Vivold Consulting

Meta adds an 'AI Mode' search to Facebook that answers from public posts across its apps

Key Insights

Meta is rolling out a wave of AI features on Facebook, headlined by AI Mode - a search that uses Meta AI to answer plain-language questions with synthesized info pulled from public posts, Groups, and Reels. It follows Meta's Reddit-style Forum app and arrives alongside new AI editing tools and photo presets. The push aims to make Facebook stickier and to diversify revenue as Meta layers in paid subscription tiers.

Stay Updated

Get the latest insights delivered to your inbox

Meta wires generative AI deeper into Facebook

Meta has announced a batch of new AI features for Facebook aimed at changing how people find information, create content, and interact - the latest sign of the company pushing to catch up in the AI race and keep users engaged.

What's new

The headline is AI Mode, a new way to search Facebook that uses Meta AI to surface answers pulled from public posts across the platform, including Groups and Reels. Rather than scrolling through results, users can ask a question in plain language and get a synthesized answer drawn from what people are actually discussing. It follows Meta's quieter launch last month of Forum, a Reddit-style app with its own AI "Ask" tab that pulls from Facebook Group discussions. Both raise a familiar reliability question: because the AI summarizes everyday user posts rather than vetted sources, there's a real risk of outdated or misleading answers slipping through - a criticism already leveled at Google's comparable AI Mode.

Beyond search

Facebook also added creative tools - collage cutouts and transition effects for video montages, plus AI photo presets that let users swap clothes, hairstyles, and accessories (sports fans can virtually don a team jersey via an "AI Edit" option). These build on a steady run of recent additions: animated profile pictures in February, an AI auto-reply for Marketplace sellers in March, and an AI creator assistant earlier this month that suggests posting times and summarizes audience comments.

The strategy underneath

Taken together, the releases point to a broader plan: make Facebook's AI tools sticky and useful while diversifying how Meta makes money. The company recently launched global subscription plans for Facebook, Instagram, and WhatsApp starting at $3.99 a month, with more AI-related tiers reportedly on the way - so the feature blitz doubles as the on-ramp to a paid AI layer across Meta's apps.

Related Articles

Microsoft has become the lone conduit for OpenAI's models in China, profiting where their makers won't go

Microsoft has quietly become the main supplier of OpenAI's GPT models in China, selling them through Azure to the country's biggest internet firms even though OpenAI and Anthropic both refuse to sell there directly on IP and misuse grounds. Per Bloomberg, ByteDance is Microsoft's largest AI customer and is on track to spend over US$1 billion a year. A unique OpenAI contract lets Microsoft set its own overseas terms - and it's simultaneously testing a Chinese DeepSeek model for Western customers, taking margin on both sides of the trade.

HSBC's multi-year Google Cloud deal targets 200+ AI use cases, some worth $100M+ each

HSBC has signed a multi-year partnership with Google Cloud to build and deploy AI across wealth management, financial-crime risk, and internal decision support, using Gemini models and the Gemini Enterprise Agent Platform. The bank expects more than 200 AI use cases over two years, with selected ones each potentially returning over US$100 million. It builds on a deep existing base - 600-plus AI use cases and a Google-built financial-crime system screening 1.2 billion transactions a month.

The EU's new AI-labelling code previews Article 50 duties that become law on August 2

The EU has published a voluntary Code of Practice for marking and labelling AI-generated content, a practical playbook for meeting transparency duties that become law on August 2, 2026 under Article 50 of the AI Act. From that date, deepfakes and AI-generated text on public-interest matters must be labelled, and users must be told when they're talking to an AI. The code splits the work - model builders mark output in machine-readable form, deployers handle visible labelling - and companies have under two months to prepare.