Meta Launches ‘AI Mode’ Search on Facebook Meta is turning years of casual posts, group discussions and Reels into something closer to a real-time, AI-driven knowledge base on Facebook — raising fresh questions about accuracy, consent, and control over user content.

Early AI experiments set the stage

In recent months, Meta has steadily embedded AI tools across Facebook. It rolled out animated profile pictures in February and an AI tool in March that auto-replies to Marketplace buyer inquiries. Earlier in June, a creator assistant began offering help with captions and engagement suggestions. Meta also quietly launched Forum, a Reddit-style app that includes an AI “Ask” tab pulling answers from Facebook Group discussions, previewing how AI might mine social content at scale.

June 15: Meta unveils Facebook “AI Mode”

On June 15, Meta announced “AI Mode,” a new search experience on Facebook that uses Meta AI to surface answers from public posts, including Groups, Reels and Marketplace listings. Instead of a list of links, users type a natural-language question and receive synthesized responses “based on what people are actually discussing.”

Technically, the feature is powered by Meta’s Muse Spark AI model, which generates results from publicly shared content across the social network. AI Mode now appears alongside existing search filters like “People” and “Marketplace,” and Meta says the system “uses Meta AI to give you answers grounded in what people are saying publicly across our apps,” with plans to eventually cite content from Instagram, Facebook and Threads.

Benefits, risks, and unanswered questions

Supporters frame AI Mode as turning Facebook into an efficient, conversational search engine that can recommend products, surface Group advice and pull relevant Reels without users scrolling through long result lists.

But critics note that, like Google’s AI overviews, the tool compresses user-generated content into answers, potentially bypassing original posts and communities. Because it relies on everyday public chatter rather than vetted sources, “there’s a real risk of outdated or misleading information slipping through,” especially as Meta races to catch up in AI. Privacy and control are also flashpoints: Meta has not clarified whether group admins or users can opt out, or how it handles content that was once public but later deleted or made private — “significant gaps for a feature that treats user content as raw material for an AI system.”

As the rollout begins in the U.S., Meta is betting that deeper AI integration — alongside photo and video editing effects and potential subscription tiers — will boost engagement, even as the company faces mounting scrutiny over how it repurposes the social lives of its users into training data.

Continue reading https://foxvector.com/stories/019ecedc-55d5-2f67-7133-34860ec80057