Argos Multilingual Names New Exec to AI Data Unit

US-headquartered language solutions integrator (LSI) Argos Multilingual has named Mitchell Ison as Vice President and General Manager, Frontier AI Data Services for its Argos Data division. 

Ison will report to Chris Phillips, Argos Multilingual’s Chief Innovations Officer and Head of Argos Data. Ison previously held go-to-market roles at AI data companies Labelbox and Sama. 

At Argos, Ison’s “sole focus is understanding how the world’s leading AI developers buy, evaluate, and consume human data,” Phillips told Slator in response to questions about Ison’s appointment. Ison will lead expansion in model evaluation, post-training data, red-teaming support, and annotation for companies building multilingual AI systems.

Argos is one of an increasing number of LSIs that have moved into the data-for-AI market, leveraging their multilingual capabilities and access to talent, operations, and existing workflows to carve out a niche in the AI ecosystem. 

The company “has been producing speech, language, and evaluation data for large technology companies since the neural machine translation era, well before generative AI made “AI data” a category of its own,” said Phillips, including with some of the world’s largest AI developers. This year, Argos brought those capabilities together under Argos Data, Phillips said.

“We have spent 30 years building the linguistic and quality infrastructure that this market now urgently needs, and Argos Data is how we put it to work for organizations building AI on a global scale,” Argos Multilingual CEO Alexander Ulichnowski said in the press release announcing Ison’s appointment. 

This emerging Data-for-AI industry — the market for datasets and human workflows used to train, adapt, align, evaluate, and deploy AI systems — is becoming a foundational layer of the AI economy.

Even as AI disrupts the traditional language solutions market, LSIs are finding they often have an edge when hiring the talent needed to produce datasets that cover many languages, dialects, and cultural contexts.

Slator research has found that about 10% of LSIs, including TransPerfect, RWS, Welocalize, and Acolad, established dedicated data-for-AI divisions. RWS’s TrainAI data-for-AI unit was one of its best-performing businesses in the first half of fiscal 2026, with the caveat that RWS said that growth may moderate over the rest of the year.

“Argos has been producing speech, language, and evaluation data for large technology companies since the neural machine translation era, well before generative AI made “AI data” a category of its own.” — Chris Phillips, Chief Innovations Officer, Argos Multilingual

An Evolving Market

Argos Data offers the Myriad platform to companies building AI systems. The platform helps teams “collect, annotate, evaluate, validate, and improve data for production AI systems,” allowing AI developers to tap into Argos’ over 80,000 vetted experts around the world, according to the Argos Data website. 

And the data-for-AI space continues to evolve. Phillips says “the center of gravity has shifted from simply producing training data to measuring and improving model behavior.”

Phillips expects three trends in the industry to accelerate over the next year: evaluation will become a continuous process, rather than a one-off, and will become increasingly multilingual; audio and multimodal data will become increasingly important; and human work will focus on “the point where automation stops”, such as in designing evaluation frameworks and handling edge cases.