – Gregory Warner ([10:08])

• Purpose & Potential:

  • “There's no way that any hardware can ever match the efficiency of your brain.” – Gregory Warner ([20:54])
  • “If you could swap out some of that hardware for wetware, then it would be so much more efficient that essentially the data centers that are currently like four or five football fields in size would only need to be the size of a Hershey bar.” – Gregory Warner ([21:21])

• Reframing AGI:

  • “You cannot model the complexity of the human world with text, video or images.” – Minas Liarokappas ([35:13])
  • “But what Minas argues is that actually the path to superintelligence is not through sort of white collar work. The path to superintelligence is through finally getting an AI that knows what the rain feels like on its face...” – Gregory Warner ([36:19])

Timestamps for Major Segments

• Computers' Transformation: [00:06]–[01:41]
• Intro to Han Wang Chong & Biological Chips: [03:12]–[05:16]
• From Drug Testing to Computation: [06:09]–[08:15]
• The Device: Wetware Computer: [08:33]–[09:59]
• Teaching Neurons to Play Pong: [10:29]–[12:55]
• Reward, Punishment, and Neurons' 'Wants': [13:01]–[16:11]
• Ethics and Sentience: [16:35]–[19:43]
• Biological Computer Commercialization: [19:46]–[22:35]
• Robotics, Handling Embodiment: [26:56]–[32:14]
• Philosophy: The Brain vs Hand Debate: [33:14]–[34:23]
• Embodiment & Path to AGI: [35:09]–[36:19]
• Wetware Brain Implants & Augmentation: [37:34]–[38:38]

Final Thoughts

"Wetwear" takes listeners on a journey from the origins of computer science to the bleeding edge of biotech. The conversation blurs boundaries: between human and machine, intelligence and instinct, cruelty and progress. Whether in robotic 'skin' or as brain enhancements, wetware may not only create more efficient or more human-like AI—it may fundamentally change what it means to be intelligent and even what it means to be human.

Recommended for listeners seeking:

• The latest in neuroscience and AI crossover
• Reflections on the ethics and future potential of biotechnology
• Engaging, sometimes darkly humorous, explorations of the weirdest edges of tech
• Big questions about intelligence, consciousness, and the future of our species