Article Earlier this year, on October 20, 2025, Amazon Web Services (AWS) experienced a massive outage in its US-East-1 region, triggered by a complex DNS failure and cascading automation error that disrupted Internet services worldwide.   Apps, websites and even smart-home devices were affected: platforms such as Snapchat, Reddit, Duolingo and Fortnite went offline, and services including voice assistants and banking apps suffered outages.   For individuals with disabilities, the effects of this disruption were more acute than for able-bodied users. While many people experienced temporary inconvenience, users who rely on cloud-based assistive technologies — such as screen-reading software, voice-controlled devices, remote tele-health platforms, and smart-home mobility aids — faced a threat to their communication, autonomy and everyday independence when those underlying services faltered. The outage serves as a striking reminder that our accessible technologies depend not just on good design, but on resilient infrastructure. This incident underscores the need for inclusive-technology providers, developers and accessibility advocates to incorporate resilience into their design thinking. It’s not enough to build accessible tools — we must build systems that assume the “cloud might fail.” Redundancy, offline fallback modes, multi-cloud strategies and thoughtful architecture become as important as interface design. When even able-bodied users felt disconnected, imagine how far worse the impact can be for people whose ability to connect, work or live independently hinges on digital services.

Question to Consider How can developers, assistive-technology providers and organisations build systems so that people with disabilities are not disproportionately impacted when major cloud services fail?

My Thoughts

In my view, this outage is a wake-up call for the accessibility community. It’s easy to celebrate how the cloud enables connection, independence and participation—but far less common to ask what happens when that infrastructure fails. For people with disabilities, losing access to a “normal” service might mean losing a lifeline: their ability to speak, control their environment, join a meeting, or engage in their career. The fact that able-bodied users were affected underscores just how pervasive the dependency is, but the consequences for disabled users are deeper. In building inclusive systems, we must design for when things fail, not just for when they work. That means offline capabilities, alternate paths, decentralized architectures and the principle that assistive systems shouldn’t rely solely on one provider’s uptime. True inclusion means designing not just for access, but for resilience, reliability and continuity of service — because autonomy and connection matter every moment, not just when the internet is cooperating.

Sources

    1.    “Amazon’s cloud unit reports outage that disrupted businesses worldwide” – Reuters, Oct 21 2025.       2.    “Amazon reveals cause of AWS outage that took everything from banks to smart beds offline” – The Guardian, Oct 24 2025.       3.    “Amazon explains how its AWS outage took down the web” – WIRED, Oct 25 2025.