Outage or cyberattack? How to tell when the internet breaks
Five IT failures hit the US in four business days, and plenty of people suspected hackers. Analysts say only about one in a thousand outages is an attack.

Five big IT failures in four business days is enough to make anyone suspect a cyberattack. A Microsoft Exchange outage hit thousands of companies, a Federal Aviation Administration technical problem grounded planes, a software issue disrupted JPMorgan Chase, United Airlines had another system glitch, and a classified NSA network went down, leaving staff at Fort Meade waiting at home.
With the US at war with Iran and fears of AI-enabled hacking rising, plenty of people read the pile-up as more than coincidence. So far, though, there is no evidence that these failures are linked, or that hackers caused any of them.
The starting point for analysts is simple: assume an accident. Doug Madory, head of Internet analysis at Infoblox, watches up to a dozen network outages a day. Only around one in a thousand, he says, turns out to be the work of hackers. Cloudflare CEO Matthew Prince calls most of what he sees right now "standard background noise." João Tomé of IPinfo says the usual culprits are a bad update, a configuration change, a cable cut or a power failure.
How can you tell if an outage is a cyberattack?
First, check what the affected organisations say. Big companies post updates on verified social media accounts and in the newsroom section of their sites. Many also run status dashboard pages, which show which parts of a service are working. AWS, Google Cloud and Microsoft Azure all have one.
The picture can be muddy. A failure at a cloud provider such as Amazon Web Services can knock over its customers. Some services only look broken because they depend on one that is down. This week's bank trouble seems to be an example: several banks had problems, but all of them involved transactions with JPMorgan Chase. False reports can pile more confusion on top.
Second, look at independent network data. Amanda Meng of Georgia Tech, who helps run the Internet Outage Detection and Analysis service (IODA), points to its data, plus Cloudflare Radar and Watch the Net. Her team looks at when an outage began and ended and which networks were affected, then compares that with news reports. Madory also posts his findings on social media.
The short version: unless there is concrete evidence of an attack, treat it as a mistake.
Apps worth a look
Tap the right tiles, add them up, hit the target. Easy to learn, hard to master. Free, for all ages.
Talk to an AI conversation partner; every sentence comes back corrected and explained in your language. 9 languages, A1 to C2.
Open EDIFACT and X12 messages, see every segment explained, convert to JSON/XML, export PDF and Excel. Runs locally, no upload.
Pick who it is for, the occasion and their personality. The search engine reads product descriptions from local shops and ranks what fits. Free, no sign-up.
More from this day
See all →
Google AI Mode temporary threads: incognito-style search

Arcom rules YouTubers must give all 2027 candidates equal time

Kerch Bridge strike: Ukraine steered missiles without Starlink

OVHcloud Olala mini: a free AI model that runs in 8 GB of RAM
More in: Security
See all →
Apple backdoor order: UK tribunal forces government to confirm it

JPMorgan Chase data breach: vendor hack leaks customer emails

Rakuten Securities data leak: why real customer data sat in a test system
