[MUSIC PLAYING]

Welcome to Electronica 2024, where I'm here with Amichai, who is the Senior Vice President of Embedded Processing at TI. Amichai, we want to talk about Edge AI today. What exactly is TI doing in this space?

Yeah, it's an excellent question-- the first mark. I'm very excited to be here today. It's nice to see that the AI booth, everything that is happening here in Electronica, but specifically Edge AI. It's something that I'm personally very excited from the amount of opportunities that it creates for our customers and for TI. And what it is is basically customers have a lot of data. So customers that are doing the solar system, they have information on how the whole solar system operates. A customer building a car, they will have a lot of sensor data on how the car behave.

We allow the customer to take this data and use Edge AI to be able to detect or to create a certain function that makes the system better. Doing it on the Edge, what it means is that you have limited resources, like any other embedded system. But still, TI offer user tools, makes it super easy to make sure that you can take all this data, create the function that allows you to detect a fault, to do a preemptive maintenance. Anything that you're trying to do in your system using your data with the TI devices, that's what we call basically Edge AI.

And we are showing several demos here on the show today on how to implement it. And I think every time that I meet a customer, I learn about a new application and how they are going to leverage the TI devices to build a new Edge AI system that will make the world more sustainable, easier to use, and also safer for everyone.

Great. So I'd love to get into the demos, but from a product perspective, what do we offer from TI that has this type of Edge AI capability?

Yeah, another excellent question. So we're trying to offer the full spectrum of devices, starting from our higher-end processor, usually used for vision system. So it can be detecting what's in front of a robot. It can be detecting what's in front of the car using a vision system and using the Edge AI capabilities to do a profiling in front of the camera. Then you go to a little bit lower end, where it's more of a time domain type of analysis. And then customers have the ability-- for example, we are showing here a demo on the solar system using our new P55 device that we just announced.

I think it's one of the best devices in the industry. But again, I'm a little bit biased. But it's used in, both, the TI real-time capabilities plus the TI Edge AI capabilities to create a great real-time system but also to be able to detect a fault in a solar system before it happens.

So if you look in your house, if you have a solar system, it's great when it operates. But if two cables are touching each other and you want to make sure that you shut down the system before any damage is created to the house. What we are showing here today is a device that can do it with over 99% accuracy and also very, very fast. So it shut down the system before any damage is created to your house or to wherever the solar system is installed.

Very good. So just picking your brain on where do you think the industry is headed with Edge AI? Where do you think TI will be in this space in 10 years?

Yeah, so it's very interesting. And I'm not very good at predicting the future, but what I can tell you is that we are doing-- we are trying to enable our customers. And every time I meet a customer-- it can be a customer building an appliance; it can be a customer building, again, a solar system; it can be a customer building an HVAC system-- all of these customers are thinking how to use the TI devices with the Edge AI capabilities to improve the system.

At the end, what it means is you build a safer system, you build a system that consumes less energy, you build a system that is easier for the consumer to use it. So it makes our life better, safer, and a lot more sustainable. And I think that's basically we're going to see it step after step, Edge AI capabilities creating a better world around us. And that's exactly what TI is trying to do, make the world better through semiconductors, make everything more affordable, more easier to use for customers. And I think Edge AI is going to be a big part of it as well.

Well, thank you, Amichai. I think, as you mentioned, we have vision, we have real-time processing. I'm actually very excited to see what happens in the future. If you'd like to learn more about Edge AI, please go to ti.com/edgeai.

[MUSIC PLAYING]