Has the era of organic computing already begun? Judging by the latest news from Switzerland, it would seem so. FinalSpark, a young and ambitious startup, has just announced that it has developed the first commercial bioprocessors in history. Devices that not only open up completely new perspectives in the field of computing, but which could also help solve one of the most pressing problems in the sector: enormous energy consumption. How do they work? Let's find out together.
A revolutionary approach to data processing
The “neuroplatform” of FinalSpark it is a radical paradigm shift compared to traditional silicon processors. Instead of relying on electronic circuits, this technology directly exploits the computational potential of biological neurons, grown in vitro in the form of human brain organoids.
An architecture defined as "wetware", which merges biology, software and hardware in a way never seen before.
According to what the startup reported in its research paper (that I link to you here), the heart of the system is made up of four “multi-electrode arrays” (MEA), each of which hosts four human brain tissue organoids. In total, therefore, the platform can count on as many as 16 interconnected "mini-brains", capable of learning and processing information autonomously.
Bioprocessors for "micro brains", potential and open questions
The implications of this technology, if confirmed, would be enormous. In fact, FinalSpark claims that its bioprocessors consume up to a million times less energy than conventional chips. An astonishing figure, which could revolutionize computationally intensive sectors such as artificial intelligence and big data analysis. Suffice it to say that, according to the estimates reported by the startup (not far off from a widely ascertained truth), training a language model like GPT-3 requires around 10 GWh of electricity, more than 6.000 times the annual per capita consumption of a European citizen.
Of course, we are still in a very early and experimental stage. Before we can talk about a real breakthrough, we will need to independently verify FinalSpark's claims and evaluate the scalability of this solution. At the moment, access to the neuroplatform has been granted to only nine institutions, with a subscription model based on a proprietary cryptocurrency (500 PCM per month per user). An approach that raises more than one question about the real maturity and sustainability of the project.
But the challenges to be faced are not only technical and economic.
Ethical and safety issues
The use of human brain tissue for computational purposes opens up a series of ethical and safety questions that cannot be ignored. What are the risks of improper or malicious use of this technology? And how to regulate such a new and complex sector?
If it is true that bioprocessors could open up extraordinary horizons for human progress, it is equally true that we cannot afford to proceed blindly, without considering all the implications and responsibilities associated with this step.
Bioprocessors, the future still to be written
In conclusion, the announcement of FinalSpark undoubtedly represents a historic moment for computing and biomedicine. But it is only the beginning of a journey to be explored, made up of opportunities and pitfalls, hopes and unknowns. It is up to us, as a society, to find the right path to reap the fruits of this revolution without losing sight of the values and principles that define us as human beings.
It is not the future of computers that is at stake, but ours: a future in which the border between biological and artificial is becoming increasingly subtle and blurred. A future in which, perhaps, we will be called to redefine the very meaning of intelligence (and perhaps conscience). Bioprocessors will help us write it one bit, or rather, one neuron at a time.
Comments are closed.