AI-Radar | Local LLMs, AI Hardware & Trends Observatory

Pavona: un ecosistema open source per il silicio, tra sovranità e certificazione

📁 Hardware AI generated ✅ Phoronix

Pavona: An Open-Source Silicon Ecosystem for Sovereignty and Certification

GlobalPlatform has announced the launch of Pavona, a new open-source silicon ecosystem. The initiative, supported by founding members such as Meta, Qualcomm, and the University of Oxford, aims to provide a certifiable foundation for hardware development. This approach promotes control and sovereignty over components, crucial aspects for on-premise LLM deployments and infrastructures requiring high security and compliance standards.

2026-05-26 📰 Source

NVIDIA Vera: i core Olympus ridefiniscono le performance ARM per i data center

📁 Hardware AI generated ✅ Phoronix

NVIDIA Vera: Olympus Cores Redefine ARM Performance for Data Centers

NVIDIA is set to launch its Vera CPU, an ARM-based data center processor designed for agentic AI workloads. Featuring proprietary Olympus cores, Vera promises competitive performance against Intel and AMD x86_64 CPUs, a significant milestone for ARM architecture in the data center sector. Early benchmarks on Linux suggest a substantial leap forward for on-premise AI infrastructures.

2026-05-26 📰 Source

ReVision Implant si assicura 4 milioni di euro per la protesi visiva corticale

📁 Altro AI generated ℹ️ Tech.eu

ReVision Implant Secures €4M to Advance Cortical Visual Prosthesis

ReVision Implant, a Belgian neurotechnology company, has secured €4 million in funding to accelerate the development of its cortical visual prosthesis. This brain-computer interface aims to restore functional vision in patients with severe blindness by bypassing retinal or optic nerve damage. The investment will support clinical and operational expansion, including the construction of its own cleanroom for greater control over manufacturing processes.

2026-05-26 📰 Source

Wall Street e l'AI: ex-banchieri formano le istituzioni finanziarie

📁 Market AI generated ℹ️ The Next Web

Wall Street and AI: Ex-Bankers Train Financial Institutions

Felipe Sinisterra and Dave Wang, former investment bankers, are capitalizing on the growing adoption of AI on Wall Street. They offer specialized training to financial institutions and funds, teaching senior staff how to leverage AI tools already acquired. With rates reaching $25,000 per day, the two experts are fully booked for the next two months, highlighting a strong demand for practical skills in integrating AI into the financial sector.

2026-05-26 📰 Source

Spotify sperimenta articoli di riviste narrati: implicazioni per l'infrastruttura AI

📁 Market AI generated ℹ️ The Next Web

Spotify Tests Narrated Magazine Articles: Implications for AI Infrastructure

Spotify has launched a test for a new content format: narrated magazine articles, integrated into its audiobook section. This strategic move, involving over 650 pieces from prominent publishers, prompts reflection on the use of AI technologies for audio production and personalization, and the related on-premise or cloud deployment decisions for the underlying infrastructure.

2026-05-26 📰 Source

InfiniteRoots acquisisce Bosque Foods per scalare la produzione industriale di micelio

📁 Market AI generated ℹ️ Tech.eu

InfiniteRoots Acquires Bosque Foods to Scale Industrial Mycelium Production

InfiniteRoots, a German mycelium foodtech company, has acquired Bosque Foods. The acquisition aims to integrate Bosque's expertise in solid-state fermentation and whole-cut product development, strengthening InfiniteRoots' platform. The goal is to accelerate the scalability and industrial reproducibility of mycelium applications, a sector moving beyond the experimental phase towards industrialization.

2026-05-26 📰 Source

Magnifica Humanitas: l'enciclica di Papa Leone XIV e la regolamentazione dell'AI

📁 Market AI generated ℹ️ The Next Web

Magnifica Humanitas: Pope Leo XIV's Encyclical on AI Regulation

Pope Leo XIV has published his first encyclical, Magnifica Humanitas, which is being read not only as a theological text but also as an analysis of technological regulation. The document directly addresses governments, parliaments, and the executives of major artificial intelligence companies, highlighting the growing importance of policy within the AI ecosystem.

2026-05-26 📰 Source

Quantinuum di Honeywell: IPO mira a 12,7 miliardi di dollari, sotto le attese iniziali

📁 Market AI generated ℹ️ The Next Web

Honeywell's Quantinuum Targets $12.7 Billion IPO, Below Initial Expectations

Quantinuum, the quantum computing company majority-owned by Honeywell, is targeting a $12.7 billion valuation for its US IPO. This figure is significantly lower than the initial estimates of over $20 billion that circulated in May. The offering still represents a 27% increase over the $10 billion pre-money valuation.

2026-05-26 📰 Source

L'interesse crescente per i Large Language Models on-premise: una discussione chiave

📁 Altro AI generated ℹ️ LocalLLaMA

The Growing Interest in On-Premise Large Language Models: A Key Discussion

The tech community is showing increasing interest in the local deployment of Large Language Models. This trend raises crucial questions related to data sovereignty, infrastructural control, and cost optimization, pushing companies to evaluate self-hosted alternatives to cloud solutions.

2026-05-26 📰 Source

Reclamare la 'Social Engineering': Controllo e Sovranità nell'Era Digitale

📁 Altro AI generated 🏆 IEEE Spectrum

Reclaiming 'Social Engineering': Control and Sovereignty in the Digital Age

The concept of "social engineering," originally intended to shape human behavior for beneficial purposes, has been distorted by authoritarian regimes and, today, by proprietary algorithms influencing our decisions. This article explores its evolution, highlighting how modern digital manipulation is often invisible and beyond control. For tech decision-makers, understanding these dynamics is crucial to ensure transparency, data sovereignty, and control over AI systems, especially in on-premise deployments.

2026-05-26 📰 Source

SK Hynix presenta iHBM: raffreddamento integrato per le memorie AI di nuova generazione

📁 Hardware AI generated ℹ️ Tom's Hardware

SK Hynix Unveils iHBM: Integrated Cooling for Next-Gen AI Memory

SK Hynix has introduced iHBM, a new thermal architecture that integrates cooling elements directly into the HBM memory interface. This innovation aims to reduce thermal resistance by 30%, enhancing efficiency for HBM5 accelerators and high-density AI data centers. The solution addresses critical thermal challenges for on-premise Large Language Model deployments.

2026-05-26 📰 Source

Hacker iraniani violano i sistemi di controllo della LA Metro

📁 Altro AI generated ℹ️ The Next Web

Iranian Hackers Breach LA Metro Control Systems

Research by Gambit Security has revealed that Iran-linked hackers were responsible for a cyber-attack targeting the Los Angeles County Metropolitan Transportation Authority in March. The operation compromised parts of the infrastructure, including rail-yard control displays, and led to the theft of 700 gigabytes of sensitive data, including emails and backups, traced back to a server previously associated with Iranian entities.

2026-05-26 📰 Source

SpaceX chiede al Pentagono un aumento del costo per i terminali Starlink in Iran

📁 Market AI generated ℹ️ The Next Web

SpaceX Requests Price Increase for Starlink Terminals in Iran

SpaceX has requested a significant price increase from the US Department of Defense for Starlink terminals, proposing a jump from $5,000 to $25,000 per terminal. This request is driven by the network's crucial role in guiding American strike drones operating over Iran, a service that SpaceX believes is currently undervalued compared to its actual worth in a critical operational context.

2026-05-26 📰 Source

BNP Paribas e Mistral AI: una risposta europea alla cybersecurity con LLM

📁 Altro AI generated ℹ️ The Next Web

BNP Paribas and Mistral AI Partner for European AI Cybersecurity Solution

BNP Paribas, the Eurozone's largest bank, is collaborating with Mistral AI to develop AI-powered cybersecurity tools. This initiative aims to create a European solution addressing data sovereignty and control needs, contrasting with the difficulties European regulators face in reliably accessing similar US-origin tools, which are already utilized by American counterparts.

2026-05-26 📰 Source

L'AI e il futuro del lavoro: le scelte infrastrutturali per le aziende

📁 Altro AI generated ✅ Wired AI

AI and the Future of Work: Infrastructure Choices for Enterprises

The advancement of artificial intelligence is redefining professional roles, compelling companies to rethink their strategies. For tech leaders, understanding the implications of LLM deployment, from choosing between cloud and on-premise to managing data sovereignty and TCO, is crucial for maintaining competitiveness and driving innovation.

2026-05-26 📰 Source

Uber frena sulla spesa AI: il CEO avverte sui limiti del 'tokenmaxxing' e il valore di prodotto

📁 Market AI generated ℹ️ Tom's Hardware

Uber Slows AI Spending: CEO Questions Link Between 'Tokenmaxxing' and Product Success

Uber is re-evaluating its artificial intelligence investment strategy. Its chief has voiced concerns about the direct correlation between aggressive AI resource optimization, termed 'tokenmaxxing,' and the actual ability to deliver successful products. This caution highlights the industry's growing focus on return on investment and the practical application of LLMs, a critical consideration for those evaluating on-premise or cloud deployments.

2026-05-26 📰 Source

Pony AI accelera: obiettivo flotta robotaxi 2026 rivisto al rialzo dopo un Q1 eccezionale

📁 Market AI generated ℹ️ The Next Web

Pony AI Accelerates: 2026 Robotaxi Fleet Goal Revised Upward After Exceptional Q1

Pony AI, a Chinese autonomous driving company, has raised its year-end 2026 robotaxi fleet target to over 3,500 units, up from the previously projected 3,000. This decision follows a 395% year-on-year increase in robotaxi revenues in the first quarter of 2026, with the current fleet already exceeding 1,700 vehicles. The announcement underscores the rapid expansion of the sector.

2026-05-26 📰 Source

TSMC: malumore tra i dipendenti per i bonus, tagli in vista nonostante i ricavi record spinti dall'AI

📁 Market AI generated ℹ️ Tom's Hardware

TSMC: Employee Discontent Over Bonuses, Cuts Loom Despite Record AI-Driven Revenues

A report indicates growing discontent among TSMC employees, with potential strikes and unionization being considered. The cause stems from bonuses: the company is reportedly considering a 15% cut to fund capital expenditures. This move comes despite TSMC achieving record revenues, driven by strong demand in the artificial intelligence sector, highlighting tensions between human resource management and strategic investments.

2026-05-26 📰 Source

Verifica dei fatti e LLM: l'AI sbaglia più spesso di quanto si pensi?

📁 LLM AI generated ✅ Wired AI

Fact-Checking and LLMs: Is AI Wrong More Often Than You Think?

A professional fact-checker from WIRED has examined the capabilities of LLMs in fact-checking, raising doubts about their accuracy. This article explores the technical challenges and implications for companies considering adopting AI-based solutions for critical tasks, highlighting how, despite advancements, Large Language Models can still present significant errors. It analyzes the trade-offs between automation and reliability, a crucial aspect for on-premise deployments where data control and accuracy are paramount.

2026-05-26 📰 Source

La Cina estende le restrizioni di viaggio ai ricercatori AI del settore privato

📁 Altro AI generated ℹ️ The Next Web

China expands travel curbs for private AI researchers

China is broadening its informal travel restrictions on senior artificial intelligence researchers. Following initial measures that involved DeepSeek, requests for passport surrender now extend to a wider group of private-sector AI firms, signaling increased state control over technological talent and the mobility of key professionals.

2026-05-26 📰 Source

SkillOpt: Ottimizzare le 'Skill' degli LLM con File Markdown Addestrabili

📁 LLM AI generated ℹ️ LocalLLaMA

SkillOpt: Optimizing LLM Skills with Trainable Markdown Files

Recent research introduces SkillOpt, an approach that treats Markdown files defining LLM agent 'skills' as trainable parameters. By using a frontier model to propose bounded edits and a validation set to accept only strict improvements, the method optimizes LLM capabilities for specific tasks. The methodology has shown cross-model transferability and significant improvements in procedural benchmarks, though it requires an automated grading system with clear answers, limiting its application to open-ended contexts.

2026-05-26 📰 Source

Padroneggiare l'AI: Competenze Essenziali per l'Enterprise e i Deployment On-Premise

📁 LLM AI generated ✅ Wired AI

Mastering AI: Essential Skills for Enterprise and On-Premise Deployments

The era of artificial intelligence demands deep expertise from tech professionals. From effectively managing Large Language Models to optimizing interactions, acquiring an "AI-native" mindset is crucial for addressing the challenges of enterprise deployments, especially self-hosted ones, ensuring control and data sovereignty.

2026-05-26 📰 Source

L'etica nell'AI: i laboratori assumono filosofi per dilemmi complessi

📁 Altro AI generated ✅ Wired AI

AI Ethics: Labs Hire Philosophers for Complex Dilemmas

Leading artificial intelligence laboratories are integrating philosophers into their teams to address ethical edge cases and fundamental questions of mind and morality. This trend raises questions about the true depth of commitment to responsible AI or if it's merely a marketing tool, but highlights the growing complexity of challenges related to LLM deployment.

2026-05-26 📰 Source

L'IA e il recupero crediti: l'automazione delle chiamate più 'odiate'

📁 Market AI generated ✅ Wired AI

AI and Debt Collection: Automating the 'Most Hated' Calls

Artificial intelligence is rapidly entering a traditionally challenging sector: debt collection. The trend points towards automating the most 'unwanted' calls, with AI poised to manage interactions with debtors. This evolution raises crucial questions about technological implications, data sovereignty, and deployment choices for companies adopting these solutions.

2026-05-26 📰 Source

L'AI e il futuro delle carriere: le scelte infrastrutturali che definiscono il domani

📁 Altro AI generated ✅ Wired AI

AI and the Future of Careers: Infrastructure Choices Shaping Tomorrow

The advancement of artificial intelligence is reshaping the professional landscape, leading to the transformation of some roles and the emergence of new opportunities. For businesses, understanding AI's impact on careers also means carefully evaluating deployment strategies and underlying infrastructures, balancing control, data sovereignty, and TCO. Today's technological decisions will directly influence an organization's ability to adapt and thrive in the AI era.

2026-05-26 📰 Source

Intel introduce pmtctl: un nuovo strumento per la telemetria su Linux

📁 Hardware AI generated ✅ Phoronix

Intel Introduces pmtctl: A New Tool for Linux Platform Telemetry

Intel has announced the development of pmtctl, a new Linux kernel tool designed to interface with its Platform Monitoring Technology. This tool will enable more granular management of platform telemetry data, offering IT professionals greater control and visibility over underlying hardware, a crucial aspect for on-premise deployments and efficient resource management.

2026-05-26 📰 Source

Il 99% dei CEO prevede tagli al personale guidati dall'AI: incertezza sul ROI

📁 Market AI generated ℹ️ Tom's Hardware

99% of CEOs Expect AI-Driven Layoffs: Uncertainty on ROI

A recent survey indicates that almost all CEOs anticipate AI-driven layoffs, as companies accelerate the replacement of junior roles. Despite this rush to adoption, many executives express uncertainty regarding the actual returns on AI investments, highlighting a tension between the imperative for efficiency and the need for a clear, measurable deployment strategy.

2026-05-26 📰 Source

Ennostar punta sugli interconnessioni ottiche: un pilastro per l'AI on-premise

📁 Hardware AI generated ✅ DigiTimes

Ennostar Pivots to Optical Interconnects: A Pillar for On-Premise AI

Ennostar, an LED manufacturer, is strategically reorienting its focus towards optical interconnect solutions. This move reflects a growing demand for high-speed, low-latency infrastructure, which is essential for artificial intelligence workloads, particularly for on-premise deployments requiring scalability and data control.

2026-05-26 📰 Source

Otomato raccoglie 2 milioni di dollari per l'intelligenza DeFi in tempo reale

📁 Market AI generated ℹ️ Tech.eu

Otomato Raises $2M for Real-Time DeFi Intelligence

Otomato, a Swiss DeFi intelligence company, has secured $2 million in funding from Improbable. The platform aims to solve the complex problem of managing DeFi positions by real-time monitoring on-chain activities across multiple blockchains and providing users with only relevant alerts, cutting through the noise of generic notifications.

2026-05-26 📰 Source

AI ed etica: Anthropic e il Vaticano a confronto sui limiti morali

📁 Altro AI generated ✅ DigiTimes

AI and Ethics: Anthropic and the Vatican Discuss Moral Limits

A significant meeting took place between Christopher Olah, co-founder of Anthropic, and Pope Leo XI. The discussion focused on the moral perils of artificial intelligence, with a call to scripture to outline the ethical boundaries of technological development. The event highlights the growing attention towards responsible AI governance, a crucial topic for companies evaluating Large Language Models deployment and data sovereignty.

2026-05-26 📰 Source

Darfon Electronics investe in GrAndvance: focus sui sistemi di raffreddamento per l'AI

📁 Hardware AI generated ✅ DigiTimes

Darfon Electronics Invests in GrAndvance: Focus on AI Cooling Systems

Darfon Electronics has announced a strategic investment in GrAndvance, a company specializing in cooling systems for artificial intelligence. This move highlights the increasing importance of advanced thermal solutions to manage the heat generated by AI workloads, particularly in on-premise deployments. Efficient cooling is crucial for the stability and performance of AI infrastructures, directly impacting TCO and operational sustainability.

2026-05-26 📰 Source

L'allarme delle autorità USA: l'estremismo anti-tecnicico minaccia l'adozione dell'AI

📁 Altro AI generated ✅ Wired AI

US Law Enforcement Warns of ‘Anti-Tech Extremism’ as AI Hatred Grows

US law enforcement agencies have issued a warning about a new category of threat: anti-tech extremism. This concern arises amidst growing public resentment towards artificial intelligence, perceived as a job threat, and the proliferation of data centers. Documents obtained by WIRED reveal how federal authorities are closely monitoring this phenomenon.

2026-05-26 📰 Source

L'investimento da 16,5 miliardi: l'ottica co-packaged trasformerà l'infrastruttura AI entro il 2030

📁 Hardware AI generated ✅ DigiTimes

The US$16.5 Billion Bet: How Co-Packaged Optics Will Rewire AI's Backbone by 2030

A US$16.5 billion investment by 2030 is set to redefine AI infrastructure through the adoption of co-packaged optics (CPO). This technology, integrating optical components directly into chips, promises to overcome the limitations of traditional electrical interconnects regarding bandwidth, power consumption, and latency. It will be crucial for large-scale LLM deployments, especially in self-hosted environments, significantly impacting TCO and data sovereignty.

2026-05-26 📰 Source

Nvidia e Marvell: da rivali a partner, un'alleanza da 2 miliardi di dollari a Computex

📁 Market AI generated ✅ DigiTimes

Nvidia and Marvell: From Rivals to Partners, a $2 Billion Alliance at Computex

Nvidia and Marvell CEOs will share the Computex stage, marking a transition from rivalry to partnership. This alliance, valued at $2 billion, highlights the evolving hardware landscape for AI. For companies evaluating on-premise solutions, such collaborations are crucial for innovation and diversifying infrastructure options, influencing TCO and data sovereignty.

2026-05-26 📰 Source

ASIC vs. GPU: Alchip e la svolta nell'hardware AI per i deployment on-premise

📁 Hardware AI generated ✅ DigiTimes

ASIC vs. GPU: Alchip and the Shift in AI Hardware for On-Premise Deployments

According to Alchip, the market for Application-Specific Integrated Circuits (ASICs) in artificial intelligence is poised to outpace the growth of the broader GPU sector. This forecast highlights a potential strategic shift for companies evaluating AI computing solutions, particularly for on-premise deployments, where efficiency, TCO, and data sovereignty are critical.

2026-05-26 📰 Source

Tonada emerge dal silenzio: l'IA generativa reinventa la musica d'ambiente per il retail

📁 Market AI generated ℹ️ Tech.eu

Tonada Emerges from Stealth: AI-Generated Music Redefines Retail Ambiance

Tonada, a Swedish AI-generated music platform, has emerged from stealth with backing from prominent figures including Vercel's CEO and Last.fm's co-founder. The company aims to transform background music for retail and hospitality by offering unique, royalty-free catalogs that adapt in real-time to physical spaces.

2026-05-26 📰 Source

Nexus Luxembourg 2026: L'agenda che definisce il futuro tech europeo

📁 Market AI generated ℹ️ Tech.eu

Nexus Luxembourg 2026: The Agenda Defining Europe's Tech Future

Nexus Luxembourg 2026 is set to host 10,000 innovators for a strategic briefing on AI integration and technological autonomy. The newly released agenda promises deep dives into scale, regulation, and impact, featuring over 150 experts and four dedicated thematic zones, including a focus on digital sovereignty. The event represents a crucial opportunity for European leaders who wish to be at the forefront of discussions shaping the continent's technological future.

2026-05-26 📰 Source

Lucis raccoglie 20 milioni di dollari per la sua piattaforma di salute preventiva basata su AI

📁 Market AI generated ℹ️ The Next Web

Lucis Secures $20M Series A for AI-Driven Preventive Health Platform

Parisian startup Lucis has successfully closed a $20 million Series A funding round, led by Singular. The platform integrates blood biomarker testing with an AI-powered companion application for preventive health. This significant investment comes just four months after the company's seed round, accelerating the development of its digital health offerings.

2026-05-26 📰 Source

Qwen3.5 27B: un LLM versatile per deployment on-premise con MTP preservati

📁 LLM AI generated ℹ️ LocalLLaMA

Qwen3.5 27B: A Versatile LLM for On-Premise Deployments with Preserved MTPs

Qwen3.5 27B, a Large Language Model optimized for general AI assistance, has been released, maintaining its full 15 Multi-Turn Preservation (MTP) capabilities. Available in various formats such as Safetensors, GGUFs, NVFP4, and GPTQ-Int4, the model is designed to facilitate self-hosted deployments and offers significant resilience to "abliteration," distinguishing itself with robustness compared to newer versions.

2026-05-26 📰 Source

Sam Altman di OpenAI ridimensiona i timori di un'apocalisse lavorativa da IA

📁 Market AI generated ℹ️ The Next Web

OpenAI's Sam Altman Downplays AI Job Apocalypse Fears

Sam Altman, OpenAI's CEO, has moderated his previous predictions regarding a massive employment collapse caused by artificial intelligence. Speaking in the Asia-Pacific region, Altman stated that current data does not support the hypothesis of a widespread "job apocalypse" due to AI. This perspective suggests an evolution of the job market rather than destruction, with implications for AI adoption strategies and skill development within enterprises, especially for those considering on-premise deployments.

2026-05-26 📰 Source

GSR Ventures raccoglie 350 milioni per un nuovo fondo cinese, puntando su RedNote

📁 Market AI generated ℹ️ The Next Web

GSR Ventures Raises $350M for New China Fund, Anchoring on RedNote Stake

GSR Ventures Management Co., an early backer of the Chinese social commerce app Xiaohonghu (internationally known as RedNote), is raising approximately $350 million. The funds are earmarked for a new China-focused venture fund, with the existing stake in RedNote serving as the central pitch. This information comes from a Bloomberg report.

2026-05-26 📰 Source

Strix Halo e LLM MoE: un PR rifiutato sblocca fino al 31% di performance in llama.cpp

📁 Hardware AI generated ℹ️ LocalLLaMA

Strix Halo and MoE LLMs: A Rejected PR Unlocks Up to 31% Performance in llama.cpp

A Pull Request for `llama.cpp`, though rejected, offers a significant performance boost for Large Language Models (LLMs) based on the Mixture of Experts (MoE) architecture. Users of AMD Strix Halo hardware can achieve up to 31% faster tokens per second, especially with smaller context windows. This manually applicable modification highlights optimization opportunities for on-premise LLM deployments.

2026-05-26 📰 Source

InPsytech e l'era dei chiplet: connettività AI al centro dell'attenzione

📁 Hardware AI generated ✅ DigiTimes

InPsytech and the Chiplet Era: AI Connectivity in the Spotlight

The rise of chiplet technology is highlighting the crucial importance of advanced connectivity solutions for artificial intelligence. In this scenario, Taiwanese company InPsytech positions itself as a key player, thanks to its specialized intellectual property (IP) in AI connectivity. Its focus addresses critical needs for efficient data transfer and processing within complex chiplet architectures, which are fundamental for next-generation AI workloads.

2026-05-26 📰 Source

Quanscient ottiene 10 milioni per la piattaforma di simulazione AI e quantum-native

📁 Market AI generated ℹ️ The Next Web

Quanscient Secures €10M for AI and Quantum-Native Simulation Platform

Finnish startup Quanscient has closed a €10 million Series A funding round. The capital will scale its cloud-based multiphysics simulation platform, which aims to reinvent physics simulation as the data engine for AI-driven hardware design, integrating AI and quantum-native capabilities. This funding will support expansion and the development of new frontiers in engineering design.

2026-05-26 📰 Source

LLM on-premise: il TCO non è l'unico fattore, la questione è il controllo

📁 Altro AI generated ℹ️ LocalLLaMA

On-Premise LLMs: TCO Isn't the Only Factor, Control Is Key

A detailed analysis challenges the common perception that self-hosting Large Language Models is always cheaper than cloud solutions. While per-token costs might favor cloud for intermittent workloads, the true motivations for on-premise deployment lie in data sovereignty, privacy, total infrastructure control, and technical learning opportunities.

2026-05-26 📰 Source

Bruxelles prepara multa record per Google sul Digital Markets Act

📁 Market AI generated ℹ️ The Next Web

Brussels Prepares Record DMA Penalty for Google

The European Commission is set to impose a fine of hundreds of millions of euros on Google for breaching the Digital Markets Act. This would be the highest penalty ever issued under the EU's new tech competition regime, based on a long-running complaint. This event highlights the increasing focus of regulatory authorities on the practices of large technology companies and their strategic implications for data sovereignty.

2026-05-26 📰 Source

Qwen3.5 35B A3B: Un Nuovo LLM per Usi Generali, Ottimizzato per Deployment Locali

📁 LLM AI generated ℹ️ LocalLLaMA

Qwen3.5 35B A3B: A New General-Purpose LLM Optimized for Local Deployments

The Qwen3.5 35B A3B model, developed by llmfan46, is now available in various configurations optimized for inference on local hardware, including GGUF and GPTQ-Int4 formats. This LLM, which preserves 785 MTPs, stands out for its `qwen35` architecture and its focus on general-purpose AI assistance, differentiating itself from Qwen3.6, which is more geared towards agentic and coding tasks. Its availability in quantized formats makes it particularly appealing for on-premise scenarios.

2026-05-26 📰 Source

Wonderful Hi-Tech: la domanda cresce tra satelliti e data center AI

📁 Market AI generated ✅ DigiTimes

Wonderful Hi-Tech Sees Rising Demand from Satellites and AI Data Centers

Wonderful Hi-Tech is experiencing a significant surge in demand, driven by the expansion of AI-dedicated data centers and low-orbit satellite connectivity projects. This trend highlights the growing need for robust and high-performance infrastructure, emphasizing strategic decisions regarding on-premise deployment and the management of LLM workloads.

2026-05-26 📰 Source

Lucis raccoglie 20 milioni di dollari per espandere la sua piattaforma di prevenzione sanitaria basata sull'AI

📁 Market AI generated ℹ️ Tech.eu

Lucis Secures $20M Series A to Advance AI-Driven Preventive Healthcare Platform

Lucis, an AI-powered preventive healthcare platform, has closed a $20 million Series A funding round, bringing its total funding to $28 million. The company analyzes over 110 blood biomarkers to provide personalized guidance, showing promising results in health improvement and user engagement. Lucis plans international expansion, highlighting the critical role of AI in proactive healthcare.

2026-05-26 📰 Source

Linux si prepara alla crittografia post-quantistica con ML-KEM e X-Wing

📁 Altro AI generated ✅ Phoronix

Linux Prepares for Post-Quantum Cryptography with ML-KEM and X-Wing

Eric Biggers from Google has released a set of patches for the Linux kernel, introducing proof-of-concept support for the ML-KEM and X-Wing cryptographic algorithms. This initiative aims to strengthen system security in anticipation of the quantum computing era, addressing potential future vulnerabilities for data and communication protection.

2026-05-26 📰 Source

Wiwynn espande la capacità produttiva negli USA: la corsa ai server AI mette sotto pressione energia e supply chain

📁 Market AI generated ✅ DigiTimes

Wiwynn Expands US Capacity Amid AI Server Boom, Straining Power and Supply Chains

Wiwynn, a key player in the server market, is increasing its production capacity in the United States. This expansion responds to the surging demand for AI-dedicated servers, a rapidly growing sector that is placing significant strain on energy infrastructures and global supply chains. The move highlights the challenges companies face in building and maintaining their AI infrastructures.

2026-05-26 📰 Source

Kinpo Electronics: l'espansione strategica tra server AI e nuove frontiere tecniciche

📁 Market AI generated ✅ DigiTimes

Kinpo Electronics: Strategic Expansion into AI Servers and New Technological Frontiers

Kinpo Electronics is redefining its core business, moving beyond traditional manufacturing. The company is focusing on high-value sectors such as artificial intelligence servers, satellite systems, and quantum computing. This strategic move aims to capitalize on growing opportunities in advanced technological fields, offering crucial infrastructural solutions for AI workload deployment, with implications for those evaluating on-premise options and data sovereignty.

2026-05-26 📰 Source

LLM on-premise: la ricerca della configurazione universale per il deployment locale

📁 Altro AI generated ℹ️ LocalLLaMA

On-Premise LLMs: The Quest for Universal Local Deployment Configuration

The `LocalLLaMA` community highlights the growing need to simplify Large Language Model deployment on local hardware. The concept of a universal "letter" symbolizes the quest for a standardized configuration that can smooth out technical complexities, offering greater control and data sovereignty for enterprises evaluating self-hosted solutions.

2026-05-26 📰 Source

Avrea raccoglie 4,7 milioni per ridefinire il CI/CD nell'era dell'AI generativa

📁 Frameworks AI generated ℹ️ Tech.eu

Avrea Secures $4.7 Million to Redefine CI/CD in the Generative AI Era

Avrea has secured $4.7 million in pre-seed funding for its Continuous Integration platform. The goal is to modernize CI/CD processes, which are struggling to keep pace with AI-driven code generation. Avrea's solution integrates with existing workflows and allows AI agents to natively participate in development, improving delivery speed and pipeline observability.

2026-05-26 📰 Source

Quanscient raccoglie 10 milioni di euro per l'ingegneria hardware basata su AI e quantistica

📁 Market AI generated ℹ️ Tech.eu

Quanscient Secures €10M to Advance AI and Quantum-Native Hardware Engineering

Quanscient, a Finnish company specializing in cloud-based multiphysics simulation and quantum algorithms, has raised €10 million in Series A funding. The investment aims to support international expansion and enhance its capabilities in simulation, quantum computing, and artificial intelligence, addressing traditional hardware engineering challenges and accelerating digital product development.

2026-05-26 📰 Source

AMD e Nvidia rafforzano gli investimenti nell'ecosistema dei semiconduttori di Taiwan

📁 Market AI generated ✅ DigiTimes

AMD and Nvidia Deepen Investments in Taiwan's Semiconductor Ecosystem

AMD and Nvidia are increasing their investments in Taiwan's semiconductor ecosystem. This strategic move highlights the island's central role in advanced chip manufacturing, which is crucial for the development and deployment of Large Language Models (LLM) and other artificial intelligence applications. The implications concern supply chain stability and the availability of hardware for on-premise AI infrastructures.

2026-05-26 📰 Source

Sistema di Dibattito Multi-Persona: LLM per Ipotesi Scientifiche Automatizzate

📁 LLM AI generated 🏆 ArXiv cs.CL

Multi-Persona Debate System: LLMs for Automated Scientific Hypothesis Generation

The Multi-Persona Debate System (MPDS) is a new framework leveraging Large Language Models to generate automated scientific hypotheses, overcoming limitations in synthesizing fragmented knowledge. Particularly useful in battery materials research, MPDS combines literature retrieval, long-context LLM reasoning, and structured multi-agent debate, based on scientific paper "snapshots." The system improves complex hypothesis formation, offering a reusable workflow for text-intensive scientific discovery.

2026-05-26 📰 Source

Raon-Speech e Raon-SpeechChat: LLM open source per la comprensione e generazione vocale

📁 LLM AI generated 🏆 ArXiv cs.CL

Raon-Speech and Raon-SpeechChat: Open-Source LLMs for Speech Understanding and Generation

Raon-Speech and Raon-SpeechChat, two 9-billion-parameter speech language models (SpeechLMs), have been introduced. Raon-Speech excels in English and Korean speech understanding and generation while retaining strong text capabilities. Raon-SpeechChat extends these functionalities to natural real-time full-duplex conversation. Both models, along with their training and inference pipelines, are open-sourced, offering new opportunities for on-premise deployments and autonomous data management.

2026-05-26 📰 Source

Rilevamento Anomalie Industriali: Un Nuovo Framework per l'Edge AI Distribuita

📁 Altro AI generated 🏆 ArXiv cs.LG

Industrial Anomaly Detection: A New Framework for Distributed Edge AI

A new framework, MODIAD, addresses challenges in multimodal industrial anomaly detection within distributed and online environments. Leveraging edge intelligence, MODIAD introduces a Multi-class Intelligent Scheduling (MIS) problem and an SMG algorithm for resource-constrained training. The REC-LoRA strategy enhances computational and communication efficiency, reducing system overhead. Validated on industrial datasets, MODIAD offers superior performance and efficiency, with significant implications for on-premise deployments and data sovereignty.

2026-05-26 📰 Source

Algometrics: Valutare i Modelli Predittivi in Mercati Algoritmici

📁 Market AI generated 🏆 ArXiv cs.LG

Algometrics: Evaluating Predictive Models in Algorithmic Markets

The new "algometrics" framework offers an approach to analyze time series where predictive models influence the data they aim to forecast. It distinguishes historical risk from deployment risk, highlighting how passive evaluation is insufficient. The framework suggests integrating feedback sensitivity into benchmarks for a more accurate assessment of operational risk, crucial for those deploying LLMs and other AI systems in dynamic environments.

2026-05-26 📰 Source

AI-Radar - Local LLMs, AI Hardware and Trends Observatory

AI-Radar for on-prem LLMs & Home AI

The Daily Signal

Presentiamo AI-Portable, il sito gemello di AI-Radar

⚡ Trending Now

Latest Analysis & Radar News