Biochips Mimic the Brain to Cut AI Energy Use

“This is an exploration of an alternative way to form computers,” says Grassyas.

How do Biochips mimic the brain?

Traditional chips have long been limited to two -dimensional layouts, which can limit how signals flow through the system. This model began to transform, as chips are now developing 3D chips to increase the power of their devices.

Likewise, Biochips is designed to simulate 3D brain structure. The human brain can support neurons that reach 200,000 connections – at the level of intercourse, Gracias says flat silicon chips cannot achieve. This spatial complexity allows biomed transmission to transmit signals via multiple axes, which may enable processing more efficient information.

The Gracias team has developed a 3D electrical shell (EEG) that wraps around my organic, allowing the richer stimulation and registration of traditional flat poles. This CAP is compatible with the curved surface of the organs, creating a better interface to stimulate and record electrical activity.

To train members, the team uses learning reinforcement. Electrical legumes are applied to the targeted areas. When the resulting nervous activity coincides with the desired pattern, it is strengthened with dopamine, which is the natural reward in the brain. Over time, the member learns to link some stimuli to the results.

Once the pattern is learned, it can be used to control physical actions, such as directing a mini -robot car through the strategically placed electrodes. This indicates nervous distortion – the ability to produce a predictive responsive responses. Consisked reactions place this basis for more advanced functions, such as discrimination in stimulation, which is necessary for applications such as face recognition, decision -making, and generalized artificial intelligence inference.

The Gracias team is in the early stages of self -driving cars controlled by Biochips: Evidence of the concept that the system can serve as a console. This experimental act proposes future roles in robots, artificial extremities, and biomedically integrated transplantation that communicates with human tissues.

These systems also carry a promise in disease modeling and drug testing. Gracias group develops organic muscles that mimic neurological diseases such as Parkinson’s. By monitoring how these sick tissues respond to various medicines, researchers can test new treatments in a dish instead of dependent only on animal models. They can also detect the potential mechanisms of cognitive weakness that fails the current artificial intelligence systems in simulating.

Since these chips are alive, they require continuous care: regulating temperature, nutrient feeding, and waste removal. The vital integrated Gracias team kept alive and practical for up to a month with continuous monitoring.

Farid Jordan (left) Martin Cotter are the founders of the finals, a Swiss startup that develops Biochips, which the company claims to store data in live neurons.finals

Challenges in the expansion of Biochip technology

However, there are still great challenges. Biochips is fragile and high maintenance, and current systems depend on huge laboratory equipment. It will require a reduction in practical use and technologies that can independently manage life support functions. Also, nerve cumin, sign noise, and the ability to training on neurons also offer obstacles to the conclusion of artificial intelligence in real time.

“There are a lot of biological questions and devices,” says Grassyas.

Meanwhile, some companies test water. Euolina Cortez, a scientist and strategic consultant at the company, says that the Swiss start -up finals claim that Biochip can store data in living neurons – a milestone called “Bio Bit”. This step indicates that biological systems can one day perform the basic computing functions that are traditionally dealt with by silicone devices.

FinalSpark aims to develop bioseerves that can be accessed from a distance of public computing about a decade. The aim of this is to match digital treatments in performance while being more efficient in energy. “The biggest challenge is to program neurons, as we need to know a completely new way to do this,” says Cortez.

However, the transition from the laboratory to the industry will require more than just technical breakthroughs. “We have enough financing to keep the laboratory operation,” says Grassyas. “But to take off, more funding is needed from the silicon Valley.”

Whether Biochips will increase or replace silicone still should be seen. But as artificial intelligence systems demand more and more, the idea of chips that think – and Sip Energy – has become increasingly attractive.

For Gracias, this technology can be charged to the market sooner than we think. “I don’t see any major removal on the way to implement this,” he says.