Screening brief
What this video covers
At SlatorCon London 2026, Toloka CTO Artem Grigorev discusses how AI data needs have shifted from simple labeling tasks to requiring expert judgment and iterative involvement to improve models. He highlights that as AI agents become more capable, evaluation must move beyond single-instance labels to assessing behavior and decision-making over time.
The conversation serves as a primer on the importance of expert contributions in training and refining AI systems, arguing that data quantity alone is no longer sufficient and that complex, ongoing evaluation processes are increasingly necessary.
Summary and topic guide by HumanData.TV, based on the original publisher’s description and our editorial catalogue. This is not a transcript or independent verification of the speaker’s claims.
