AI took control at CES this year. AI was mentioned everywhere, from huge language model-powered voice assistants in cars to the Rabbit R1. It was a touch excessive.
It may be the year of artificial intelligence at CES, but many of these “AI” capabilities have been there for a while – it’s just that businesses are just now adopting the branding of artificial intelligence. AI has reached the public consciousness: putting it front and center in a product is chic and hip, indicating that organizations are ambitious and forward-thinking. As a result, the word has been used whenever feasible, even when it does not refer to the AI that most people are familiar with.
But, as more corporations rebrand anything involving algorithms such as AI, how can we tell the wheat from the chaff? And, more significantly, wouldn’t this lead to overestimation of AI’s capabilities?
Whether or not new goods use generative AI, the name AI conveys the impression that a feature is novel and intriguing. Generative AI is likewise still in its throw-it-at-everything stage of development. People want to know how far they can push technology and whether it will be a major differentiator. This is why we’re seeing everything from Walmart restocking your cupboard with AI models to vehicle manufacturers packing ChatGPT into their dashboards to offer driver’s something to talk to.
According to Arun Chandrasekaran, an analyst at Gartner, this is normal for many organizations, but it does run the danger of overpromising to customers when they discover something labeled as AI isn’t like ChatGPT.
“There is a conflation now of generative AI and other AI that could muddle the field a little bit,” says Chandrasekaran. “Marketers might be shooting themselves in the foot when they advertise something that ends up not being what people expected.”
For better or worse, most people associate AI with generative AI — specifically, ChatGPT. This gives the perception that whenever a consumer uses an AI-branded product, they anticipate it to behave similarly to a chatbot that “thinks” like a human.
This is a disservice to products that use equally amazing forms of AI. Many of the robots on display at CES, such as Samsung’s Ballie and LG’s AI agent robot thing (it’s not precisely an AI agent; AI agents refer to AI software, usually a chatbot of sorts, that can do things such as book a trip or get a table at a restaurant), are adorable and engineering marvels. However, their emergence is due to developments in robotics and even computer vision rather than the rise of LLMs. (At least we don’t know if Samsung utilized LLMs to assist in training Ballie).
There’s also machine learning. Many AI researchers will argue that generative AI and the foundation models that underlie many implementations of it are simply the next stages of machine learning advancement. However, no one wants to talk about machine learning any longer. It’s deemed ancient and “traditional,” yet I’m sure it’s powering many of the pattern recognition features on display at CES.
“Technology has lifecycles, and yes, we may reach a point in AI where people are disillusioned with its promise after failing to solve many of the problems people expect it to solve.” But that’s when you start seeing a lot of fantastic solutions and better-fitting use cases,” Chandrasekaran explained.
In the coming years, we’ll see features and solutions that don’t require a chatbot or a huge language model. It’s just not at this year’s CES. No, not yet.
