The social media platform X, formerly known as Twitter, has experienced a widespread global outage on Friday, January 16. This latest technical failure comes just three days after thousands of users were locked out of the site earlier this week.
Reports began flooding in shortly after 4.30pm CET, with over 77,000 users flagging issues on the tracking site Downdetector. Visitors to the platform were met with a variety of error messages, including “connection timed out,” “something went wrong,” and “this page isn’t working.”
While the site’s host, Cloudflare, reported that its systems were operating normally, users attempting to access X saw notices indicating that the platform’s own servers were experiencing a critical error.
A Pattern of Disruptions
This disruption marks the second time this week that the platform has faltered. On Tuesday, January 13, more than 25,500 users reported similar access issues. Since the $44 billion acquisition of the company in late 2022, X has weathered a series of significant technical stumbles:
- March 2025: A massive outage resulted in over 1.6 million reports on Downdetector, which was later attributed to a “massive cyberattack.”
- May 2025: A data centre failure disrupted service for thousands across the U.S., Canada, and Germany.
- November 2025: A faulty security configuration involving Cloudflare briefly crashed the platform’s traffic system.
Infrastructure and Cost-Cutting
Industry analysts have frequently pointed to aggressive cost-cutting measures as a primary factor in the platform’s declining stability. Following the 2022 takeover, the company’s engineering staff was slashed from approximately 7,500 to fewer than 2,000 employees.
In addition to personnel cuts, several physical infrastructure changes were made to save an estimated $1 billion:
- Data Centre Closures: The Sacramento data centre was abruptly closed, and the Atlanta facility was significantly downsized.
- Hard-coded Errors: Following the Sacramento closure, it was revealed that the site’s code contained roughly 70,000 hard-coded references to that specific location, leading to persistent “broken” features.
- Cloud Spend: Expenditures on Google Cloud services were sharply reduced, further straining the platform’s remaining resources.
As 5pm Friday afternoon, X has not issued an official statement regarding the specific cause of today’s blackout. Service appears to be slowly returning for some, though many users continue to report intermittent failures.
