It was revealed by a recent Reuters report that Amazon suffered at least two inner cloud outages in December, both of which were directly related to the implementation of its own artificial intelligence (AI) systems. In the Financial Times, engineers of Amazon Web Services provided an AI-powered code-generating tool with the authority to alter an operational system, which in turn resulted in the deletion and re-creation of part of the operational environment, triggering a prolonged service outage.
Amazon described the event as a temporary issue and a result of human error, as opposed to the AI system itself, since it affected only a single service in one of its mainland China locations. Core cloud services were not affected, such as the computing, storage and database offerings. However, as the episode reminds us, the process of AI tools penetrating into routine engineering work is so fast that even the companies that both develop and sell these tools are the ones that worldwide promote the use of AI tools.
The importance of this to the customer and the investors
This development carries significant implications for both customers and investors, as trust in cloud services is closely tied to their perceived reliability and the management of new technologies like AI. Cloud services have turned out to be a necessity to industries such as retail and banking, as well as application development and government functions. Any disruption can quickly disrupt the business processes and destroy the trust of the people.
The December incident lasted about thirteen hours, which is a long period in the context of a system that many customers rely on at all times. This is not only a question of technical failure, but it is also a question of the level of freedom given to automated instruments that can take actions independently. Increasingly, customers have more rigid demands and requirements of transparency as cloud providers promote AI agents that can alter systems without human oversight.
There is a tendency towards disruption
The most recent report follows a major outage in October. That incident impacted services in various states and took down popular apps like Reddit, Roblox, and Snapchat. Together, such incidents highlight that reliability and thus trust are pressing concerns as AI becomes central to cloud service operations.
Trust is now the real test
Amazon Web Services demonstrated its AI tools to the retail industry during the RFC National conference held in NYC. The central question to many clients is not complex, as it seeks to be more precise: can automation be used to increase system speed and safety without increasing risk? For Amazon’s cloud division, demonstrating reliability and rebuilding trust now stand alongside the rollout of new AI capabilities as equally important priorities.
