AI was everywhere at CES 2025 in Las Vegas from industry giants to small appliance makers like smart vacuums that pick up dirty socks or AI-powered pellet barbeques . But the more serious battleground was between mega chip makers like Intel (124,800 employees – $87 billion), Nvidia (22,500 employees – $3.387 billion), AMD (13,000 employees – $340 billion) and Qualcomm (41,000 employees – $170 billion). It was an eye opener focused on human and financial resources, not for new chip technology but more AI capabilities. Each company is pushing the AI envelope beyond better gaming performance, general computing or battery efficiency to powerful dynamic object and environment awareness and more.
Intel Eyes Chips And Cars
Intel’s AI was bristling in next-gen desktop and notebook processors for professionals and casual consumers. Fresh off of its worst year since going public in 1971 with overheating and underperformance issues in last years chips (almost resolved as you read this, as I wait for Intel’s final BIOS 200S series update any day now for my test rig). Intel has a lot to prove with this fresh crop of new chips at CES 2025, hoping to turn it’s fortunes around after dismissing CEO Pat Gelsinger with new Interim Co-CEO’s David Zinsner and Michelle Johnston Holthaus.
Overall, Intel’s CES 2025 offerings show a clear progression in terms of AI integration and computing performance levels compared to their 2024 lineup. Let’s take a look.
The new Core Ultra 200V desktop series processors were unveiled at the show along with Core Ultra 200H and 200HXseries, targeting mainstream and high-end notebooks, respectively. These processors are built to handle on-device AI tasks, better battery life and performance compared to previous models for both business and consumer users.
“Intel Core Ultra processors are setting new benchmarks for mobile AI and graphics, once again demonstrating the superior performance and efficiency of the x86 architecture as we shape the future of personal computing,” said Michelle Johnston Holthaus, interim co-CEO of Intel and CEO of Intel Products. “The strength of our AI PC product innovation, combined with the breadth and scale of our hardware and software ecosystem across all segments of the market, is empowering users with a better experience in the traditional ways we use PCs for productivity, creation and communication, while opening up completely new capabilities with over 400 AI features. And Intel is only going to continue bolstering its AI PC product portfolio in 2025 and beyond as we sample our lead Intel 18A product to customers now ahead of volume production in the second half of 2025.” Processor details:
- New IntelCore Ultra 200V series mobile processors with Intel vPro are empowering businesses with AI-driven productivity and enhanced IT management. The combination of performance, efficiency and industry-leading business computing with advanced security and manageability – all while enabling a seamless Microsoft Copilot+ experience – helps to deliver a robust platform for modern workplaces.
- New Intel Core Ultra 200HX and H series mobile processors feature improved Performance-cores and Efficient-cores, an integrated neural processing unit (NPU) for AI acceleration and available Intel Arc graphics.
- New Intel Core Ultra 200U series mobile processors blend performance and power efficiency for mainstream mobile users.
- The newer Intel Core Ultra 200S series desktop processor (launched last year at 125 watts) portfolio expands with 12 new 65-watt and 35-watt offerings for mainstream desktop users.
- New Intel Core Ultra processors, Intel Core™ processors, and Intel Processors for Edge computing prioritize scalability and performance across edge-relevant applications. These processors, particularly the IntelCore Ultra 9, excel in AI workloads, showcasing significant gen-over-gen performance advancements in key areas like media processing and AI analytics.
- Some come with extras like Intel’s newest Wi-Fi technology, Intel Wi-Fi 7, and memory on the package.
- New updates on Intel’s 18A manufacturing and next-generation processor family.
As with other recent generations of Intel chips, the new lineups have different types of cores that make up the chip package. Some are “P-cores,” which are focused on performance, while others are “E-cores,” architected for efficiency. A third category of cores, “low-power E-cores,” are even more power-efficient than the standard E-cores.
Intel claims that its newest generation of P-cores have been re-engineered with AI-based power management and other optimizations, while its new E-cores are its “most efficient” yet. “Next-gen low-power E-cores optimized for tasks requiring less power, along with new power, thermal, and acoustic features, help create powerhouse thin and light systems,” wrote the company in a release.
The Core Ultra 200H chips get Intel’s flagship built-in graphics in certain configurations: Intel Arc with XMX. The Core Ultra 200S, Core Ultra 200U, Core 200S, and Core 100U have the company’s less powerful Intel Graphics brand of GPUs. But the Core Ultra 200S and Core Ultra 200U, along with the Core Ultra 200H, ship with Intel’s AI accelerator tech, AI Boost, under-hood.
Certain processors, like those in the Core Ultra 200HX, Core Ultra 200H, Core Ultra 200U, and Core Ultra 200V series, sport Intel’s neural processing unit (NPU), a dedicated chip for running AI applications and features in Windows 11. Intel claims that the new NPU packed into the Core Ultra 200V has 2x the bandwidth compared to the previous version.
Intel is pitching its Core Ultra 200V, HX, H, U, and S chips as best for “commercial” applications. By comparison, the Core Ultra 200S series should deliver “desktop performance.” The Ultra 200HX is aimed at “enthusiast notebooks,” according to Intel, while the Ultra 200H is designed for devices with “thin and light” form factors.
An important differentiator for the Core Ultra 200V – which also has Intel Arc graphics – is the inclusion of Intel’s vPro enterprise-focused ecosystem. That spans hardware-based threat detection tech, apps optimized for the chip, and stability and validation programs.
The Core Ultra 200V can also be configured by manufacturers to support Pluton, Microsoft’s hardware-based security co-processor, Intel said. Pluton acts as a hardware root-of-trust, which in simple terms protects a device’s hardware from tampering.
Intel is embracing the automotive industry introducing the Whole-Vehicle Platform, a comprehensive platform designed to accelerate the transition to electric and software-defined vehicles (SDV). This platform includes high-performance compute, discrete graphics, AI, power management, and zonal controller solutions. The new Adaptive Control Unit (ACU) U310 enhances energy efficiency and reduces costs in electric vehicles
Partnerships with Stellantis Motorsports and Karma Automotive were announced to integrate their technology for improved performance and efficiency and also announced Second-Generation Intel Arc B-Series Graphics for Automotive: These graphics are designed to enhance AI workloads and in-vehicle experiences.
A new collaboration with Intel and Amazon Web Services (AWS) co-developed the Intel Automotive Virtual Design Environment to revolutionize automotive software development, reducing R&D costs and speeding up time-to-market. The company cites these innovations highlight its commitment to advancing automotive technology and supporting the shift towards more efficient, software-driven vehicles.
NVIDIA
NVIDIA impressed with a rich portfolio showcasing its latest advancements in AI and gaming and was recognized with the Best of CES award not just for AI but the entire show. Powered by its new Blackwell architecture with high-performance computing and AI workloads for professionals it showed significant advancements in enhanced AI performance for complex workloads, suitable for data centers, scientific computing, and AI research.
Say hello to Nvidia’s Cosmos, a new AI platform unveiled at CES 2025. It’s designed to accelerate the development of physical AI systems, such as robots and autonomous vehicles in how they react in the real physical world. Here are some key points about Cosmos:
- Generative World Foundation Models: Cosmos uses state-of-the-art generative models trained on over 20 million hours of video data to create photo-realistic simulations. These models help robots and autonomous vehicles understand and interact with the physical world more effectively.
- Integration with Omniverse: Cosmos works in conjunction with Nvidia’s Omniverse platform, which generates simulations that Cosmos then turns into realistic video imagery for training purposes.
- Open-Source Availability: The Cosmos code is available under an open-source license on GitHub, making it accessible to developers worldwide.
- Applications: Cosmos can be used for various applications, such as training robots for warehouse environments or improving the performance of autonomous vehicles.
- Partnerships: Nvidia has partnered with major companies like Toyota to use Cosmos for developing next-generation autonomous vehicles.
Nvidia’s CEO, Jensen Huang, believes that Cosmos could be as transformative for robotics and AI as large language models have been for enterprise AI. Not solely robots, either. Check out smart glasses to process speech and images with Nvidia technology. Nvidia and Toyota have already inked a deal to use the Cosmos AI model to train cars.
As ZDNET Editor-in-Chief Jason Hiner put it “Nvidia Cosmos demonstrates the biggest and boldest ambition we’ve seen at CES 2025 for how technology could help people and communities in the years ahead.” Nvidia’s Huang calls it “the ChatGPT moment for robotics.”
Nvidia also unveiled its new GeForce RTX 50 Series graphics cards. These GPUs are powered by Nvidia’s new Blackwell architecture and are designed to deliver game-changing performance for both gamers and creators. The flagship model, features 92 billion transistors, 32GB of GDDR7 VRAM, and 3,352 trillion AI operations per second (TOPS). It’s designed to handle native 4K gaming and even 8K gaming with ease. It’s also worth noting how Nvidia’s tech has leaped ahead other players in the GPU space. Nvidia is so heavily dominating the GPU market that AMD and Intel and have been relegated to competing in the midrange category.
Huang also announced a new chip named Thor for cars and trucks that uses AI to process visual information coming in from cameras and lidar sensors to lead the way in Level 4 autonomous driving. And he revealed the 50-series lineup of gaming and laptop GPUs that promises to deliver massive leaps in performance and “breakthroughs in AI-driven rendering” at a lower cost than the 40 series (in most cases).
Qualcomm
Qualcomm entered the fray with its Snapdragon X Elite line of semiconductors, designed for Windows-based computers. These chips are also focused on AI capabilities and improved battery life as well as collaborations across various industries. Qualcomm unveiled the Snapdragon X platform, the fourth in its high-performance PC portfolio, bringing industry-leading performance, multi-day battery life, and AI to more of the Windows ecosystem.
Qualcomm expanded its automotive collaborations with companies like Alpine, Amazon, Leapmotor, Mobis, Royal Enfield, and Sony Honda Mobility. They showcased the Snapdragon Digital Chassis solutions for AI-powered in-cabin experiences and advanced driver assistance systems. Smart Home 2.0: Qualcomm introduced new AI chatbots integrated into appliances, advanced smart TVs, and humanoid robots, envisioning 2025 as the start of “Smart Home 2.0” with significant advancements from generative AI integration.
Qualcomm also announced new services for its Qualcomm Aware platform to drive connected intelligence across industries and introduced an on-premises AI inference solution for enterprises. Qualcomm’s CEO, Cristiano Amon, emphasized that AI is creating a generational shift in technology, with on-device AI becoming the next user interface, transforming experiences in PCs, automotive, smart homes, and more.
AMD
AMD countered with its Ryzen 9950X3D CPU and Ryzen 9 9900X3 high-performance desktop users and gamers with Zen 5 architecture. These CPUs feature AMD’s 2nd Gen 3D V-Cache technology, offering impressive performance for both gaming and content creation. With significant improvements in single-thread and multi-thread performance over previous generations boasting advanced cooling solutions for sustained high performance and optimized for gaming with enhanced graphics and processing power with specs like 16 cores, 32 threads, and a max boost of 5.7 GHz.
Also new, the Ryzen AI Max Series processors, combine leading CPU core performance, fast built-in graphics, and a powerful Neural Processing Unit (NPU) for AI tasks. These processors are designed for next-gen AI PCs and offer workstation-level performance in ultra-portable form factors.
For handheld gaming, AMD announced the Ryzen Z2 series processors, which bring powerful graphics and processor performance to handheld gaming devices.
The Ryzen AI 300 Series include models in the Ryzen 7 and Ryzen 5 segments, bringing AI performance and multitasking capabilities to a wider range of users.
AMD also revealed its new RDNA 4 architecture, with the Radeon RX 9070 XT and RX 9070 GPUs set to launch in Q1 2025. These GPUs are designed to improve performance-per-watt and offer better upscaling with AMD Fidelity FX Super Resolution 4
They also announced the Radeon RX 9000 series GPUs, skipping the 8000 series and aiming to compete directly with Intel and NVIDIA.