Alexander Fritsch, Senior Product Sales Manager at Advantech Europe featured on the latest episode of IoT Unplugged to talk through the evolution of Edge AI, architectural considerations, and the future of the market.
In covering the evolution of Edge AI, Fritsch broke it down into three phases: beginning with on-device AI, which was isolated; moving to a greater connection between Edgeand the Cloud, and culiminating in AI at the Edge.
The reasons behind data processing moving to Edge AI are multiple; Fritsch cited reduced latency, increased efficiency, cost savings, and enhanced data privacy as all drivers.
Fritsch explained that while early applications like number plate recognition relied on classic algorithms, the goal now is to enable devices at the Edge to perform local simulations, training, and self-learning, reducing reliance on Cloud processing.
Despite these advancements, full autonomy in AI – referring to systems operating without human oversight – is still a considerable amount of time away. Human control is still essential, particularly in areas such as cybersecurity and self-driving vehicles.
It’s important organisations pick the right processor architecture for Edge AI deployments. The choice depends on the specific use cases, and required performance and scalability needs.
Looking ahead, Fritsch predicted that the next three to five years will see more self-learning AI at the Edge, increased efficiency across industries, and a gradual reduction in public resistance as people experience the benefits of AI in everyday life.
To learn more about what Fritsch had to say about the Edge AI journey, listen on Spotify, Apple Podcasts, and at the link below.
There’s plenty of other editorial on our sister site, Electronic Specifier! Or you can always join in the conversation by visiting our LinkedIn page.
