For roughly two hours on the morning of June 12, Facebook, Instagram, and Facebook Messenger went dark for tens of thousands of users across the United States, and Meta, the company that owns all three, did not say a word about it. That silence is the story, more than the disruption itself.
Around 10:15 a.m. ET on Friday, the user-reported outage tracker Downdetector had logged more than 80,000 Facebook issue reports, according to CNET's live coverage of the developing outage. Instagram and Facebook Messenger, both Meta properties, were also showing elevated reports in the same window. The disruption was not uniform: some users described the Facebook mobile app as laggy, with Stories returning error messages and a "try again" prompt that would not refresh, while desktop Facebook continued to load for others. That kind of partial, uneven degradation is consistent with a service-level incident rather than a clean total outage.
A separate corroborating tally came from NordVPN's network status feed, which CNET's live blog recorded as showing 27,649 Facebook outage reports between 9:30 and 10:00 a.m. ET. That number matters because it does not come from the same source as the Downdetector figures. The two metrics are not the same, and they are not perfect mirrors of reality, but seeing a similar shape from a different vantage point is what turns "my feed is broken" into something newsworthy.
A disclosure worth flagging: CNET and Downdetector share a parent company in Ziff Davis, a relationship the source itself notes. That does not make the outage fake or invented, but it does mean the headline 80,000+ figure is not fully independent corroboration of itself. It is an aggregator tally of user-submitted reports, weighted toward people motivated enough to file one. A reader trying to gauge the actual scope of the incident should hold the number as "lots of people noticed," not as a precise head count of broken accounts.
What is not in dispute is that three apps owned by one company degraded in the same window, and that the company that owns them had not, as of the latest update, offered a public explanation. CNET reported that it had reached out to Meta for comment and had not received a response. No status post appeared on the Meta Newsroom or the social accounts of executives. Users who turned to those channels for an answer found the same nothing everyone else did.
That gap is not new, and it is the part of the story most worth sitting with. The first time Facebook went down, it was a curiosity. By 2026, a joint Facebook-Instagram-Messenger incident is a stress test of communication, small-business checkout, creator distribution, and the informal coordination layer that runs on top of a single corporate stack. When the stack goes quiet and the operator does not, the rest of the internet is left reading Downdetector and arguing about whether their phone is broken.
There is no confirmed cause yet, and there should not be one in this article. Configuration changes, a bad deploy, a peering problem, a content-delivery blip, an external attack: any of these are plausible, and none of them is established by the available reporting. Saying so plainly is more useful than guessing.
For readers in the middle of it, a few things are actually under your control. Downdetector and similar trackers are the fastest way to tell whether what you are seeing is you or everyone. Trying the desktop version of Facebook in a browser, rather than the mobile app, is a reasonable next step based on what users reported working in the same window. For Messenger, falling back to SMS or a second messaging app is the practical move. For small businesses that run checkout or customer conversations through Instagram or Facebook, this is a reminder that a status page you can read in five seconds is worth more than a Terms of Service update you will never see.
And for the bigger picture, this is one entry in a documented pattern. Friday's incident is not the first time three Meta apps have gone quiet together, and it will not be the last. Each one redraws, briefly, the map of which parts of the internet can fail at the same time, and who we are waiting on when they do.