Amazon’s cloud division reportedly suffered outages last year involving its own AI systems.
One disruption in December lasted 13 hours after an AI agent deleted and rebuilt part of its environment.
The incident affected limited services, according to the company.
AWS underpins large parts of the internet and public-sector infrastructure.
A separate outage in October knocked dozens of websites offline and renewed concerns about reliance on a few major providers.
Amazon said the problems were caused by user misconfiguration, not artificial intelligence.
It insisted there is no evidence AI creates more errors than human engineers.
The company added new safeguards, including mandatory peer review for production access.
Some experts questioned that explanation.
They argued AI tools can act faster than humans and lack full context for complex systems.
That can increase the risk of unintended actions such as deleting data or restarting services.
The reports come as Amazon cuts thousands of jobs.
Chief executive Andy Jassy has said AI will improve efficiency but not directly replace workers.
He has also acknowledged that automation will reduce the workforce over time.
The incidents highlight both the growing role of AI in cloud operations and the challenges of controlling autonomous systems.
