In September 2026, MIT researchers announced that they had engineered colonies of bacteria to behave like transistors: the tiny switches at the heart of every electronic device. Printed onto an agar plate in precise arrangements, these living colonies form “circuit boards” that pass chemical messages to one another and perform genuine computations, including addition and signal routing.
The work makes no claim on silicon’s dominance; its ambition is relocating computation into environments electronics cannot readily inhabit: the wet, growing surfaces of living organisms. What follows explains how these living transistors function, how they measure against their semiconductor counterparts, and what applications their creators envision.
What MIT Actually Built
The research team, led by postdoctoral researcher Hamid Doosthosseini and Christopher Voigt, head of MIT’s Department of Biological Engineering, engineered five bacterial strains that function as modular computing components. The study was published in Nature Chemical Biology in August 2026, and the funding came in part from the U.S. Defense Advanced Research Projects Agency (DARPA) and the Intelligence Advanced Research Projects Activity (IARPA).
The five engineered strains cover three roles:
- Two transistor strains: one switches on when it detects a signal molecule, the other switches off.
- Three relay strains: these receive a transistor’s output, convert it, and pass it to the next component, like wires connecting parts of an electronic board.
Together, these five strains act as a reusable toolkit. By rearranging them in different layouts, the researchers built circuits that add two or three inputs, process several signals at once, and even operate as a demultiplexer, which routes one incoming signal to a chosen destination based on a control signal. The largest circuit demonstrated contained 24 interconnected bacterial colonies wired together to add two inputs.
“We’ve built some initial computer architecture components that are commonly used, but any operation can be built with these five strains,” Doosthosseini said in MIT’s announcement of the work.
How a Bacterial Transistor Works
In a silicon chip, a transistor controls the flow of electrical current. In MIT’s living version, engineered cells control the flow of small signaling molecules instead. The team built its components from Pantoea agglomerans, a bacterium that commonly grows on plant surfaces, making it a natural candidate for circuits meant to live on crops one day.
The mechanism works in three steps:
- The two transistor strains respond to a molecule called OC 6. One strain activates when it encounters OC 6; the other deactivates.
- Each transistor also senses a second molecule, OC 12. Depending on OC 12’s presence and its own on or off state, the cell produces an output molecule called OHC 14.
- The relay strains convert OHC 14 into a signal that can trigger the next transistor, chaining individual cells into a larger circuit.
To assemble the circuits, the researchers printed bacterial colonies onto agar plates, spacing each colony about 5 millimeters from its nearest neighbor. That precise spacing matters: it keeps chemical signals flowing to only the next colony in the sequence, so information moves in one direction, just like current following traces on a circuit board.
A notable finding was that the same bacterial transistor could perform different logic operations, including multi-input, OR, and imply gates, depending only on where it sat in the circuit. In living electronics, physical layout replaces wiring diagrams.
How Living Transistors Compare With Silicon Transistors
The headline claim from the researchers is audacious but precise. “Computationally, there’s nothing that your iPhone can do that these circuits couldn’t do,” Voigt said. The catch is speed, and the difference is enormous.
| Feature | Bacterial transistor (MIT, 2026) | Silicon transistor |
|---|---|---|
| Signal carried | Chemical signaling molecules | Electrical current |
| Computation time | About 8 hours per calculation | Less than a nanosecond per switch |
| Assembly method | Colonies printed onto agar, ~5 mm apart | Lithography on silicon wafers |
| Largest demonstrated circuit | 24 interconnected colonies | Tens of billions of transistors per chip |
| Self-renewal | Living colonies grow and replenish themselves | None; components degrade permanently |
| Operating environment | Wet, biological surfaces, ambient conditions | Sealed, dry, temperature-controlled |
| Best suited for | Control and sensing inside living systems | General-purpose computing |
Read that table as a story of two niches, not a contest. Silicon wins every speed and density metric by many orders of magnitude. The living version wins somewhere silicon cannot compete at all: environments that are wet, growing, and biological.
Breaking the Single-Cell Complexity Ceiling
Why not simply pack a whole circuit inside one bacterial cell? That is exactly what synthetic biology has tried for two decades, and it runs into hard limits. Traditional genetic circuits rely on transcription factors and proteins that interact inside a single cell, and only a limited number of these parts exist that do not interfere with one another. Cram in too many circuits and the cell’s protein-making machinery overwhelms.
MIT’s approach sidesteps the ceiling by distributing the workload. Each cell handles one simple job, transistor or relay, and complexity emerges from how colonies are arranged rather than from how much machinery is stuffed into one genome. It is a shift from designing genes to designing layouts, and it is why the same five strains can be reassembled into nearly any circuit.
This modular strategy places the work firmly in the emerging field of wetware computing, where living cells become the substrate for information processing rather than merely the subject of it.
Which Bacteria Generate Electricity?
A common question raised by this research is whether the MIT bacteria produce electricity. They do not. MIT’s living transistors communicate purely through chemical messages, not electron flow. But a separate lineage of bioelectronics does exploit microbes that generate electricity.
Bacteria such as Shewanella oneidensis and Geobacter sulfurreducens are electrogenic: they transfer electrons outside their cells as part of their metabolism, a process researchers have harnessed in microbial fuel cells. Scientists have even used dead bacterial biofilms to generate electricity from sweat on wearable patches.
So bioelectronics currently advances along two parallel tracks: electricity-producing microbes that can power small devices, and chemical-signal microbes like MIT’s that can compute. Combining the two is one of the field’s obvious next frontiers.
Living Circuits in Agriculture and Beyond
The application MIT points to most directly is agriculture. The vision looks like this: living circuits coat plant roots or leaves, continuously processing chemical information about conditions around them. When the system recognizes a stress signal, such as drought or a pest attack, it triggers a matching response, for example, producing a fungicide on the spot.
The eight-hour computation speed sounds crippling until you consider the clock it needs to beat. “If you have bacteria on the root of a plant, or the plant itself is doing the computing, running a simple calculation overnight is fast enough relative to a growth season,” Voigt explained.
Beyond smart agriculture, potential applications include:
- Living environmental sensors: bacterial circuits that detect contaminants in soil or water and release a visible or chemical warning signal.
- Bioprocess control: circuits that monitor fermentation vats or bioreactors and adjust chemical outputs without external electronics.
- Medical surfaces: coatings for implants that sense infection markers and release defensive compounds on demand.
- Programmable materials: surfaces embedded with living logic that responds to touch, moisture, or specific molecules.
The Honest Limits: Speed, Scale, and Survival
Reportage on biological computing skews toward hyperbole, so a candid inventory of the technology’s current limitations is warranted.
- Speed: an eight-hour calculation versus nanoseconds for silicon means bacterial circuits will never run anything resembling a modern application.
- Scale: 24 colonies is the largest circuit demonstrated, all on a laboratory agar plate under controlled conditions.
- Range: diffusion-based chemical signals are slow and short-ranged, which is why 5-millimeter spacing works, but longer “wires” do not yet exist.
- Stability: living systems evolve. Engineered functions can degrade over many generations of bacterial growth, a problem silicon simply does not have.
- Environment: the circuits need nutrients and hospitable conditions. Outside the lab or a plant surface, keeping them alive and functional is an unsolved engineering problem.
The researchers are explicit that they are not competing with electronic computers. As Voigt put it, “We’re not trying to replace computers, but rather put computational control into biology.”
From Bell Labs to Biofilms: Where Transistors Go Next
The transistor was invented at Bell Laboratories in 1947, and it became the most manufactured device in human history, with modern chips packing tens of billions of them onto a fingernail-sized slab of silicon. Transistors are very much still used today: they are the foundation of every processor, phone, and sensor in modern life.
What MIT’s work represents is a widening of the transistor concept itself. First the switch moved from vacuum tubes to doped silicon; now it has been rebuilt out of living cells and chemical messages. In the same way that researchers have shown it is possible to store data inside the DNA of living organisms, this study shows that computation, not just memory, can be embedded in biology.
Frequently Asked Questions
What are the three types of transistors?
The three main transistor families in electronics are the bipolar junction transistor (BJT), the field-effect transistor (FET, including MOSFETs), and the insulated-gate bipolar transistor (IGBT). MIT’s bacterial transistor is a biological analog that fills the same switching role, but it belongs to no silicon family: it switches chemical signals instead of current.
Did Bell Laboratories invent the transistor?
Yes. John Bardeen, Walter Brattain, and William Shockley invented the transistor at Bell Labs in 1947, work that earned them the 1956 Nobel Prize in Physics. MIT’s 2026 study updates the concept nearly 80 years later by rebuilding the switch out of living bacterial cells.
Are transistors still used today?
Absolutely. Transistors remain the fundamental building block of all modern electronics. The new development is not a replacement but an expansion: transistors made of bacteria open up biological environments where silicon devices cannot operate.
How long does one calculation take in MIT’s living circuit?
About eight hours per calculation. That is far too slow for electronics but practical for biological settings where a plant responds on the timescale of a day or a growing season.
Could living transistors ever replace computer processors?
No. The researchers are explicit that the goal is to add computational control to biology, not to rival silicon. Bacterial circuits will never approach the speed of electronic chips; their advantage is operating inside wet, living systems.
What bacterium is used in MIT’s living transistors?
Pantoea agglomerans, a species that commonly grows on plant surfaces. Its natural affinity for plants makes it well suited for the team’s agricultural vision of circuits living on roots and leaves.
Conclusion
MIT’s living transistors will not end up in your phone, and that was never the point. By turning five engineered bacterial strains into a modular, printable toolkit, the team showed that computation can be woven directly into biological environments, solving the complexity ceiling that has limited genetic circuits for two decades. The speed is glacial and the circuits are small, but the direction is unmistakable: the future of computing may include chips that grow, heal, and live on the roots of crops.
