GPUS Go BIOLOGICAL: BBB reveals Bionode Bionode, Laboratory growth, Living neurons account for artificial intelligence applications

Graphics processing units (GPU), expensive computer chips made by proverbs Nafidiaand AMD and Sima.aiThe only way to train and spread artificial intelligence is no longer.

Biological Black Box (BBB)Emerging from its establishment in the Timoor, developing a new category of artificial intelligence devices, it came out of the ghost through the Bionode-a computing system that merges the living neurons and equipped with the laboratory with traditional processors.

The company, which works quietly while providing patents and improving its technique, believes that the biological computing approach-the cultivation of new neurons specifically to serve as computer chips using donor-made human stem cells and mice-derived cells-can provide a low-energy and air conditioning alternative to traditional adaptation.

“Over the past twenty years, three independent fields-biology, tools and mathematical tools-have advanced to the point where biological computing has now been possible,” Alex Cesnderskki, co-founder and CEO of BBB.

Member Nafidia starting incubatorBBB plays itself as a progress and an increase in the dominant silicon -based potatoes produced by NVIDIA and others.

By taking advantage of the ability of neurons to re -empty itself physically, the company aims to reduce energy costs, improve treatment efficiency, and accelerate the training of the artificial intelligence model – challenges that have become increasingly necessary with the expansion of artificial intelligence adoption.

This is not a science fiction, despite the amazing hypothesis: BBB nerve chips are already running the computer and LLMS vision for customers, and have entered talks with two partners to license its technology for computer vision applications-although the company refused to nominate its customers and partners specifically, referring to secrecy agreements. It accepts inquiries from partners and potential clients as well Website.

Mix biology and devices

At the core of the BBB approach, there is a bionode platform, which uses neurons implanted in the laboratory in computing systems.

“We have multiple models that we use,” Ksindzovsky told me. One of these models of mice cells.

The co -founder said that “hundreds of thousands of them” are built into a plate containing 4,096 pillars, which is mainly of one Bionode chip. He also said they live for more than a year before needing to replace it.

The idea is to harness the ability to naturally adapt to the neurons to treat artificial intelligence, creating a mixed computing system that is mainly different from today’s solid chips.

Ksindzovsky, which has been working with neurons on electrodes since 2005, originally enjoyed the idea of ​​using it to predict the stock market. female teacher , Dr. Steve PotterRefuse the idea at that time.

“Why don’t we use neurons to predict the stock market so that we can all be rich?” Remember Ksindzovsky’s question Potter, which he laughed as an impractical. “At the time, he was right,” he confessed to Ksenzovsky.

Since then, the improvements in electronic technology, mathematical tools, and the longevity of the life of biological computing cells have made viable. “The biological network has evolved over hundreds of millions of years to the most efficient computing system that has ever created,” explained by KSIndzovsky.

This setting provides two Foreign models:

• Seeing the most efficient computer: Bionode was tested as a pre -processing process for the tasks of artificial intelligence classification, which reduces both inference and GPU.

• Training of the Right Grand Language Model (LLM): Unlike graphics processing units, which require frequent re -training courses, neurons adapt to the fly. This can significantly reduce the time and energy needed to update the LLMS, and treat the main bottle cervix in scaling artificial intelligence.

“One of our biggest breakthroughs is the use of biological networks to train LLMS more efficiently, which reduces the massive energy consumption required today,” said Ksindzovsky.

Building a viable living graphics processing unit with the help of NVIDIA

NVIDIA graphics units played a fundamental role in AI’s rapid progress, but their high energy consumption and cost -increasing concern raised concerns about expansion.

BBB sees an opportunity to provide a more efficient energy alternative while continuing to work in the NVIDIA ecosystem.

“We do not see ourselves as direct competitors for Exlevia, at least in the near future,” Ksandzovsky pointed out. “Biological Computing and Silicon Computing will coexist.

In fact, according to the co -founder, “we can use our biological networks to increase and improve silicone artificial intelligence models, making it more accurate and more efficient in energy.”

He said that the long -term vision of artificial intelligence devices will be a normative environmental system, as it plays biological computing, silicone slices, and even quantitative computing for each role.

He said: “The future of computing will be a normative environmental system, as the traditional silicone, biological computing and quantum computing play a role based on their strengths.”

Although BBB has not yet revealed the date of the commercial launch, the company moves from Baltimore, Maryland to the Gulf region as it facilitated to expand technology.

The future of hybrid artificial intelligence

Although silicon-based graphics processing units remain the standard of industry, the BB’s Brain-ON-CIP concept provides an overview of a future as artificial intelligence devices are no longer limited to transistors and circles.

The ability of neurons to restore itself dynamically can enable the most efficient, adaptive, adaptive intelligence systems, and capable of continuous learning.

“We are already applying biological computing to the computer’s vision,” said Kesenzovsky.

Besides efficiency gains, BBB also believes that his biological approach can provide a deeper view of how to treat artificial intelligence models.

“We have built a closed episode system that allows nerve cells to re -connect themselves, and increases efficiency and accuracy of artificial intelligence tasks.”

Despite the capabilities, Ksandzovsky acknowledges that ethical considerations will be a continuous discussion. BBB is already working with ethics and organizational experts to ensure the development of its technology responsibly.

“We do not need millions of neurons to treat the entire environment as the brain does.

BBB bets that live tissues, not just silicone, can be the following AI jump key.

Daily visions about business use cases with VB daily

If you want to persuade your boss at work, you have covered VB Daily. We give you the internal journalistic precedence over what companies do with obstetric artificial intelligence, from organizational transformations to practical publishing operations, so that you can share visions of the maximum return on investment.

Read with us privacy policy

Thanks for subscribing. Check more VB news bulletins here.

An error occurred.