On-device AI is transforming computing for hybrid workforces

Presented by Qualcomm Technologies Inc.

Traditionally, laptop performance has been measured by CPU and GPU, but on-device AI processing is now a critical third measure. And that’s especially crucial for enterprise PCs as hybrid work challenges crystalize, and the way we use our laptops has fundamentally changed over the last several years. More than ever before, cutting-edge PC technology is our most important and effective business communication and productivity tool.

Without the right tools to be effective at work, Gallup found that employees often feel less connected to an organization’s culture, their experience impaired collaboration and relationships, and work processes were often disrupted. That’s particularly true for PCs without AI capabilities, which suffer from lagging, blurring noise and video quality, multiple distractions, unsecured privacy, unstable connectivity and short battery life.

“Having high-performing on-device AI processing alongside the CPU and GPU is crucial for enterprise users,” says Kedar Kondap, SVP and GM, Compute and Gaming at Qualcomm Technologies, Inc. “Productivity today requires more: faster, smarter, more reliable connectivity, higher-quality cameras and clear and crisp audio, to fuel vital human-to-human interaction. It requires advanced security tools, and it flourishes with dedicated AI processing to power natural interactions with incredible power efficiency.”

If the consumer is mobile, the device needs to be intelligent enough to manage power for every individual user; on every platform it needs to understand user preference with on-device learning.

“Our vision is to drive the convergence of mobile and PC, bringing the best of the smartphone to your laptop,” Kondap says. “That means enhanced software, custom hardware, unprecedented connectivity and broad ecosystem support.”

Powering hybrid work with AI

AI-enabled PCs are stepping up to the challenge that enterprises face. They offer premium performance, multi-day battery life and fast 5G connectivity to help increase productivity, collaboration and security. Qualcomm Technologies itself has partnered with leading PC OEMs to launch several generations of Snapdragon compute powered laptops in thin and light designs.

Here’s a look at how AI features are improving work for hybrid employees:

Enhanced and more natural video conferencing. Video quality in video conferencing has been a long-time industry challenge, but now it’s become a crisis. Millions of people turn on their cameras and microphones every day across multiple applications, from Discord and Google Meet to Microsoft Teams, Slack, Zoom and many more. A recent study showed that 75% of people judge colleagues based on their audio quality in meetings, while 73% judge based on video quality.

Noise suppression, background blur, auto-framing and eye gaze correction change the game for video conferencing. Qualcomm and Microsoft collaborated on Snapdragon to enable these new AI-accelerated experiences for Windows 11 — without impacting performance and power efficiency. Snapdragon intelligently offloads these computationally intensive tasks to a dedicated AI engine, which frees up the CPU and GPU resources.

Lightning-fast connectivity. A PC that uses multiple radio bands and AI can take full advantage of that for video conferencing, to aggregate wireless connections and get a near optical cable-level of performance. Redundancy means if most connections fail, the video will continue without dropping the call. Not only will future PCs be able to connect to multiple Wi-Fi 6 connections, but also 5G simultaneously.

Advanced security and privacy. AI is powering new remote device management protocols, zero touch deployments and advance endpoint security, because a large portion of the workforce is no longer on-site. AI in PCs can also use Wi-Fi as a proximity detector to better secure the privacy of PCs. The PC can wake when it detects a user sitting down, and suspend and secure the PC when the user leaves their desk.

More power for intensive computing tasks. Adding on-device AI processing to powerful CPU and GPU capabilities adds the ability to offload compute-intensive processes. Performance is dramatically increased, delivering a higher level of user experience. Snapdragon compute platforms offer best-in-class AI accelerated user experiences to reach a new level of mobile computing performance: The pre-released Procyon AI Inference Benchmark results show the Hexagon AI processor scores up to 5x faster than the competitor’s CPU and up to 2.5x faster than the competitor’s GPU.

Power saving and efficiency. Traditional x86-based systems primarily rely on CPU and GPU for compute intensive modeling. By leveraging the dedicated AI engine, Snapdragon delivers all these experiences on the AI engine while freeing up CPU and GPU resources and optimizing for long battery life.

Tools to improve productivity and power innovation. The application ecosystem is the lifeblood of the end-user experience. As applications are optimized for Snapdragon, developers can start to take advantage of the on-device AI, and have access to new ways to personalize the user experience. For instance, when a user is editing photographs, the creative application will be able to automatically make adjustments that improve the image, or when they’re writing a document, it can auto-fill the end of sentences.

To ensure apps are able to take advantage of the computing that Qualcomm has to offer, including new intelligent capabilities, the company provides models to third-party app developers, and works with them to ensure that they’re optimized for performance and able to take advantage of the AI on the device to improve the end-user experience. Qualcomm and Microsoft are offering tools like the Windows Dev Kit 2023 to third-party app developers, for optimizing applications for Windows on Snapdragon.

For instance, Adobe used the Snapdragon 8cx Gen 3-powered developer kit to ensure their Creative Suite utilized dedicated AI processing capabilities for more personalized and intuitive experiences that pair with Adobe Sensei. And Qualcomm partners at Adobe are at the forefront of utilizing AI to enable creators. Later in 2023, key Adobe Creative Cloud applications will become native for Windows on Snapdragon.

Next-generation AI capabilities for enterprise PCs

This is only the beginning. Chip developers are integrating AI capabilities that will become more pervasive in PCs, enabling users to conduct the kind of generative AI that has emerged, performing tasks such as automatically drafting essays from outlines, or creating unique images through a text prompt.  

“Enterprises need the right compute to run the experiences that power their workforces, which is what we’re highlighting when we say on-device AI is the third crucial metric,” Kondap says. “We’re putting the software applications that can take advantage of AI in place and working with strategic partners that can take advantage of the on-device silicon we have. Our vision is optimizing the experience for consumers, enhancing software, hardware, connectivity and more, so that there are no barriers between users and their aspirations.”

Learn more here about how on-device AI is transforming what the enterprise workforce can achieve.

Sponsored articles are content produced by a company that is either paying for the post or has a business relationship with VentureBeat, and they’re always clearly marked. For more information, contact [email protected].

Source : https://venturebeat.com/ai/on-device-ai-is-transforming-computing-for-hybrid-workforces/

Leave a Comment

SMM Panel PDF Kitap indir