Meta just dropped a bold new feature on Threads, and it is already causing a firestorm. The company is testing public AI replies through an official @meta.ai account that users can tag in posts and replies. But there is a catch. Threads users have discovered they cannot block the AI account, and more than a million posts are now flooding the platform in protest.
How the @meta.ai Account Works on Threads
Meta has created a dedicated Threads account with the username @meta.ai. Users can mention this account in any post or reply to get instant answers powered by artificial intelligence.
Think of it like having a search engine right inside your conversations. Someone could type, “Hey @meta.ai, why are people talking about the World Cup this month?” and the AI would jump in with a public reply. It works in the language of the original post, so users across different countries can get responses in their native tongue.
The feature is currently in early beta testing across five countries: Malaysia, Saudi Arabia, Mexico, Argentina, and Singapore. Meta plans to expand the rollout based on feedback from these markets.
Meta announced this update on May 12, 2026, through a revised version of its Muse Spark blog post from early April. The company sees this as a way to bring real-time context about trending topics and breaking news directly into Threads conversations.
Meta AI public replies feature testing on Threads app
Why Threads Users Are Pushing Back Hard
“Users cannot block Meta AI” became the number one trending topic on Threads within hours of the announcement. Over one million posts have been shared about this single issue.
The anger is simple. Unlike every other account on Threads, the @meta.ai account cannot be blocked. Users tried reporting it for spam, hoping to trigger the block option, but the button never appeared. The block simply did not go through.
Replies under the @meta.ai account’s first post tell the story clearly:
- “We don’t want you here.”
- “Give us the ability to block you, you cowards!”
- “Why can’t I block you?”
- “Either I get to block you or I uninstall this godforsaken app.”
Meta responded by saying users have other options. They can mute the account, tap “Not interested” on Meta AI posts, or hide AI replies on their own posts. But for many users, these workarounds feel like a half-measure when blocking is the standard tool for controlling what you see on social media.
This is not the first time Meta has faced this kind of pushback. Last year, the company launched AI-generated profiles on Instagram that also could not be blocked. Meta later called it a “bug” and shut the entire project down. Whether history repeats itself here remains to be seen.
How Threads AI Stacks Up Against Grok on X
The feature is clearly inspired by Grok, the AI assistant built into X by Elon Musk’s xAI. On X, users regularly tag @grok in posts and replies to fact-check claims, explain trending topics, or get quick context on viral content. It has become one of the most used features on the platform.
Here is how the two compare:
| Feature | Meta AI on Threads | Grok on X |
|---|---|---|
| How to use | Mention @meta.ai | Mention @grok |
| Reply visibility | Public by default | Public by default |
| Can users block it? | No | Yes |
| Safety guardrails | Stronger safeguards reported | History of controversial outputs |
| Data training | Not confirmed for this feature | xAI confirms using interaction data |
| Current availability | 5 countries (beta) | Global |
One major difference stands out. Grok has a troubled track record with safety. It has produced inappropriate content on multiple occasions, including biased rants and harmful material. Meta claims its AI has stronger safeguards in place, though only time will tell how the system handles the messy reality of public conversations.
On the data side, xAI openly says it uses Grok interactions to train its AI models. Meta has not confirmed whether conversations with @meta.ai on Threads will be used the same way. That silence is making some privacy-conscious users nervous.
Muse Spark Is Powering Everything Behind the Scenes
The Threads AI feature is just one piece of a much bigger puzzle. Meta is rolling out upgrades across its entire product lineup, all powered by Muse Spark, the company’s newest and most capable AI model.
Muse Spark was first introduced in April 2026 through Meta’s superintelligence lab. It is designed to handle complex tasks across every Meta product, from messaging to shopping to wearable tech.
Here are the key upgrades arriving alongside the Threads feature:
- Faster voice chats: Meta AI can now hold natural conversations with interruptions, topic changes, and language switching built in.
- AI vision: Users can point their phone camera at objects or locations and ask Meta AI questions in real time. Snap a photo of a snack shelf, and it can rank items by protein content.
- Shopping tools: Meta AI can now search Facebook Marketplace alongside web listings, showing both used and new products with a map view for local results.
- Smart glasses integration: Muse Spark is rolling out on Ray-Ban Meta glasses and upcoming Oakley Meta devices in the US and Canada over the coming weeks.
- Nutrition tracking: Users aged 18 and over in the US can log food through a photo or voice command on Meta smart glasses.
These updates show that Meta is not just experimenting with AI on Threads. It is embedding artificial intelligence deeply into every corner of its ecosystem.
The Bigger Question About AI on Social Media
Threads now has over 400 million monthly active users and recently surpassed X in daily mobile users for the first time. That is a massive audience, and the stakes for getting this right are high.
The backlash against @meta.ai echoes a pattern playing out across the entire social media industry. When Bluesky introduced its own AI assistant called Attie earlier this year, it became the second most blocked account on the platform within days. More than 125,000 users blocked it in a coordinated rejection of unwanted AI presence.
The message from users is consistent across every platform: they want control over their AI interactions, not forced participation.
Meta is betting that once people see the AI in action, answering questions about trending topics, providing real-time context during breaking news, and making conversations richer, the resistance will fade. That same bet has failed before. The company’s AI-generated Instagram profiles crashed and burned just months ago after similar outrage.
What makes this moment different is scale. With Muse Spark powering everything from Threads posts to smart glasses, Meta is going all in on AI integration. If the Threads rollout succeeds in those five test countries, expect @meta.ai to show up in feeds around the world very soon.
For now, millions of Threads users are making one thing very clear: they want the choice to say no. Whether Meta listens could define how the next chapter of AI-powered social media plays out, not just for Threads, but for every platform trying to bring bots into our conversations. What do you think about AI chatbots showing up in your social media feeds? Drop your thoughts in the comments below.
