Want smarter insights in your inbox? Sign up for our weekly newsletters to get only what matters to enterprise AI, data, and security leaders. Subscribe Now
At VentureBeat’s Transform 2024 conference, Nitin Kumar, Qualcomm‘s VP of Product Management, outlined the company’s vision for AI-powered PCs and Qualcomm’s role in reshaping computing.
Kumar didn’t sugarcoat the state of PC innovation. “PCs really haven’t evolved, especially when you compare to the smartphone space,” he stated. He positioned AI PCs as a turning point that will “drive growth and new experiences across all variety of users of PCs and improve their lives and productivity.”
Highlighting Snapdragon’s strengths, Kumar emphasized robust performance without draining battery life and an integrated high-performing NPU (neural processing unit). He noted tight integration with Windows and Microsoft’s ecosystem as another key advantage. “That’s a fundamental value proposition of Snapdragon PCs,” Kumar explained. “Delivering all performance at a much lower power consumption.”
Local LLMs and 45 TOPS: Inside Qualcomm’s AI capabilities
On-device AI took center stage in Kumar’s remarks. He outlined benefits like faster processing, cost savings, enhanced privacy and improved personalization. Kumar noted that Qualcomm’s latest Snapdragon X series processors feature a 45 TOPS NPU, enabling local execution of complex AI models. “You can run [Meta’s large language model] Llama, a 7 billion parameter model locally on the device, without hitting the cloud,” he said.
Kumar discussed Qualcomm’s AI Hub, which gives developers tools to optimize and deploy AI models on Snapdragon devices. “We have an initiative. We call it AI hub and we have launched that publicly,” he said. “It’s essentially a portal that has more than 80 models already optimized.”
AI Scaling Hits Its Limits
Power caps, rising token costs, and inference delays are reshaping enterprise AI. Join our exclusive salon to discover how top teams are:
- Turning energy into a strategic advantage
- Architecting efficient inference for real throughput gains
- Unlocking competitive ROI with sustainable AI systems
Secure your spot to stay ahead: https://bit.ly/4mwGngO
Kumar emphasized the privacy advantages of on-device AI compared to cloud-based solutions. Keeping data and AI processing on the device itself means sensitive information never leaves the user’s control. This approach addresses growing concerns about data privacy and security. By running AI models locally, users can benefit from personalized experiences without exposing their data to potential breaches or misuse in the cloud. This is especially crucial for enterprise applications dealing with confidential information.
Additionally, on-device AI allows for personalization using data that users may not want to upload, further enhancing the user experience while maintaining privacy. As Kumar put it, “What happens on the edge, stays on the edge.”
Re-imagining PCs for the AI age
The Qualcomm executive spotlighted the company’s partnership with Microsoft, particularly regarding Windows and Copilot+. “Microsoft announced a whole set of new set of Copilot+ PCs, and they exclusively run on [Snapdragon] processors,” Kumar noted. “With Copilot+ PCs, there are more marquee AI models that are running on the NPU… running pervasively all the time.”
Qualcomm isn’t just dipping its toes into AI PCs – it’s diving in headfirst. Drawing on its mobile tech prowess, the company is making bold moves to reshape personal computing. With major PC makers already on board, Qualcomm seems poised to take advantage of the deployment of generative tech in everyday devices.
“This is truly what we believe— [it’s a] a pivotal moment in the PC industry,” Kumar concluded.
