Latest news with #neuromorphic
Forbes
2 days ago
- Science
- Forbes
Hardware, Software, Meet Wetware: A Computer With 800,000 Human Neurons
The Cortical Labs CL1 biological computer with human brain cells. The world's first 'code-deployable' biological computer is now for sale. The Cortical Labs CL1 costs $35,000 and has 800,000 human brain cells living and growing in a nutrient solution on a silicon chip. Computer scientists can deploy computer code directly to these neurons, which have been integrated into a 'biOS' or Biological Intelligence Operating System with what the company says is a mixture of hard silicon and soft tissue. The goal, according to the companies' founder? Smarter AI that drops some of the A and adds more of the I. Maybe, eventually, smarter brains than the ones we currently walk around with. 'The only machine or the only thing that we know of that actually has true intelligence is the brain,' founder and CEO Hon Weng Chong told me when I interviewed him five years ago, while he was still using mice neurons. 'So we said, let's start with the basic building structure, the building blocks being neurons, and let's build our way up and maybe we'll get there along the way.' Our human brains have neurons connected together in hierarchies, and from that emerges intelligence and consciousness, he adds. This approach is similar to neuromorphic computing architectures, which attempt to mimic biological brains with silicon-based hardware, but of course different in that neuromorphic chips do not typically use actual living brain cells. Cortical Labs, based in Australia, says scientists can solve today's most difficult problems with their biological computers, which they say are self-programming and infinitely flexible. A key difference between biological computers and silicon-based chips, of course, is that biological computers last even less time. The neurons that ship with your CL1 will live for 'up to six months,' at which point you'll likely have to invest in a refresh or refurbishment which provides new neurons for continued compute. And yes, biological computers need food and water and nutrients, all of which are supplied onboard via a life-support system the keeps them at optimum temperature. Plus, it filters out waste byproducts of living human cells: the kind of work kidneys might do in a full living organism. A Cortical Labs chip under a high-powered electron microscope. You can see tight connections between ... More neurons and the silicon substrate, the company says. In some ways the CL1 is more like a space ship than a computer, because it's a self-contained life support system that requires few external inputs. A key difference: the need for external power. From the outside, though, you treat the CL1 as a typical computer. You can plug in USB devices, cameras, even actuators if you want your CL1 to control a physical system. (Which, frankly, human neurons are typically pretty good at.) And there's a touchscreen so you can see system status or view live data. Five years ago, Cortical Lab's then-CTO Andy Kitchen told me they were deploying systems with tens of thousands of neurons to hundreds of thousands of neurons, but that their roadmap included 'scaling that up to millions of neurons." Now Cortical Labs sees their biological computers growing to hundreds of millions of cells, and with different technologies, billion or trillion-cell levels. However, there's not a direct one-to-one equivalent with neuromorphic neurons in a silicon-based system, he added. Biological neurons are much more powerful, he says. Interestingly, communicating with physical human neurons in a biological computer is vastly different than writing computer code to an artificial computer. 'The premier way would be to describe your task somehow, probably through some sort of very high-level language, and then we would turn that into a stimulus sequence which would shape biological behavior to fit your specification,' Kitchen told me. Part of the difference is how to encode and communicate the problem, and part of the difference is that the CL1 neurons, like the ones in your brain right now, have some plasticity: they can essentially reprogram themselves for different tasks. Essentially, the neurons learn how to solve your problem, just like you learn how to do new things. You won't likely see CL1 systems in general use anytime soon: currently, the targeted customers are in medical fields like drug discovery and disease modeling, says IEEE Spectrum. There's the added value that scientists can perform experiments on a little synthetic brain as well. If all of this seems on the edge of creepy, or even right over, that's likely because it is. CL1 says they don't do any animal testing, although they did start with mouse neurons, and they say that the human brain cells in their biological computers are lab-grown. But clearly the first human neurons came from somewhere. Cortical Labs says customers have to get 'ethical approval' to general cell lines, and require buyers to have proper facilities to maintain the biological chips. What exactly that means, however, is unclear. Soon we may see physical system in the world, like humanoid robots, with partially organic components to their brains.
Yahoo
24-05-2025
- Business
- Yahoo
Sam Altman's $150M AI Chip Bet Crashes: Rain AI Faces Sale As OpenAI, Nvidia, And Microsoft Circle The Wreckage
Rain AI, a San Francisco-based chip startup backed by OpenAI CEO Sam Altman, is exploring a sale after its ambitious $150 million Series B funding round failed to secure investors. Despite early backing from Altman in a $25 million seed round in 2022, Rain AI has struggled to convert technical ambition into scalable business outcomes. The company, which aimed to rival industry giants like Nvidia (NASDAQ:NVDA) with its energy-efficient chips, is now in discussions with potential buyers, including OpenAI, according to New York Post. Don't Miss: Hasbro, MGM, and Skechers trust this AI marketing firm — 'Scrolling To UBI' — Deloitte's #1 fastest-growing software company allows users to earn money on their phones. Wired reports that Rain AI aimed to develop neuromorphic chips designed to process AI workloads more efficiently by mimicking how the human brain functions. According to Wired, the company hoped these chips would deliver better performance at a fraction of the power consumption compared to traditional graphics processing units, making them ideal for generative AI models. According to the Post, testing showed promising results, but the company failed to secure major commercial contracts or letters of intent, a problem that hampered investor confidence. One source close to the company told the Post that the founders were talented engineers but lacked the sales acumen to close enterprise deals. Trending: Maker of the $60,000 foldable home has 3 factory buildings, 600+ houses built, and big plans to solve housing — Rain AI's co-founder, Jack Kendall, acknowledged to stakeholders that the company was rapidly depleting its cash reserves and urgently required a $3 million bridge fund to sustain operations while negotiating a potential acquisition. The Post says that the company has since secured the emergency funding and entered what was described as 'good progress' in acquisition talks with multiple interested parties. These talks involve several high-profile technology firms that view Rain AI's hardware as a strategic foothold in the increasingly competitive chip market. While the company had originally planned to launch its Series B round in December, the raise was delayed repeatedly due to leadership instability and difficulty closing large institutional checks, the Post reports. The company also hired former Apple (NASDAQ:AAPL) veteran Jean-Didier Allegrucci to lead chip development efforts, bringing experience from Apple's custom silicon division, the Post Passo, Rain AI's co-founder and original CEO, stepped down citing personal reasons, and Kendall has since taken over as CEO, the Post reports. Rain AI is now in discussions with multiple technology giants about a potential acquisition. OpenAI, which has begun interviewing Rain employees to assess talent fit, is a top contender. Altman had previously pitched OpenAI investors to support Rain AI's Series B, aiming for a $600 million valuation, according to the Post. Rain AI's uncertain future unfolds against a backdrop of aggressive expansion in the AI hardware ecosystem. The Post reports that Nvidia, Elon Musk's xAI, Microsoft (NASDAQ:MSFT), and BlackRock (NYSE:BLK) are among the giants investing more than $30 billion to build the next generation of AI infrastructure. Rain AI may not have achieved its original vision, but its core technology and its team remain valuable assets for any player looking to gain ground in the competitive AI chip race. Read Next:Deloitte's fastest-growing software company partners with Amazon, Walmart & Target – Image: Shutterstock UNLOCKED: 5 NEW TRADES EVERY WEEK. Click now to get top trade ideas daily, plus unlimited access to cutting-edge tools and strategies to gain an edge in the markets. Get the latest stock analysis from Benzinga? APPLE (AAPL): Free Stock Analysis Report TESLA (TSLA): Free Stock Analysis Report This article Sam Altman's $150M AI Chip Bet Crashes: Rain AI Faces Sale As OpenAI, Nvidia, And Microsoft Circle The Wreckage originally appeared on © 2025 Benzinga does not provide investment advice. All rights reserved.
Forbes
20-05-2025
- Science
- Forbes
Brain-Inspired AI Chip Enables Energy-Efficient Off-Grid Processing
AI Processor In an advancement for efficient and secure AI, researchers at the Technical University of Munich (TUM) have unveiled the AI Pro chip, a neuromorphic processor that functions independently of cloud servers and without internet connections. Designed by Professor Hussam Amrouch, this brain-inspired chip performs on-device computations, enhancing cybersecurity and energy efficiency. The AI Pro's architecture emulates the human brain, integrating computing and memory units to process data locally. This design eliminates the need for data transmission to external servers, reducing latency and potential security vulnerabilities. By employing hyperdimensional computing, the chip recognizes patterns with minimal data, streamlining the learning process. Neuromorphic processors are computer chips designed to mimic the structure and function of the human brain. Unlike traditional processors, which separate memory and processing units, neuromorphic chips integrate them, similar to neurons and synapses in the brain. This allows them to process information more efficiently, especially for tasks like pattern recognition, learning from small data sets, and operating with very low power. Professor Amrouch explains, "humans draw inferences and learn through similarities" in much the same way as the AI Pro chip. This approach allows the AI Pro to function effectively with fewer training examples, making it suitable for applications where data availability is limited. The AI Pro demonstrates remarkable energy efficiency, consuming just 24 microjoules for specific tasks, up to ten times less than comparable chips. This efficiency is crucial for enabling powerful AI capabilities on battery-powered devices and scenarios where power resources are constrained. By processing data on-device, the AI Pro enhances cybersecurity. Sensitive information remains within the device, mitigating risks associated with data transmission and storage in external servers. This feature is particularly beneficial for applications in healthcare, environmental monitoring, and autonomous systems. 'While Nvidia has built a platform that relies on cloud data and promises to solve every problem, we have developed an AI chip that enables customized solutions. There is a huge market there,' explains Prof. Amrouch. The AI Pro's capabilities align with the growing demand for edge computing solutions. Its ability to operate without internet connectivity makes it ideal for remote or mobile applications, such as drones, wearable health monitors, and IoT devices. Compared to general-purpose GPU chips like those from Nvidia, which rely heavily on cloud-based processing, the AI Pro offers a specialized solution focused on efficiency and security. While it contains around 10 million transistors, significantly fewer than the over 200 billion on Nvidia's latest Blackwell B200 GPU, the AI Pro's design prioritizes targeted performance over broad applicability. However this power and efficiency comes at a price, with the one square millimeter chip currently priced at 30,000 euros. The AI Pro has progressed beyond the conceptual stage, with prototypes manufactured by Global Foundries in Dresden. This development indicates the chip's readiness for integration into commercial products. As industries seek to enhance data security and reduce energy consumption, the AI Pro represents a promising advancement in AI hardware. Its design philosophy underscores a shift towards localized, efficient processing, potentially setting a new standard for future AI applications. With the increasing capability and power of AI-specific processors, Prof. Amrouch is convinced that 'the future belongs to the people who own the hardware.'
