Biocomputing: Ethically Grey, Energy Efficient Technology

As our digital world grows, so does the demand for faster and more powerful computers. But this growth comes with a cost. Data centres and artificial intelligence models require huge amounts of electricity, contributing to environmental challenges. To address this, scientists are exploring a radical alternative: biocomputing.

Biocomputing uses biological systems such as DNA, enzymes, or even living brain cells to process and store data. Unlike traditional computers, biocomputing can operate with very low energy consumption. But with this innovation comes a set of complex ethical questions, especially when living human cells are involved.

Why Biocomputing Matters

Traditional computers rely on electricity and generate heat during processing. In contrast, biocomputing mimics how biological organisms function, using chemical reactions or living cells to carry out computations. These natural processes use far less energy than silicon chips.

One example is the DishBrain project by Cortical Labs, which developed a computer made from about 200,000 live neurons. This system learned to play Pong by reacting to electrical signals and used a fraction of the energy a conventional computer would need. Another project by FinalSpark uses human brain organoids to run tasks with up to one million times less energy than silicon processors.

Types of Biocomputing

Biocomputing can take many forms depending on the biological material and the goal of the research.

Organoid Intelligence involves using mini-brains grown from stem cells. These organoids can show learning behaviour, respond to stimuli, and may one day support more advanced AI systems.

DNA and enzyme computing uses chemical reactions to build logic gates. These systems can carry out operations like AND or OR in a test tube without needing any living cells.

Neuron-based processors are created by connecting brain cells in patterns that perform memory and logic functions. These networks are being used to build simple computing systems from living cells.

Where Biocomputing Could Be Used

Although still in development, biocomputing could become valuable in a number of areas.

In medicine, researchers can use brain organoids to test how individual patients might respond to drugs. This helps create safer, more personalised treatments (Smalley, 2024).

In artificial intelligence, biological computing could reduce the energy needed to train and run large models. This would make AI more sustainable. 

In environmental monitoring, bacteria or enzyme systems can detect pollution or measure nutrients in soil. These systems work naturally and break down harmlessly after use 

The Ethical Questions

Biocomputing introduces questions that science and society have never faced before.

There is concern about consciousness. If a system made from human neurons can learn or remember, does it have some form of awareness? Scientists stress that current organoids are not conscious, but future advances could raise serious concerns (Smalley, 2024).

There are also issues with donor consent. Many of these organoids come from human stem cells. It is important to ensure that donors fully understand how their cells may be used and who owns the outcomes.

Finally, there is little agreement on how to regulate biocomputing. Should these systems be treated like software, like human tissue, or as something entirely new? Laws and guidelines will need to evolve quickly to keep pace with the science.

References

Bouchard, K E, Greaves-Tunnell, A, Fong, R and Lin, J (2024) ‘Neural logic and memory using living cells: A biological computing framework’, arXiv preprint. Available at: https://arxiv.org/abs/2408.07628 

Cortical Labs (2023) DishBrain: Biological computing with neurons. Available at: https://corticallabs.com 

FinalSpark (2024) Neural processors for sustainable computing. Available at: https://www.finalspark.com 

Harding, A (2023) ‘Enzyme-based computing: A new way to use biology as a processor’, Phys.org. Available at: https://phys.org/news/2023-05-biocomputing-method-enzymes-catalysts-dna-based.html 

Nguyen, M, Adewale, O and Zhang, Y (2023) ‘Vernacular bio-sensing for agriculture’, Buildings, 13(4), pp. 1 to 18. doi:10.3390/buildings13040878.

Smalley, E (2024) ‘Organoid intelligence: Should we be worried about brain tissue in machines?’, Financial Times, 15 March. Available at: https://www.ft.com/content/713eab47-a1f1-4477-a7de-f2b150e2faac.