This blog post was originally published at Qualcomm’s website. It is reprinted here with the permission of Qualcomm.
Explore topics about on-device generative AI in 2023 with our subject matter experts
Welcome back to AI on the Edge. In our last roundup of posts, we introduced our new series with topics relating to the ever-growing impact of on-device artificial intelligence (AI). Since then, AI continues to buzz throughout all industries and we’ve been hard at work gathering data and interviewing our subject matter experts to help the world better understand the AI landscape unfolding before our eyes: from the importance of offloading the cloud with AI-enabled edge devices, to predicting how AI will evolve in 2024 and beyond.
Through all of these insights, it’s clear that on-device AI is transforming how businesses and consumers operate, providing opportunities for better experiences across the board. We hope this will help you better understand and decide how on-device AI can benefit you, your work and society.
Preserving privacy and security in a generative AI world
As generative AI’s footprint grows, concerns about privacy and security can’t be overlooked. Generative AI models hosted on the cloud, for example, require user data to leave the device. However, on-device generative AI can mitigate these challenges by keeping user data directly on the device and powering decisions based on local information. On-device security features, including biometric passwords and data encryption, offer additional protection for AI models.
Learn more about how on-device AI enhances privacy and security
Getting personal with on-device AI
AI at the edge promises greater personalization while maintaining privacy and security. How? On-device AI leverages local data about the individual users and their current environment. Smartphones, for example, can take data from built-in sensors like cameras or microphones for on-device inference to give more relevant and personalized responses, which would otherwise be costly, inefficient and impractical on the cloud. The result? An enhanced, more enjoyable and customized user experience.
Learn more about the benefits of personalized AI at the edge
The evolution of generative AI: From the generic cloud to the personalized intelligent device
In this interview, Qualcomm Technologies’ Senior Vice President of Product Management, Snapdragon Technologies and Roadmaps, Ziad Asghar, discusses the value that on-device generative AI brings to our lives now and in the future, moving toward models that are contextually aware of their environment and who’s using them. This offloads the cloud while still providing a personalized yet private experience. Asghar also shares his thoughts on how on-device AI will change how we interact with our devices.
https://www.qualcomm.com/news/onq/2023/10/the-evolution-of-generative-ai-from-cloud-to-device
Everyone becomes more efficient with generative AI
We have been a leader in technology since the wireless and mobile revolution and now play a significant role in the AI revolution. Qualcomm Chief Marketing Officer, Don McGuire, shares his thoughts on the future of AI and what it means for us, the industry and society. This includes how the AI hype cycle may be sustainable, the plethora of exciting AI applications available today and how AI will scale to video, gaming, the metaverse and more in the future.
Learn about the efficiency AI unlocks now and in the future
The positive social impact of AI
There’s been a lot of talk about AI’s opportunity to increase efficiency and productivity. But what about the opportunity to create positive change in society? Learn about how the power of AI can be used to advance scientific innovation, supercharge creativity and accessibility and provide valuable insights more quickly than before.
In short, AI’s ability to collect and process vast amounts of data, make predictions and automate tasks makes it a valuable tool for tackling the world’s greatest issues.
Read about the opportunity to leverage AI for the greater good
Advancing perception across modalities with state-of-the-art AI
Our Qualcomm AI Research team has been making waves with state-of-the-art (SOTA) results, poised to transform some of the buzziest tech innovations, including video streaming, autonomous driving and augmented reality. Read about several specific highlights from our recent research, including creating and understanding 3D, computer vision, wireless sensing, efficient speech recognition and more.
Hear about our leading innovations and state-of-the-art AI
The state of AI and where it’s heading in 2024
The adoption of on-device generative AI is driving the democratization of AI on consumer devices to further enhance user experiences. Qualcomm Technologies’ Senior Vice President and General Manager, Technology Planning and Edge Solutions, Durga Malladi, shares his first-hand insights on how the technology is evolving as we head into 2024, its growing impact on language and speech processing, what’s driving large-scale deployment, the diverse workloads AI can support and more.
Explore more about the current state of AI and where it’s heading in 2024
The generative AI economy: Worth up to $7.9T
As tech companies double down on generative AI, analysts are working to determine the economic impact of technology. This is challenging given generative AI’s broad value proposition; the opportunities to impact how we learn, work, play and live are endless. The only consensus is that the value of generative AI cannot be measured in dollars alone but that its economic potential is huge.
Learn more about the generative AI economy
Optimizing generative AI for edge devices
Generative AI applications are quickly gaining popularity, necessitating the move for some or all of AI workloads to edge devices like smartphones, PCs, XR headsets and more. Quantization, pruning and knowledge distillation are effective techniques to help run generative AI models more efficiently on device while maintaining accuracy, resulting in reduced latency, enhanced privacy and other advantages for users.
Learn more about optimizing AI models for on-device processing
Pat Lawlor
Director, Technical Marketing, Qualcomm Technologies, Inc.
Jerry Chang
Senior Manager, Marketing, Qualcomm Technologies