Presented by Qualcomm Technologies, Inc.
The applications that on-device AI enable are capturing enormous consumer attention. Learn more about this innovative emerging technology that’s already at the fingertips of developers, OEMs, and ISVs in this VB Live event.
The movement of AI from the cloud to on device is accelerating, says Anshel Sag, analyst at Moor Insights & Strategy, and the catalysts keep stacking up, from the growing emphasis on privacy and improvement of on device performance, to the more sophisticated smart hardware coming on the market.
“More and more compute is moving closer to the edge, which means that more AI has to happen at the edge,” Sag says. “It helps achieve those latency goals that a lot of OEMs and ISVs are hoping to accomplish by deploying new technologies and leveraging 5G.”
Most of the benefits to consumers will actually be pretty invisible, and that’s on purpose. The obvious benefits will be that things are done more quickly and at lower power, which gives consumers more battery life on their devices. But ultimately, on device AI automates functions that would otherwise have to be done manually, or wouldn’t be done at all, and adds more features and more capabilities to devices without having to add more hardware or affect the size of the device.
One big area that on-device AI is impacting for consumers is photography, with AI filters offering scene detection, object detection, and smart camera settings in response to those observations. It’s also accelerating the efficiency and productivity of users by automating functions that they would otherwise have to do manually, as well as reduced latency for things like language processing in voice commands, personal assistants, and dictation.
Security cameras are also getting smarter with on-device AI. Doing more of the compute on device rather than sending the images to the cloud and having them processed and sent back means detection is near instantaneous and potential security risks are identified immediately.
For businesses, on-device AI will help them keep their sensitive data from being transmitted over the network as often as it is today. There’s a lot of potential vulnerabilities when you’re transmitting data over open WiFi networks, or even through a compromised cell tower. Transmitting less data generally results in less vulnerability. It will also theoretically allow for more offline AI functions to occur in areas where connectivity may not be as reliable, or there’s no opportunity for connectivity at all.
“Having the ability to still have these AI functionalities and accelerated compute capabilities will be extremely valuable, because then these applications can be built knowing that they’ll work everywhere, not just where you have cloud connectivity,” he explains.
Leveraging this technology requires the right hardware — a processor that can do AI inference at low power, and relatively quickly –and the right SDKs, supporting the right frameworks for AI.
“It’s important for OEMs and ISVs who want to leverage this technology to be patient and experiment and think about what benefits you want to accomplish using on-device AI,” Sag says. “Way too many people try to push AI to the cloud or tick a box to claim they have AI functionality. It’s much better to develop something that fits your needs than compromise with something that doesn’t. But ultimately, it’s a massive benefit for whomever does it right.”
To learn more about how on-device AI is unlocking a world of intelligent, connected devices that can perceive, reason, and take intuitive actions based on awareness of the situation with immediate response, enhanced reliability, increased privacy, and efficient use of network bandwidth, catch up on this VB Live event now.
Don’t miss out!
- How AI-powered smartphone apps are igniting the consumer imagination – and why
- How on-device AI is changing the future of connected devices
- How developers, OEMs, and ISVs can take advantage of on-device AI processing
- The technology you need to drive the best performance of on-device AI
- Gary Brotman, Sr. Director Product Management, Qualcomm Technologies
- Anshel Sag, Moor Insights & Strategy