Introduction
The Internet of Things (IoT) is changing how we interact with technology. With the rise of generative artificial intelligence (AI), we see new possibilities for smart devices. This blog explores three generative AIoT applications that can potentially transform how we use technology.
Application #1: Code Generation for IoT Generative AI can be used to create, complete, or combine software code, aiding developers in constructing innovative applications. GitHub Copilot is an example of this application, which suggests software code and entire functions in real time. While generative AI is not likely to replace developers anytime soon, it can be another tool in the toolbox for code generation.
Application #2: Robot Control Generative AI can generate control logic and commands for robots based on motion data from humans or animals. Instead of deterministically programming movements for each leg of a robot, generative AI models can generate the movements of individual parts and make the robot walk complicated steps. Additionally, generative AI models can help robots make sense of their surroundings and connect horizon goals with more intermediate steps to reach the goals.
Application #3: Social IoT Devices Generative AI can make device communication “more social” by allowing devices to answer complex questions, allowing the user to talk to the device to change settings, and allowing devices themselves to use generative AI to generate answers. Amazon has developed the DialFRED framework, which will enable robots to ask questions if they are unsure. Through reinforcement learning, the questioner model is fine-tuned to ask the right types of questions at the right time to benefit task completion.
Challenges: While generative AI is a hot topic, it is not yet widely adopted in IoT applications. The existing generative AI landscape is not focused on IoT data except image data. Moreover, the accuracy of generative AI solutions in IoT settings must usually be very high. A “most likely” answer is often not good enough, and the result must be available quickly.
Conclusion
Generative AI is changing how we interact with technology, and we are seeing new possibilities for smart devices. While there are challenges to implementing generative AI in IoT applications, there is an opportunity for the first IoT-focused vendors to utilize generative AI at scale and generate a first-mover advantage. As technology evolves, we expect to see more innovative uses of generative AI in IoT applications.
