Meta has switched on Meta AI inside Threads direct messages for all users globally, quietly completing its takeover of every private inbox across its social empire and putting its chatbot in direct competition with ChatGPT for daily query volume.
The rollout landed on July 27, 2026. Threads users can now open a private conversation with Meta AI, share posts, images, links, and videos directly into the chat, and get responses that stay completely off their public feed. Meta framed it as giving people "greater flexibility in how and where they interact with Meta AI." That's the official line. The real story is simpler: Threads was the last surface Meta owned where its AI wasn't already embedded, and now it isn't.
Meta AI is already woven into Facebook, Instagram, WhatsApp, and Messenger DMs. With Threads now added, Meta has closed the loop. Every major inbox across its platforms feeds into the same AI assistant, and with Threads reporting roughly half a billion users, this is a meaningful single-day expansion of Meta AI's reach. The company already claims 1.2 billion monthly active users for Meta AI across its apps, a figure it hit in roughly 18 months, according to data from AI Funding Tracker. ChatGPT crossed 1 billion monthly active users in June 2026. The race is close, and inboxes are where the next leg of it gets run.
Meta first tested AI in Threads publicly in May 2026, limited to five countries: Argentina, Malaysia, Mexico, Saudi Arabia, and Singapore. That version worked similarly to Grok on X, where the bot replies directly under posts in the public feed. As TechCrunch noted at the time, it was a visible, social-facing integration. The DM version is different in character. Private messages carry more intent, more context, and more sensitive detail than anything a user would post publicly. Someone asking Meta AI about flight prices, medical symptoms, or relationship advice in a DM is handing over a richer signal than almost any public interaction could provide.
Meta updated its privacy policy in late 2025 to cover its generative AI products explicitly, stating that interactions with Meta AI can be used to personalize and improve its services, which feeds into ad targeting. According to a detailed breakdown published by Dataslayer and confirmed by Auditsocials' analysis of Meta's policy changes, AI-derived signals - what topics you ask about, what products you reference, what links you share - flow into the same unified interest profile that shapes ad delivery across its platforms. A user asking Meta AI about running shoes generates a footwear and fitness signal. That's not a bug. It's how the business works.
Meta has been careful to say that person-to-person DMs remain private and are not collected for ad targeting. But conversations with Meta AI are a different matter: you're talking to the product, not to another person, and the product is listening in the way all AI assistants listen. That distinction matters, and it's one most users won't notice when they tap the new AI icon in their Threads inbox.
The competitive pressure behind the move
Don't mistake this for a features update. It's a distribution play. Meta AI's daily active users sit at around 40 million, with the majority of interactions happening on WhatsApp, according to AI Business Weekly. Google's Gemini app passed 750 million monthly active users earlier this year, as TechCrunch reported in February, and ChatGPT is now approaching 900 million weekly active users. The gap between monthly reach and daily habit is where the real competition lives, and that's what Meta is trying to close by putting its AI one tap away inside the inbox users already check constantly.
The Threads DM integration matters for another reason too. Threads has been Meta's answer to X since 2023, and it's been building steadily. Adding AI to the DM layer gives Threads something X's Grok integration doesn't have in the same form: a private, conversational AI experience separate from the public timeline noise. Whether users actually want that, or whether they'll engage with Meta AI in their Threads inbox the way they do on WhatsApp, is still an open question. WhatsApp's share of Meta AI interactions, at 63% of all usage, suggests the messaging-native context is where people are most comfortable treating an AI as a utility.
Frankly, the inbox is the most valuable real estate in social media. It's where people are least performative and most honest about what they actually want. Meta has spent three years building Threads as an alternative to X, and it's now layering its AI directly into that space at scale. The company didn't do this because the Threads feed needed more features. It did it because the DM inbox is where intent lives, and intent is what the ad machine runs on.
Also read: Dario Amodei says he doesn't want open-weight AI banned but fears what China just released for free • The EPA just gave AI data centers a pollution exemption that may not survive a courtroom • OpenAI's first hardware product sold out in 12 hours and is now flipping on eBay for $1,850
