Major Internet Outages: How AWS, Azure, and Cloudflare Disrupt Global Connectivity

AWS, Azure, and Cloudflare are critical components of the Internet's infrastructure. However, they all contributed to significant Internet outages within a month of each other: AWS on October 20th, Azure on October 29th, and Cloudflare on November 18th. Each incident brought a significant portion of the Internet and various businesses to a standstill, as many rely on at least one of these providers.

Back in the 1990s, when the Internet was emerging, it was designed to be resilient, capable of surviving even a nuclear war. Today, however, a single typo in the code can disrupt large parts of the World Wide Web for an unpredictable amount of time. This is especially concerning given that there are approximately 5.45 billion users globally, and these outages (from AWS, Azure, and Cloudflare) are not limited to the U.S.; they impact users worldwide.

AWS and Azure

AWS, Microsoft Azure, and Google Cloud Services are the three leading providers in cloud computing, with AWS holding 29% of the market, Azure 20%, and Google Cloud Services 13% as of Q3 2025. The remaining 38% is shared among Alibaba, Oracle, Salesforce, IBM, and Tencent. When AWS experienced downtime, it impacted 29% of the global Internet. This included essential resources for schools, such as ClassLink, which connects various services like Canva, Infinite Campus, online textbooks, and other educational materials, as well as Zoom, which schools depend on for virtual classrooms. Similarly, Azure's outage affected Canvas, a web-based platform widely used by educational institutions. Such disruptions not only hinder workflow but also negatively impact education, potentially putting students at a disadvantage, even if only for a short period.

Cloudflare

The Cloudflare outage was linked to issues in cloud infrastructure, specifically related to cybersecurity and traffic management. This incident bore unsettling similarities to the 2024 Cloudstrike outage, which affected 8.5 million Windows systems. The Cloudflare disruption affected about 20% of all Internet traffic, impacting users on Windows, Linux, and Apple devices. Popular websites like X and Spotify were rendered inaccessible, preventing users from engaging in activities like scrolling through social media or listening to music. Additionally, it took down ChatGPT, which holds a 60.5% share of the AI chatbot market. This outage severely disrupted transit systems, affecting services such as NJ Transit and France’s SNCF.

Conclusion

These outages highlight a significant concern: the fragility of the internet. When a simple typo can disrupt 20% of the internet and cost businesses over $360 million in just six hours, the issue of centralization comes to the forefront. While centralizing cloud computing can facilitate quicker trend analysis, such outages affect a larger number of people than ever before. As artificial intelligence becomes part of the discussion, companies must consider what measures they are taking to prevent these disruptions.