I did say that chips are getting interesting. Apple announced the launch of the M5 chip, which promises to deliver a significant leap in artificial intelligence performance for its silicon products. The M5 chip features a next-generation ten-core graphics processing unit with a dedicated Neural Accelerator in each core, enabling more than four times the peak graphics compute performance compared to its predecessor, the M4. Additionally, the M5 chip is built using advanced three-nanometer technology and includes a powerful sixteen-core Neural Engine, providing up to fifteen percent faster multithreaded performance over the M4. Johny Srouji, Apple’s senior vice president of Hardware Technologies, emphasized that the M5 chip not only boosts AI workloads but also enhances graphics performance with a thirty percent increase in speed compared to the M4. The new chip will be available in Apple’s latest devices, including the fourteen-inch MacBook Pro and iPad Pro, both of which are available for pre-order starting today.
Intel has unveiled its latest chip, Panther Lake, which is poised to be a significant development in the company’s efforts to compete with rivals such as Apple, AMD, and Qualcomm. Scheduled for release in late 2025 and early 2026, Panther Lake will be Intel’s first chip built on its 18A process and is designed to enhance battery life, performance, and gaming graphics. The Panther Lake series will offer various models, including eight-core and sixteen-core processors, featuring new graphics technology with up to twelve integrated graphics cores. Intel claims these new chips will deliver 40 percent lower power consumption with comparable single-threaded performance and 50 percent more multi-threaded performance over its predecessor, Lunar Lake. The chips will also include advanced features such as an intelligent bias control system for gaming and a new neural processing unit for enhanced artificial intelligence tasks.
Why do we care?
I told you chips were getting interesting. Apple just dropped the M5, and it’s a beast — Neural Accelerators in every GPU core, a sixteen-core Neural Engine, and four times the graphics power of the M4. Translation? They’re not building faster laptops — they’re building edge AI machines.
Intel’s not sitting still either. The new Panther Lake chips, due next year, will use their 18A process — supposedly 40 percent more efficient and 50 percent faster in multi-threaded workloads. They’re even tossing in an NPU and some clever gaming optimizations. Bold claims — we’ll see if they deliver.
Here’s the real story: AI is moving off the cloud and into the device. Apple, Intel, Qualcomm — they’re all baking AI right into the silicon. That’s going to matter for MSPs because endpoint management just got more complicated. You’ll need to know which machines can run local AI, what models they can handle, and how to secure those NPUs.
This is the next refresh cycle. Clients won’t just buy new devices for speed — they’ll buy for AI capability. The smart move for providers? Build AI-readiness assessments, refresh plans, and governance services around these new chips. The AI race isn’t just in the data center anymore — it’s in your client’s laptop bag.