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.'
Hashtags
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
Yahoo
an hour ago
- Yahoo
Scientists Calculate That the Entire Big Bang Must Have Taken Place Inside a Black Hole
The standard model of cosmology may be the best explanation we've got for why the universe is the way it is and how it all came to be. But it's not the only explanation. Enter black hole cosmology. It's a radical idea which proposes that the Big Bang — the rapid unraveling of an infinitely dense point, believed to have given birth to the cosmos as we know it — actually took place in a black hole, which itself formed inside a larger "parent" universe. Ergo, all of us — and every star, planet, galaxy, and internet rando — are living inside one of these mysterious singularities. Enrique Gaztanaga, lead author of a new study published in the journal Physical Review D and a professor at the Institute of Cosmology and Gravitation at the University of Portsmouth, isn't the first to propose this controversial idea. But his team's research offers a new model for imagining how this hypothetical scenario took place. "Our calculations suggest the Big Bang was not the start of everything, but rather the outcome of a gravitational crunch or collapse that formed a very massive black hole — followed by a bounce inside it," Gaztanaga wrote in an essay for The Conversation. Certainly, there are a lot of holes you could poke in the standard model. Why is there more matter than anti-matter, when the universe should be uniform? Why did the universe undergo a period of "cosmic inflation" in which it expanded at faster than light speeds, then stopped? And why does its present day rate of expansion appear to be different depending on how we measure it? Gaztanaga's main gripe seems to be with our current understanding of a singularity. To him, the idea of the universe starting as a point of infinite density is immensely unsatisfying. "This is not just a technical glitch; it's a deep theoretical problem that suggests we don't really understand the beginning at all," he wrote. Gaztanaga also takes aim at other convenient cosmological constructions like dark energy, which is intended to explain why the universe's expansion is mysteriously accelerating. This hypothetical force is thought to make up 68 percent of the universe but is completely unobservable, leaving room for different-minded scientists to call its existence into question. Rethinking singularities could neatly resolve many of these conundrums. We return to Gaztanaga's paper. "Gravitational collapse does not have to end in a singularity," he wrote for The Conversation. "Our maths show that as we approach the potential singularity, the size of the universe changes as a (hyperbolic) function of cosmic time." This is a bold claim. The consensus is that gravitational collapse — like a star imploding into a black hole — must result in an infinitely dense singularity. What Gaztanaga is arguing happens instead is that the collapse not only halt short of completely crushing the matter, but reverses course — a "bounce," in his terminology. "What emerges on the other side of the bounce is a universe remarkably like our own," Gaztanaga explains. "Even more surprisingly, the rebound naturally produces the two separate phases of accelerated expansion — inflation and dark energy — driven not by a hypothetical fields but by the physics of the bounce itself." It's a fascinating explanation, but there's a lot that remains to be proved. It relies on discounting some very well-established physics behind singularities. The standard model may not be perfect, but it's the standard for a reason. It'll take a lot more to dethrone it, and Gaztanaga is optimistic that future missions like the European Space Agency's ARRAKIHS, which will study invisible structures of dark matter to test the model, could reveal the answers we're looking for. More on cosmology: Astronomers Confused to Discover That a Bunch of Nearby Galaxies Are Pointing Directly at Us
Yahoo
an hour ago
- Yahoo
Why Nebius Group N.V. (NBIS) Soared On Thursday
We recently published a list of . In this article, we are going to take a look at where Nebius Group N.V. (NASDAQ:NBIS) stands against other best-performing stocks on Thursday. Nebius Group grew its share prices by 17.54 percent on Thursday to close at $46.30 apiece after raising as much as $1 billion through the issuance of convertible notes. In a statement, Nebius Group N.V. (NASDAQ:NBIS) said that the funds will be raised in two tranches, the first of which amounts to $500,000 covering 2 percent convertible notes due 2029, while the remaining $500,000 will cover 3 percent convertible notes due 2031. A close-up of a businesswoman using a laptop, being illuminated by the AI-enabled cloud interface sponsored by the company. 'The fresh capital we are raising now gives us more firepower to go faster, paving the way for increased revenue opportunities in 2026 and further accelerating us toward our medium-term target of mid-single-digit billions of dollars in revenue as a high-margin business, with potential upside,' said Nebius Group N.V. (NASDAQ:NBIS) founder and CEO Arkady Volozh. 'Building foundational AI infrastructure is a capital-intensive business. In addition to access to the capital markets, we are fortunate to have non-core assets and equity stakes with significant growth profiles that can be used to support the future funding requirements of our core business,' he added. Overall, NBIS ranks 1st on our list of best-performing stocks on Thursday. While we acknowledge the potential of NBIS as an investment, our conviction lies in the belief that some AI stocks hold greater promise for delivering higher returns and have limited downside risk. If you are looking for an extremely cheap AI stock that is also a major beneficiary of Trump tariffs and onshoring, see our free report on the best short-term AI stock. READ NEXT: 20 Best AI Stocks To Buy Now and 30 Best Stocks to Buy Now According to Billionaires. Disclosure: None. This article is originally published at Insider Monkey. Sign in to access your portfolio
Forbes
2 hours ago
- Forbes
Trump Drops A Cybersecurity Bombshell With Biden-Era Policy Reversal
Less than 24 hours after President Trump's public feud with Elon Musk, a new cybersecurity executive order was issued on June 6, 2025, introducing major revisions to the Biden administration's final cybersecurity directives. The order not only modifies key elements of Biden's January 2025 framework but also signals a broader realignment of federal cybersecurity priorities. It shifts focus away from federal digital identity initiatives and revises compliance-heavy software security mandates. Officially titled 'Sustaining Select Efforts To Strengthen The Nation's Cybersecurity And Amending Executive Order 13694 And Executive Order 14144,' the order represents a strategic departure from prior approaches, emphasizing operational pragmatism over regulatory expansion. Notably, it comes at a time when President Trump's nominee to lead the Cybersecurity And Infrastructure Security Agency, Sean Plankey, has yet to be confirmed due to opposition and delay tactics from both sides of the aisle. President Biden's Executive Order 14144 was issued on January 16, 2025, just four days before President Trump's inauguration. It was interpreted by many observers as an effort to define long-term cybersecurity direction before the change in administration. The order included measures to bolster software supply chain security, expand digital identity infrastructure and accelerate post-quantum cryptography adoption. However, this latest Trump order criticized several of these elements as overreaching or insufficiently vetted, characterizing them as 'problematic and distracting' and specifically noting that they were 'sneaked' into policy in the final hours of Biden's presidency. The language used in the accompanying fact sheet is unusually blunt for a federal document, suggesting a clear intent to publicly distance the new administration from its predecessor's policy posture. 1. Attribution Of Threats: Direct Language On Foreign Cyber Aggressors The executive order opens with unusually direct language, identifying the People's Republic of China as the most 'active and persistent' cyber threat to U.S. government systems, private sector networks and critical infrastructure. It also names Russia, Iran and North Korea as continuing sources of malicious cyber activity. This blunt attribution departs from the more generalized threat descriptions of previous administrations. By naming adversaries explicitly in the policy preamble, the administration signals a shift toward greater transparency in threat acknowledgment and a hardening of posture. The message is clear: U.S. cyber strategy is now being framed not only by evolving technologies but by intensifying geopolitical realities. 2. Software Security Compliance: Shifting From Mandated Attestations To Voluntary Implementation: Biden's order imposed a layered framework requiring federal contractors to submit attestations, artifacts and documentation tied to NIST's Secure Software Development Framework. Some would say that these requirements risked turning development teams into compliance teams. Trump's order eliminates attestations entirely. NIST will still provide guidance through the National Cybersecurity Center Of Excellence, but reporting is no longer mandatory. This reflects a shift toward flexibility over formality. 3. Digital Identity Verification: A Full Repeal Rooted In Fiscal And Legal Concerns: The Biden administration had envisioned digital credentials as a gateway to streamlined government services. Trump's order reverses course, citing concerns about entitlement fraud and improper access. The fact sheet explicitly warns that Biden's policy could have enabled unauthorized immigrants to obtain digital IDs. As a result, pilots on interoperability and identity federation are halted. 4. Artificial Intelligence In Cybersecurity: Tighter Focus On Defense And Vulnerability Management: Biden's order encouraged AI-driven collaboration across academia and industry. Trump's order takes a narrower view. It requires agencies to track vulnerabilities in AI systems, integrate them into incident response pipelines and limit data sharing to only what is feasible under security and confidentiality constraints. AI is repositioned as a potential liability to be secured, not a universal defense engine. 5. Post-Quantum Cryptography: A Deadline Remains But The Path Is Streamlined While both administrations agree on the risk posed by quantum computing, Trump's order simplifies the roadmap. By December 2025, CISA and NSA must publish a list of product categories ready for quantum-safe encryption. TLS 1.3 or its successor must be adopted by 2030. Oversight is split between NSA for national security systems and OMB for civilian agencies. 6. Cyber Sanctions Policy: A Narrowed Scope One of the more politically sensitive changes lies in how sanctions are applied. Biden's order allowed for cyber sanctions against any person involved in disinformation or cyber-enabled threats. Trump's revision limits this to foreign persons only. Domestic political activity is explicitly excluded, a move the administration describes as a safeguard against misuse of cyber enforcement tools. Initial industry feedback has been swift. The executive order's reorientation of cybersecurity priorities is already reverberating across the federal ecosystem, private sector and innovation community. From compliance-light procurement to a tighter national focus on AI risk, the changes are reshaping expectations. Defense integrators and established IT vendors are among the most immediate beneficiaries. By removing detailed compliance documentation, particularly attestations tied to secure software development, the order reduces friction in procurement and lowers operational risk. Contract cycles may accelerate as audit-readiness gives way to implementation focus. This shift rewards incumbents with mature delivery models and embedded federal relationships. With CISA's role redefined and federal oversight of digital identity rolled back, state and local governments may gain more autonomy to design cybersecurity programs that fit local contexts. For well-resourced jurisdictions, this could spur innovation. But for others, especially those lacking talent or funding, decentralization could create new coordination gaps. Additional federal guidance may be needed to prevent fragmentation in national critical infrastructure protection. For enterprises, the EO's elimination of standardized compliance frameworks is a mixed bag. Under the previous EO, the bar for secure software delivery was clear, particularly for organizations that invested in transparency and attestation. Without a common benchmark, proving trustworthiness becomes more subjective. Kevin Bocek, CyberArk's Senior Vice President of Innovation, emphasized that the industry is entering a new era of cybersecurity not only dominated by AI and automation, but also by emerging risks that are not yet widely addressed. 'It is affirming that the EO is serious about safe and secure AI, hopefully laying the foundation to critically address one of the most urgent and overlooked threats: machine identity sprawl,' Bocek noted. According to CyberArk, machine identities now outnumber human identities 82 to 1 within enterprises, yet 68% of organizations lack security controls to protect them. Without federal guidance and clear identity accountability, Bocek warns that this vulnerability could become a significant blind spot in national cybersecurity. His comments underscore the risk of prioritizing operational efficiency over foundational security controls, a concern shared by many CISOs facing exponential identity growth from cloud and AI platforms. Digital identity initiatives long supported by privacy advocates, civic technologists and digital modernization leaders were seen as critical to enabling secure, user-friendly access to government services. They aimed to streamline verification, reduce fraud and close equity gaps in federal access. The Biden administration had embraced digital IDs as the backbone of modern digital government. The Trump administration, however, rescinded these efforts. The accompanying fact sheet expressed concerns that digital identity mandates could be exploited to extend entitlements improperly, particularly to unauthorized immigrants. This decision reflects a broader skepticism toward centralized identity infrastructure and a desire to limit the federal government's role in managing citizen-level credentials. The Biden-era policy positioned artificial intelligence as a strategic asset for defense, encouraging public-private collaboration, dataset sharing and predictive threat detection at scale. The Trump administration's new directive narrows that scope significantly. Instead of promoting AI as a systemwide defense multiplier, the EO limits AI's use to managing system vulnerabilities and tracking indicators of compromise. This reflects concerns about over-reliance on technologies that are still evolving, opaque and in some cases unregulated. As Bocek noted, 'Proper AI development is a tool for predictive defense,' but without protections for the AI itself, it could become a new risk vector. The administration's position is clear: AI should be secured before it is scaled. This AI reframing also signals a philosophical divergence between leveraging AI as a force for innovation versus containing it as a potential liability. Whether that caution slows adoption or increases security maturity remains to be seen, but the message is unambiguous: the era of unchecked AI optimism in federal cybersecurity is over. This executive order is not a one-off. It is part of a broader realignment consistent with the principles laid out in Project 2025, a policy blueprint advocating for streamlined federal governance, stronger executive control, and targeted decentralization of agency authority. More orders are expected, particularly in areas such as offensive cyber capabilities, state-level infrastructure resilience, and the restructuring of agencies like CISA. Trump's June 2025 cybersecurity order is more than a policy shift. it is a recalibration of federal cyber strategy that prioritizes execution over oversight, industry collaboration over mandates, and sovereignty over standardization. For industry leaders, innovators, and government stakeholders alike, the takeaway is clear: cybersecurity is no longer just about compliance. It is about preparedness, adaptability, and national competitiveness in an AI-driven world. The next wave of policy will not be about fine-tuning compliance frameworks but will be about defending digital sovereignty. Those who can pivot fastest, and secure what matters most, will shape the next chapter of America's cyber future.
