neil_gershenfeld's picture
Physicist, Director, MIT's Center for Bits and Atoms; Co-author, Designing Reality

Life is defined by organic chemistry. There's software for artificial life and artificial intelligence, but these are, well, artificial — they exist in silico rather than in vivo. Conversely, synthetic biology is re-coding genes, but it isn't very synthetic; it uses the same sets of proteins as the rest of molecular biology. If, however, bits could carry mass as well as information, the distinction between artificial and synthetic life would disappear. Virtual and physical replication would be equivalent.

There are in fact promising laboratory systems that can compute with bits represented by mesoscopic materials rather than electrons or photons. Among the many reasons to do this, the most compelling is fabrication: instead of a code controlling a machine to make a thing, the code can itself become a thing (or many things).

That sounds a lot like life. Indeed, current work is developing micron-scale engineered analogs to amino acids, proteins, and genes, a "millibiology" to complement the existing microbiology. By working with components that have macroscopic physics but microscopic sizes, the primitive elements can be selected for their electronic, magnetic, optical or mechanical properties as well as active chemical groups.

Biotechnology is booming (if not bubbling). But it is very clearly segregated from other kinds of technology, which contribute to the study of, but not the identity of, biology. If, however, life is understood as an algorithm rather than a set of amino acids, then the creation of such really-artificial or really-synthetic life can enlarge the available materials, length, and energy scales. In such a world, biotechnology, nanotechnology, information technology, and manufacturing technology merge into a kind of universal technology of embodied information. Beyond the profound practical implications, forward- rather than reverse-engineering life may be the best way to understand it.