Artificial Intelligence

Boyue Du created Symphonia Studio on an HP ZGX Nano (a GPU compute companion device) as part of NVIDIA GTC Hack for Impact. Boyue's project uses AI to let users input popular artists and music genres to recommend classical artists that most similarly pair with the current songs and genres users like. His project one HP's prize!Check out Boyue's open source code for Symphonia-Studio on git. 

Calling AI / ML practitioners: I’m looking for 15–25 minutes of your perspective (no scheduling). Interested?My team at HP is running a short, self‑guided AI‑moderated interview to better understand how teams approach AI development, experimentation, and production inference, and where today’s compute setups fall short.We aim to learn directly from the AI community to identify real needs and inform the future of our AI products.What to expect:✅ ~15–25 minutes if you’re a fit✅ No scheduling. Complete it anytime✅ Optional opt-in to be considered as a lighthouse account if your use cases alignTake the interview here:🚩 Start the 15-25 minute interview  

 NVIDIA shared an easy to deploy Video Search and Summarization (VSS) as Inspector demo that anyone can do in 30 minutes using the ZGX Nano and ZGX Toolkit VS Code extension.The demo uses Docker and Docker Compose for service packaging and orchestration, requiring CUDA and specific GPU drivers to work with NIM containers. It uses the cosmos-reason2 model on HuggingFace (be sure to request access and create a Read token).The VSS demo’s architecture is containerized, GPU-accelerated, and configured via scripts and Compose files for scalable video analysis. Give it a try! Git link:  https://gist.github.com/davrollins/b162595cc289a1b7a2cec8d14332a04d

Google researchers have unveiled GameNGen, a groundbreaking game engine entirely driven by AI that can simulate intricate video games in real-time. This innovative technology was showcased by recreating the iconic first-person shooter, Doom.GameNGen employs a neural network to generate game frames at more than 20 frames per second, creating visuals that closely mirror the original game. The AI model underwent a two-phase training process: initially, an AI agent was taught to play Doom, followed by training a separate model to predict the next frame based on previous actions and frames.“We introduce GameNGen, the first game engine entirely powered by a neural network, capable of real-time interaction within a complex environment over extended periods and at high quality,” the researchers detailed in their paper.This technology holds the potential to revolutionize game development by enabling the creation and modification of games through text descriptions or example images, moving away

 HP’s Andrew Hawthorn joins the Intelligent Machines team to talk private AI and Edge AI super computers. Check out the show. Let us know what you think…  Intelligent Machine is a part of TWiT.tv is a technology podcasting network located in the San Francisco Bay Area with the #1 ranked technology podcast This Week in Tech hosted by Leo Laporte. Every week we produce dozens of hours of content on a variety of programs including Tech News Weekly, MacBreak Weekly, Windows Weekly, Security Now, Intelligent Machines, and more.Follow Intelligent Machines: Website: https://twit.tv/ Bluesky: https://bsky.app/profile/twit.tv Mastodon: https://mastodon.social/@TWiT Facebook: / twitnetwork Instagram: / twit.tv X (Twitter): https://x.com/TWiT TikTok: / twittok LinkedIn: / twit-llc

HP is working to help users get a head start on customizing AI models with open-source AI blueprints available to anyone on Github. Search “Github AI blueprints” on Google or Chat-GPT to quickly find links to HP AI Blueprints for a variety of AI use cases on Github or visit Github directly: https://github.com/HPInc/AI-Blueprints.Happy coding! 

I tried a few free AI tools to prototype small local models. Some outputs were useful, others sounded like a motivational coach I didn’t ask for.Has anyone else tested these? Did it actually help, or just add extra comedy to your workflow?

 Image credit — GPT-5: What’s New in OpenAI’s Latest ChatGPT Model? | Built InThe buzzword in the AI for this week is GPT — 5, while there are mixed reviews about the model itself, the conversation is shifting from sheer performance benchmarks to critical questions of reliability and trust. For anyone who uses large language models (LLMs) regularly, their flaws — confident hallucinations, sycophantic agreement, and unhelpful refusals — are all too familiar.OpenAI’s flagship model isn’t just about getting bigger; it’s about getting smarter, safer, and more honest. Based on recent explainers, GPT-5 is being engineered with five targeted improvements designed to address these persistent issues. Here’s a practical, theory-grounded tour of what those changes are and why they represent a crucial step toward AI we can actually trust.Understanding the Core FlawsBefore diving into the solutions, it’s important to understand the problems. Here’s a quick breakdown of the key issues GPT-5 is desig

Dear Community Members,   We are delighted to invite you to an insightful talk presented by our talented student team.How will Neural Processing Units (NPUs) redefine the landscape of local AI? Our student team has delved into this question by conducting comprehensive performance comparisons of CPUs, GPUs, and NPUs on Z by HP mobile workstations. Their research focuses on a practical application: leveraging Large Language Models (LLMs) and Retrieval Augmented Generation (RAG) for efficient document interaction.During this session, you will gain valuable insights into:- Performance Comparisons: Detailed findings on processing speed across CPUs, GPUs, and NPUs.- Energy Efficiency: Real-world data on the energy consumption of these processing units.- Local AI Capabilities: A practical demonstration of LLMs and RAG for document interaction on mobile workstations.- Future of Computing: An engaging discussion on the potential of NPUs and the future of local AI. Event Details:Date: April 4th,

 A recent video post by Akash James of SparkCognition shows how he was able to get the GPU he needed to work on his multi-modal LLM called VERsatile ONline Intelligent Cognitive Agent (VERONICA) by accessing compute he owned in India from the Himalayas. Join the Z by HP Boost webinar Oct 4th @ 9 am PST to hear how Akash is accessing his GPU remotely to create AI.  

I'm a Researcher, We want to work on Artificial intelligence, Deep learning and Machine Learning algorithms on healthcare databases. We have a budget of about 30,00,000/- Lakhs Indian rupees to spend. Please suggest a Workstation and an NVIDIA GPU under this budget. 

Imagine teaching a robot to boil water. It might first search for a pot, fill it with water, then place it on the stove. But what if it mistakenly turns on the microwave instead of the stove? Traditional AI agents often struggle to recover from such errors, leading to cascading failures. This is the problem Agent-R—a novel framework for training self-reflective language model agents—aims to solve.In this article, we’ll unpack the theoretical backbone of Agent-R by focusing on how this framework redefines error recovery in AI.Why Error Recovery Matters in Interactive EnvironmentsMost AI agents learn by mimicking expert trajectories (e.g., perfect step-by-step guides). But real-world tasks are messy. Errors are inevitable, and waiting until the end of a task to correct them is like letting a typo in a sentence propagate into a garbled paragraph.The Core Challenge: Timely InterventionProblem: In multi-step tasks (e.g., crafting items in Minecraft or navigating a virtual lab), errors early

Multi-agent systems have become increasingly important in solving complex problems through distributed intelligence and collaboration. However, coordinating multiple agents effectively remains a significant challenge, particularly in the field of reinforcement learning. In this article, we'll explore a novel approach called Shared Recurrent Memory Transformer (SRMT) that aims to enhance coordination in decentralized multi-agent systems.The Challenge of Multi-Agent CoordinationImagine a group of robots trying to navigate through a crowded warehouse. Each robot has its own goal and can only see what's immediately around it. How can they work together efficiently without bumping into each other or getting stuck? This is the essence of the multi-agent pathfinding problem.Traditional approaches often struggle with this task, especially when the environment is complex or the number of agents is large. They may rely on centralized control, which doesn't scale well, or on explicit communicatio