Neil Gershenfeld on Self-Replicating Robots, Digital Materials, and Rethinking the Foundations of Computing

Neil Gershenfeld, director of MIT's Center for Bits and Atoms (CBA), discusses his lifelong work at the intersection of the digital and physical worlds, challenging foundational concepts in computer science and proposing a new paradigm for fabrication. He argues that pioneers like Turing and Von Neumann made fundamental mistakes by separating the 'head' (computation) from the 'tape' (information storage), leading to an artificial distinction between software and hardware. This separation, he contends, is a fiction that limits scaling and understanding of computation, which is inherently physical. Gershenfeld highlights that both Turing and Von Neumann ultimately explored the embodiment of computation through self-reproducing automata and morphogenesis, a path largely forgotten by mainstream computer science but central to his work.

Gershenfeld's work at CBA aims to bridge this divide by literally exploring how bits become atoms and vice-versa. He introduces the concept of "digital materials," analogous to Lego bricks or the amino acids in biology, where discrete, reversibly joined parts with local constraints determine global geometry. This approach, inspired by the ribosome's error-correcting assembly, allows for reliable construction from unreliable components, a principle Shannon and Von Neumann applied to communication and computation, respectively. This contrasts with traditional analog manufacturing (like 3D printing or CNC machining) where errors accumulate, much like Vannevar Bush's analog differential analyzer.

Practical applications of this philosophy are diverse and impactful. Gershenfeld recounts how instrumenting Yo-Yo Ma's cello to understand its computational capacity led to a hundred-million-dollar-a-year auto safety business for airbag sensors. His lab has developed ultralight, high-modulus materials using carbon fiber 'Lego,' created morphing airplanes, and designed super-efficiency race cars. The ultimate goal is to enable self-replicating robots that can build giant structures, from space telescopes and habitats to entire civilizations on Mars, by creating a hierarchy of robots that can make themselves from the very parts they are assembling, moving from micro-scale to macro-scale construction.

The broader implications of this work are profound, suggesting a future where fabrication is democratized and highly efficient. By digitizing materials and enabling self-assembly and self-replication, the limitations of traditional manufacturing are overcome, allowing for the creation of complex, large-scale structures with minimal resources and maximum reliability. This vision extends to solving global challenges by enabling local, on-demand fabrication of almost anything, fostering a new era of innovation that re-integrates the physical reality of computation and construction, much like the biological processes that have built life for billions of years.

I learned why Von Neumann and Turing made fundamental mistakes.

the head is distinct from the tape which means Persistence of information is separate from interaction with information.

all of the Canon of computer scientists credits them for what was never meant to be a computer architecture both Turing and Von Neumann ended their life studying exactly how software becomes Hardware.

the embodiment of computation is embodied in these really profound ways.

this is a mistake that dates back to the Renaissance... this Dreadful concept of the ill liberal arts.

what's the computational capacity of a musical instrument?

if you instrument everything he does and connect it to almost anything it sounds like yo-yo that that the magic is in the control not in ineffable details in how the wood Wiggles.

the real birth of computerized digital manufacturing is four billion years ago that's the evolutionary age of the ribosome.

the Tower made by a child is more accurate than their motor control because the act of snapping the bricks together gives you a constraint on the joints.

the ribosome who I mentioned a little while back can make an elephant one molecule at a time... because of the robot can make the robot.

how a computation communicates its own construction.