AI Is a Game Changer, But It’s Not an Application, Part I

July 27, 2023

Sponsored Blog

If you took a general poll in the embedded space and asked, “What’s currently the hottest application,” a majority of the responses would be AI. However, that response would be inaccurate because AI (or its subset machine learning) is not an application. It may be the means to some other application, but on its own, it’s a technology.

In Part I of this two-part blog, I’m going to explain why AI is the enabler to an application, and not the end application. I’ll discuss some of the near-term uses of AI and how it fits into the consumer and industrial realms. In Part II (which is now posted), I’ll look a little further down the road, dive into the AI ecosystem, and provide some useful tools for engineers looking to implement this essential technology.

In terms of AI’s enablement, it should not be considered as being any different than low-power techniques, faster compute, or the latest security algorithms. In short, AI enables you to run your real end application better, faster, and more accurately. By that definition, which is the correct one, AI in some form will likely start appearing in many applications, just like low-power techniques, faster compute processing, or the latest security algorithms.

One reason AI is currently thought of as the application is because many new algorithms, and even processors themselves, are aimed specifically at AI. And frankly, implementing the technology is still not for the faint of heart. It’s getting easier, but the tools are still not that mature and many of the processes remain under improvement. Even when and where AI should be applied is an area of contention. Some applications can be maximized using simpler machine-learning techniques, which are easier and more cost effective to implement than a full-blown AI implementation.

AI in the Consumer Space

In the near future, say 24 months, expect to see AI continuing to play a prominent role in consumer applications, where it can enhance various aspects of our daily lives. For example, the algorithms can make personalized recommendations based on individual tastes and interests for things like e-commerce goods, music, movies, and content streaming services. It would do this by analyzing user behavior, preferences, and historical data.

Smart-home automation is another area where AI can have an impact, by automating and optimizing daily functions, including adjusting the lights and temperature, handling irrigation, and home security, by learning from the occupants’ behavior patterns and more accurately detecting intruders.

AI can transform healthcare by assisting doctors in diagnosis, treatment planning, and personalized medicine. AI algorithms will analyze medical data, patient records, and even the latest medical research papers to aid in accurate diagnoses and suggest appropriate treatments. Real-time data will come from wearable devices and health apps to monitor vital signs, detect anomalies, and provide personalized health recommendations.

AI in Industrial

On the industrial and consumer ends, AI will continue to be used for predictive maintenance. With more and better data thanks to upgrades in sensor technology, combined with newer/better AI algorithms, this application will be significantly improved. By detecting anomalies and predicting maintenance requirements, AI can optimize maintenance schedules, minimize downtime, and reduce costs.

In manufacturing applications, AI can automate and enhance quality control processes by analyzing images, videos, or sensor data to detect defects, anomalies, or inconsistencies in products or production lines. AI algorithms can also optimize supply-chain management by analyzing historical data, market trends, and other factors that influence demand and supply. This would result in more accurate demand forecasting, inventory management, logistics optimization, and efficient resource allocation, including energy optimization by running machinery at times when resources are more readily available and more cost effective.

Education is Needed

There’s an obvious need to get the design community thinking more about how AI can enhance their application, rather than thinking solely about AI as the application. A great start is to provide comprehensive education and training in what AI is and what it can do. It’s essential to empower engineers with knowledge about AI principles, algorithms, tools, and futuristic technologies from companies like Synaptics, to create AI enhanced products that ultimately offer a better outcome.

To really build AI-enabled solutions, there’s a need for efficient hardware coupled with software that runs on the hardware. Synaptics provides the multi-modal scalable Edge AI platform (see above image). Coupled with the hardware is the software that runs on the platform to enable engineers to build AI-enabled applications. Synaptics provides its open-sourced frameworks, tools, models, and techniques that are needed to create AI-enabled applications. In addition, the company is forging partnerships with providers of AI libraries and development tools to augment and complete the value proposition that is required to quickly implement a wide variety of AI applications.

The community should take the opportunity to learn from and collaborate with experts like those at Synaptics, taking advantage of R&D investment the company has made (and continues to make). By focusing on AI as the enabler, rather than the end application, industry leaders can cultivate the proper culture of AI awareness, competence, and enthusiasm amongst the engineering community and help drive engagement with and the adoption of AI.