Nokia to roll out its own AI assistant at MWC, but it’s not what you think

admin14 February 2024Last Update :
Nokia to roll out its own AI assistant at MWC, but it’s not what you think

Nokia to roll out its own AI assistant at MWC, but it’s not what you think،

Besides the anticipated morning of flagship unveiling (we can't wait for the Honor Magic 6 Pro and Xiaomi 14 Ultra, among others, to go global), this year's MWC (Mobile World Congress) has something else up its sleeve.

Something from Nokia – no, not another easily repairable handset like the Nokia G22 (which recently received a new So Peach color option). Instead, the legendary Finnish company will present its own AI assistant.

Before you say “ugh” and “meh”: it’s not what you think. This isn't Nokia's answer to Google's Gemini or the rest of the hot (“hot” as in “popular”, don't get any ideas now!) AI assistants.

Reuters reports that Nokia's AI-based tool generates messages for industrial workers, “including warnings about faulty machines, based on real-time data” and even recommends ways to increase factory production .

The tool is called “MX Workmate” and will extend Nokia's existing communications technology used by industrial customers by leveraging AI generative extended language models (LLMs) to write human-like text, the company said in a press release.

The first version of the tool will be presented at the next MWC at the end of February in Barcelona. Some aspects of the AI ​​assistant are still in the research stage, as developers want to be sure the bottom line doesn't succumb to AI hallucination – that's when the AI ​​provides a convincing but completely invented answer.

“The tool must be precise, clear and fair. And it must be traceable and moderated,” explained Stéphane Daeuble, head of enterprise solutions marketing at Nokia. He said there would be upfront safeguards, such as a person validating the AI ​​prompts.

Nokia already provides 4G and 5G technologies for internal communications that help industrial companies connect to machine sensor data. “Now the idea is we have an assistant who is there to help the worker make sense of all this data,” Daeuble said.

“Maybe a year, a year and a half before we see the first real implementation,” he concluded.