WIRED senior writer Reece Rogers has ranked six AI agents on how well they complete basic tasks, testing products ranging from Meta's Muse to Anthropic's Claude Duet.
The comparison picks up a conversation that had been going on in the WIRED newsroom for months. Rogers, who covers software for the publication, said that last year he spent a lot of time in the office complaining to colleagues about AI agents. In his telling, they were simply bad at most of the things he asked them to do.
This year's test was meant to measure whether that picture still holds. Rogers evaluated each agent on basic tasks and ranked them from strongest to weakest. The excerpt available for this report does not include the full scoring details, so readers should refer to the complete article for the specific results and for the agent that finished first.
Rogers focuses on helping readers get more out of the technology they already use. Before joining WIRED, he covered streaming at Business Insider.
Cet article a été rédigé avec l'assistance de l'IA.
News Factory APP - actualités agentiques pour booster votre SEO et AEO.