➀ A German-Danish research group is working on a novel freshness sensor for meat and fish in the PRECISE project. The researchers aim to determine the shelf life of meat and fish using a new 'Electronic Nose' and thus predict precisely to prevent tons of food from being wasted. Experts from TH Lübeck are part of the research project and are currently testing and optimizing the freshness sensor at the CoSA Competence Center. The researchers are funded with 1.885.263 Euro by the EU Interreg Germany-Denmark program until March 31, 2026.

➁ The current methods to determine if meat or fish is still edible involve trained personnel using their nose to check for freshness and expensive, time-consuming microbiological analyses that take days to produce results. What is lacking is a technical solution that can accurately and quickly determine and predict how long meat and fish can be stored.

➂ The project aims to save up to 25,000 tons of meat and fish from waste. The team, led by Roana de Oliveira Hansen from SDU, brings experience in sensor development. The CoSA team at TH Lübeck contributes to the prediction of shelf life. The Fraunhofer ISIT in Itzehoe produces the piezoelectric cantilever sensors, while the Danish startup AmiNIC brings expertise in prototyping and software updates based on partner data.

➀ Yole Group forecasts the Panel Level Packaging (PLP) market to grow from $160 million to $600 million by 2030 with a CAGR of 27%;

➁ High-density fan-out and ultra-density platforms will dominate the market by 2030, driven by AI;

➂ Samsung Electronics leads the market in 2024, followed by ST, with PTI, SIPLP, and ASE holding less than 10% market share.

➀ Keysight Technologies has introduced KAI Data Center Builder, a new software suite designed to emulate real-world AI training workloads, aiding in assessing the impact of changes to algorithms, components, and protocols on AI system performance.

➁ The tool helps align hardware design with training algorithms, improving overall system performance by integrating training workloads from large language models (LLMs) and other AI models.

➂ KAI Data Center Builder allows AI operators, cloud providers, and infrastructure vendors to test realistic workloads in lab environments, validate new AI cluster designs, refine partitioning strategies, and optimize performance before full deployment.

➀ The dilemma facing Big Tech is the cost of building AI, with estimates ranging from tens of millions to tens of billions;

➁ Some investments in cheap, effective Chinese products have raised concerns about the wisdom of some investments;

➂ Microsoft is withdrawing from data center projects in various countries, and expectations for GenAI are declining due to high failure rates and dissatisfaction with current results;

➃ OpenAI is not yet profitable and expects positive cash flow only in 2029, despite spending $644 billion on AI programs and data centers.

➀ The Deepseek-R1 671B model has captured industry attention for its advanced reasoning capabilities;

➁ The challenge of running this model is its high memory requirements, which can be addressed through quantization;

➂ This article discusses a solution to run the model on an AMD Volcano platform with virtualized workloads, showcasing the use of Docker and Open WebUI for deployment.

➀ The NVIDIA GeForce RTX 5090 has become scarce and hard to find, but not for everyone as an AI farm in Vietnam demonstrates. The farm showcases a large number of these rare GPUs, reminiscent of the crypto mining farms seen during GPU shortages years ago.

➁ The GeForce RTX 5090, the best gaming GPU on the market, is also well-suited for AI and machine learning applications due to its impressive 32GB of GDDR7 VRAM.

➂ Despite its high price and limited availability, the RTX 5090 is still in high demand for AI work, and the market conditions for GPUs remain challenging.

➀ U.S. President Donald Trump has signed an executive order to create the United States Investment Accelerator Office;

➁ The office aims to negotiate better deals for taxpayers;

➂ There are concerns about the impact on existing contracts and awards.

➀ The article discusses the challenges and opportunities for Europe in the global AI competition.

➁ It mentions the large-scale investment in AI infrastructure in the US and the emergence of cost-effective language models from China.

➂ The EU's response with the 'InvestAI' initiative and the 200 billion Euro investment package is highlighted, along with the EU Commission's plan to invest 1.3 billion euros in AI by 2027.

➀ Battery production is crucial for industrial and climate policy, with a growing demand for energy storage systems. Challenges include raw material dependence and supply chain fragility.

➁ Advanced laser technology offers solutions for efficient, precise, and sustainable battery production, addressing issues in material processing, electrode production, and recycling.

➂ Innovations in electrode production, cell assembly, and module/pack production are highlighted, along with the integration of AI for process optimization and predictive maintenance.

➀ This article discusses a reference design for secure telehealth solutions, offering a robust framework for building advanced connected medical devices.

➁ The design features a WBZ350 module with Bluetooth LE 5.2 capability, ensuring efficient communication and enhanced security with the ATECC608C cryptographic co-processor.

➂ The design supports multiple connectivity options, including AWS or Azure cloud connectivity, and also includes bio-signal and temperature sensors for accurate health monitoring.

➀ The MLPerf Inference v5.0 results have been released, showing that NVIDIA Hopper systems dominate the rankings;

➁ Intel and AMD also have their own submissions, focusing on different market segments;

➂ The article discusses the performance of various systems and the focus on Deepseek-R1 671B inference performance.

➀ Cover Story: Quantum Chips - The latest developments in quantum computing.

➁ Future: How AI will transform everyday life.

➂ Strategy: Global chip race shifts, spotlight moment for India?

➀ This article explores the implementation of an IoT-based health monitoring system using the statistical programming language R. The system aims to process and analyze real-time health data from various IoT devices, with a focus on predicting heart attack risks.

➁ The proposed system integrates wearable sensors, mobile applications, and cloud-based infrastructure to connect individuals with healthcare providers. R's analytical capabilities, including statistical functions, visualization tools, and machine learning models like GLMs, random forests, and decision trees, are utilized to generate insights and predictive models.

➂ The system emphasizes data security and privacy, incorporating encryption and authentication mechanisms. It also includes a user-friendly mobile app for real-time health monitoring and personalized feedback.

➀ This article introduces an AI-based predictive maintenance system designed to detect engine faults in vehicles through sound analysis. The system uses Google Teachable Machine to train an AI model capable of distinguishing between normal and abnormal engine sounds.

➁ The AI model runs in a browser, continuously monitoring the engine's sound. Upon detecting an anomaly, it triggers an immediate alert and sends an HTTP request to the IndusBoard Coin, which acts as a web server to process the alert.

➂ The IndusBoard Coin also hosts a webpage that visually indicates the system's status. If a fault is detected, the webpage turns red and displays an alert message while activating a blinking LED as a warning indicator.

➀ This article introduces a device that uses YOLOv8 for real-time face detection and landmarking, suitable for various applications such as doorbells and attendance systems for small businesses. The device features a 5-point facial recognition system and operates on a MaixCam development board.

➁ The hardware setup involves a Sipeed microcontroller with dual-core CPU, NPU, and 256MB DDR3 RAM, running a Linux-based OS. The system supports YOLOv8, enabling efficient real-time face recognition.

➂ The article provides detailed instructions on how to set up and use the MaixCam board, including connecting via Wi-Fi, uploading files using MaixVision IDE, and managing face databases.

➀ Retrieval Augmented Generation (RAG) is being developed by Fraunhofer IWU to streamline the process of finding crucial information in extensive technical and legal texts.

➁ The technology is designed to complement Large Language Models (LLMs) by providing precise and exhaustive information.

➂ The team at IWU is using the EU Machinery Regulation (2023/1230) as a demonstration of the technology's capabilities.

Retrieval Augmented Generation (RAG) is making it easier to find information in extensive documents by using large language models (LLMs) and a retrieval system. This method ensures precise and comprehensive information retrieval, which is particularly useful for legal texts and user manuals. The Fraunhofer IWU is developing this technology, which can be used on standard PCs and in the cloud, ensuring data security and privacy.

➀ Investment in GenAI is expected to increase by 76% this year, but expectations for its capabilities are declining, according to Gartner.

➁ This paradox is expected to continue throughout this year and next.

➂ Despite the high failure rates in initial proof-of-concept (POC) work and dissatisfaction with current GenAI results, foundational model providers are investing billions annually to enhance GenAI models.

➀ Oura's LLM-based AI Oura Advisor has exited beta and is now available to Oura subscribers, specifically owners of the Oura Ring Gen 3 and Oura Ring 4.

➁ The feature works like a LLM chatbot, providing actionable, personalized guidance based on aggregated data.

➂ Currently, the Advisor is limited to English and focuses on Activity, Readiness, Resilience, and Sleep.