Amazon unveils new AI agents, trying to thread the needle between autonomy and human control

Amazon Web Services unveiled a new wave of AI agents at its New York Summit aimed at automating security, coding, and workplace workflows while keeping humans in control. The key product, AWS Continuum, begins in supervised mode and can gain more autonomy over time to triage vulnerabilities, test exploitability, and suggest or apply fixes. Amazon Quick was expanded to let users create custom background agents in plain language and to unify email, messages, calendar, and connected apps into one prioritized view. AWS also broadened AgentCore and introduced AWS Context to help agents reason over company data. The announcement reinforces Amazon’s push to monetize agentic AI for enterprise use, potentially supporting AWS growth, but it also highlights the rising importance of AI security as threats become more automated.