A Downdetector report on the largest global outages in 2025 ranks AWS, PlayStation, and Cloudflare among the top three largest service outages of the year, based on the number of reports. According to the data, Downdetector received 17 million AWS reports, 3.9 million PlayStation reports, and 3.3 million Cloudflare reports.
However, the scale of an outage is not only measured by report counts, but by how many services depend on a single provider to function.
On November 18, 2025, and December 5, 2025, American company Cloudflare experienced downtime, which affected many of the Internet services millions of users worldwide rely on daily, including ChatGPT, Canva, Claude, and X (formerly Twitter). Interestingly, even Downdetector, the site many depend on to know which sites are down, was unavailable due to the Cloudflare issue.
Following the first downtime, Cloudflare revealed in a statement that the issue originated from an internal server error.
“The issue was not caused, directly or indirectly, by a cyber attack or malicious activity of any kind. Instead, it was triggered by a change to one of our database systems’ permissions, which caused the database to output multiple entries into a “feature file” used by our Bot Management system. That feature file, in turn, doubled in size. The larger-than-expected feature file was then propagated to all the machines that make up our network,” Mathew Prince, Cloudflare’s Co-founder & CEO, said in a statement.
While many people experienced the outages, few understand why tools from different companies and industries could all stop working at the same time. The answer lies in how much of the Internet depends on Cloudflare.
What is Cloudflare?
Cloudflare is an American company that offers a range of services for website speed and security, including Content Delivery Network (CDN) and other privacy features that protect sites from DDoS attacks and other malicious activities.
Think of Cloudflare as a middleman. Without the middleman, when a user sends a request to a website, the request goes directly to the website’s server and returns with content. If the website receives a massive amount of traffic, responses might be slow, or the site could crash.
As a middleman, Cloudflare routes every user’s request through its servers. Before it gets to the website’s server, it passes through one of Cloudflare’s data centres in 330 cities worldwide. With this expanse of data centers, when a user sends a request to a website like ChatGPT, it goes to Cloudflare’s servers, which deliver the content from the closest server to the user, making the site load faster.
Moreover, because of this process, websites can continue running despite heavy traffic.
Cloudflare’s position in the middle also blocks harmful traffic and malicious attacks before they reach the website.
These services are especially crucial to websites because they eliminate the need to install tools for speed and security individually and can plug into Cloudflare.
What Cloudflare does not do is host the data or applications for most websites. Companies like OpenAI, Meta, or X still run their own servers or rely on cloud providers such as Amazon Web Services, Google Cloud, or Microsoft Azure.
Instead, Cloudflare operates as a layer in front of these services, handling traffic before it reaches the website itself.
What happens during a Cloudflare downtime?
In both incidents, the problem did not begin with ChatGPT, Canva, or X themselves. Their servers did not suddenly fail, nor were they under direct attack. Instead, the shared layer they rely on to deliver content to users temporarily broke down.
Because Cloudflare sits between users and many websites, its downtime meant that requests could not reach their final destination. To users, this appeared as a widespread failure across unrelated platforms.
Why does so much of the internet depend on Cloudflare?
Cloudflare was founded in 2009, and today, it handles Internet requests for millions of websites and claims to serve 81 million HTTP requests per second on average.
While Cloudflare seems to be the choice for almost every website on the Internet, it has a few competitors, such as Akamai and Amazon CloudFront.
Cloudflare’s appeal is mainly practical. Its services are relatively easy to set up, often requiring only a change in DNS settings. For many websites, Cloudflare offers immediate improvements in speed and security without the need to build complex infrastructure in-house.
Cost is another factor. Cloudflare offers free and low-cost plans that are attractive to startups, media organisations, and small businesses. Even large companies use Cloudflare to offload traffic and protect against attacks.
Over time, this convenience has made Cloudflare a default choice for millions of websites. But what this inevitably means is that if the company encounters a fault, as it recently did, almost everything is likely to shut down until Cloudflare comes back up.
What this means for the future of the Internet
When many websites depend on the same infrastructure layer, failures are amplified. A single outage does not affect just one company or service; it affects many at once.
Because Cloudflare handles traffic for a wide range of platforms, its downtime can create the impression that the Internet itself is broken.
Some organisations reduce risk by using multiple CDNs or building fallback systems. However, while this approach improves resilience, it requires more technical expertise and higher costs that smaller organisations and websites cannot afford.
Cloudflare outages are rare, but when they happen, they reveal how the Internet depends on a few major infrastructure providers.
While outages are unlikely to disappear, it is essential to consider how prepared the Internet is to absorb them without total disruptions.