Meta’s new ‘AI Mode’ on Facebook pulls from public info across its platforms
- company Google
- company Meta
- location Facebook Groups
- location Groups
- location Reels
Meta is rolling out an AI-powered search feature on Facebook called AI Mode that synthesizes answers from public posts, Groups, and Reels rather than returning a list of links, the company announced Monday [1]. The feature lets users ask questions in plain language and receive a single, AI-generated response drawn from discussions happening across the platform [1]. It follows Meta’s quiet launch last month of Forum, a Reddit-style app that includes its own AI “Ask” tab, which also pulls answers from Facebook Groups conversations [1]. Both tools raise questions about reliability, because the AI summarizes content from everyday users rather than vetted sources — a concern already raised about Google’s AI Mode on Reddit [1]. Facebook reported approximately 3.07 billion monthly active users worldwide as of December 2023, and was ranked the third-most-visited website globally as of July 2025 [2]. The platform has long faced criticism over the spread of misinformation, fake news, and hate speech [2]. Meta’s decision to build a search experience atop user-generated chatter places those content-moderation challenges at the center of a product that could shape how millions of people find information. Alongside AI Mode, Facebook added editing tools for video montages — including collage cutouts and transition effects — and AI-powered photo presets that let users alter their appearance with different clothes, hairstyles, and accessories [1]. Sports fans can virtually wear a team jersey by tapping the “AI Edit” icon in Stories and selecting “Wear It,” or by choosing “Restyle profile picture with AI” and “Wardrobe” from their profile picture settings [1]. The updates extend a series of AI releases Meta has shipped on Facebook in recent months. In February, the company introduced animated profile pictures that bring still photos to life with waves or virtual party hats [1]. In March, it added an AI feature to Facebook Marketplace that automatically replies to buyer messages on sellers’ behalf [1]. Earlier in June, Facebook launched an AI assistant for creators that offers personalized suggestions — including optimal posting times and summaries of audience comments — based on a creator’s content and performance history [1]. Meta’s broader strategy ties these AI tools to revenue diversification. The company recently launched global subscription plans for Facebook, Instagram, and WhatsApp starting at $3.99 a month, with more AI-related subscription tiers reportedly on the way [1]. Instagram, which Facebook acquired for approximately $1 billion in cash and stock in April 2012, now counts more than 1 billion users and supports 33 languages [3]. WhatsApp, another Meta property, is not detailed in the primary announcement but forms part of the subscription bundle [1]. Meta’s AI push arrives as the open-weight model economy undergoes a documented rebalancing. An analysis of 851,000 models and 2.2 billion downloads on the Hugging Face Model Hub found that U.S. open-weight industry dominance by Google, Meta, and OpenAI has declined sharply in favor of unaffiliated developers, community organizations, and, as of 2025, Chinese industry players such as DeepSeek and Qwen [7]. The same study noted a 17-fold increase in average model size and rapid growth in multimodal generation, quantization, and mixture-of-experts architectures [7]. Meta is shipping consumer-facing AI features while the underlying model landscape fragments and competitive pressure intensifies.
product-launchapplication
Background sources we checked (10)
- en.wikipedia.org ↗ Facebook is an American social networking service owned by the American technology conglomerate Meta Platforms. It was founded in 2004 by Mark Zuckerberg, along with his Harvard College roommates and fellow students Eduardo Saverin, Andrew McCollum, Dustin Moskovitz, and Chris Hu…
- en.wikipedia.org ↗ Instagram is an American photo and short-form video sharing social networking service owned by Meta Platforms. It allows users to upload media that can be edited with filters, be organized by hashtags, and be associated with a location via geographical tagging. Posts can be share…
- en.wikipedia.org ↗ Telegram (also known as Telegram Messenger) is a cloud-based, cross-platform social media and instant messaging (IM) service. It launched for iOS on 14 August 2013 and Android on 20 October 2013. It allows users to exchange messages, share media and files, and hold private and gr…
- en.wikipedia.org ↗ Google LLC ( , GOO-gəl) is an American multinational technology corporation focused on information technology, online advertising, search engine technology, email, cloud computing, software, quantum computing, e-commerce, consumer electronics, and artificial intelligence (AI). It…
- arxiv.org ↗ This paper presents FormationEval, an open multiple-choice question benchmark for evaluating language models on petroleum geoscience and subsurface disciplines. The dataset contains 505 questions across seven domains including petrophysics, petroleum geology and reservoir enginee…
- arxiv.org ↗ Since 2019, the Hugging Face Model Hub has been the primary global platform for sharing open weight AI models. By releasing a dataset of the complete history of weekly model downloads (June 2020-August 2025) alongside model metadata, we provide the most rigorous examination to-da…
- arxiv.org ↗ Large language models (LLMs) have raised hopes for automated end-to-end fact-checking, but prior studies report mixed results. As mainstream chatbots increasingly ship with reasoning capabilities and web search tools -- and millions of users already rely on them for verification …
- arxiv.org ↗ Despite widespread deployment of Large Language Models, systematic evaluation of instruction-following capabilities remains challenging. While comprehensive benchmarks exist, focused assessments that quickly diagnose specific instruction adherence patterns are valuable. As newer …
- arxiv.org ↗ The use of large language models (LLMs) in hiring promises to streamline candidate screening, but it also raises serious concerns regarding accuracy and algorithmic bias where sufficient safeguards are not in place. In this work, we benchmark several state-of-the-art foundational…
- en.wikipedia.org ↗ Meta elements are tags used in HTML and XHTML documents to provide structured metadata about a Web page. They are part of a web page's head section, the term meta indicating that they are a form of self-reference. Multiple Meta elements with different attributes can be used on th…