Brain-Inspired AI Chip Enables Energy-Efficient Off-Grid Processing
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.'
Hashtags
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
Yahoo
24 minutes ago
- Yahoo
JD.com to Acquire European Retailer CECONOMY in $2.B Takeover Bid
Inc. (NASDAQ:JD) is one of the best upside stocks to invest in now. On July 30, announced a decision to make a voluntary public takeover offer for all shares of CECONOMY AG. CECONOMY is the parent company of European consumer electronics retailers MediaMarkt and Saturn. The offer is being made through a wholly-owned subsidiary, called JINGDONG Holding Germany GmbH (the Bidder), for a cash consideration of EUR 4.60 per share. The Bidder and CECONOMY have signed an investment agreement related to the takeover offer and their future cooperation. The Bidder also has agreements with several CECONOMY shareholders, who have irrevocably committed to accept the offer for a total of 31.7% of the shares. This includes a 3.81% stake from CECONOMY's largest shareholder group, Convergenta Invest GmbH. A wide and imposing view of a supply chain distribution center, illustrating the company's technology capabilities. Post-takeover, Convergenta will hold 25.35% of the CECONOMY shares. This secures a combined shareholding of 57.1% for future partners and the company itself before the offer's launch. CECONOMY's Supervisory Board and Management Board are in full support of the takeover. plans to contribute its advanced technology, e-commerce expertise, and logistics capabilities to strengthen CECONOMY's business, which currently operates over 1,000 retail stores in 11 countries. Inc. (NASDAQ:JD) is a supply chain-based technology and service provider in the People's Republic of China. It operates through three segments: JD Retail, JD Logistics, and New Businesses. While we acknowledge the potential of JD as an investment, we believe certain AI stocks offer greater upside potential and carry less downside risk. If you're looking for an extremely undervalued AI stock that also stands to benefit significantly from Trump-era tariffs and the onshoring trend, see our free report on the . READ NEXT: and . Disclosure: None. This article is originally published at Insider Monkey. Sign in to access your portfolio
CNET
25 minutes ago
- CNET
New Study Shows Smartwatch Stress Sensors Have No Idea What They're Doing
You might want to think twice before you put a lot of stock in the latest stress charts from your fitness wearable. A recent study from the Netherlands' Leiden University, published in the Journal of Psychopathology and Clinical Science, has found that when smartwatches and similar devices record readings on stress, fatigue, or sleep, they're frequently getting it wrong. Researchers studied 800 young adults using the same Garmin Vivosmart 4 smartwatch model. They compared the data the smartwatches produced with the reports that the users created four times per day about how sleepy or stressed they were feeling. Lead author and associate professor Eiko Fried said the correlation between the wearable data and the user-created data was, "basically zero." A representative for Garmin did not immediately respond to a request for comment. Stressed or sex? Your watch doesn't know So why do wearables like fitness smartwatches get it so wrong? Their sensors are fairly limited in what they can do. Watches like these need to be worn correctly at all times (a loose or tight watch may give poor readings, for example), and they typically use basic information like pulse rate and movement to make guesses about health. Those guesses don't always reflect real-world scenarios. A wearable may identify high stress when the real cause of the change was a workout, excitement over good news, or sex. There are so many potential alternatives to stress or fatigue that the watches in the study never really got it right -- and the devices sometimes guessed the complete opposite emotional state from what users recorded. The Dutch study did note that Garmin's Body Battery readings, which specifically measure physical fatigue, were more reliable than stress indicators, but still inaccurate. And sleep sensing performed the best of them all, with Garmin watches showing a two-thirds chance of noting the differences between a good night's sleep and a bad one. It's also worth noting that smartwatch sensors can become more accurate as technology improves. It would be interesting to run a similar study with the much newer Garmin Vivosmart 5 to see if anything has improved, as well as see if other models like the latest versions of the Apple Watch have similar accuracy results.
Gizmodo
an hour ago
- Gizmodo
Doctors Were Worse at Spotting Cancer After Leaning on AI, Study Finds
Artificial intelligence tools have been shown to help doctors detect pre-cancerous growths in the colon—but don't even think about taking those tools away once you've introduced them. A new study published this week in The Lancet found that doctors who are given AI tools to assist with identifying potential cancer risks in patients get worse at making those same observations when they go back to doing it without AI's help. The study looked at four endoscopy centers in Poland, tracking the success rates of detecting colon cancer for three months before AI tools were introduced and three months after. Once AI was introduced, colonoscopies were randomly assigned to either receive AI support or not. The researchers found that doctors who gave colonoscopies without AI after having its assistance available saw their detection rates drop, producing outcomes 20% worse than what they were before AI was introduced. Making the results all the more troubling is the fact that the 19 doctors who participated in the study were all very experienced and had performed more than 2,000 colonoscopies each. If those doctors can fall prey to de-skilling, seeing their own abilities erode because of reliance on AI tools, the outcomes from inexperienced doctors could be even worse. There is little doubt that AI tools can help in medical settings. There have been numerous studies that suggest that AI can facilitate everything from the detection of cancers to the diagnosis of illnesses based on a patient's medical history. Analyzing information based on a whole wealth of previous examples is kinda the bread and butter of AI (you know, as opposed to generating braindead content slop), and there is evidence that suggests that humans can augment their own abilities by using AI tools. Studies in medical settings have found that doctors who use these tools can produce better outcomes for their patients. But no one, including doctors, is immune to the risk of shutting your brain off and relying on AI rather than their own skills. Earlier this year, Microsoft published a study that found knowledge workers who lean on AI stop thinking critically about the work they are doing and feel confident that the assistance from AI will be enough to get the job done. Researchers at MIT similarly found that relying on ChatGPT for essay writing resulted in less critical engagement with the material. In the long term, there's a real risk that reliance on AI will erode our ability to problem solve and reason, which is not ideal when AI continues to generate bad information. The American Medical Association found that about two in three physicians have already adopted AI to augment their abilities. Hopefully, they're still able to identify when it does something like hallucinating a body part that doesn't exist.
