It works on its own: Amazon introduces three new AI agents

Amazon Web Services (AWS) has announced the release of three new artificial intelligence (AI) agents, including Kiro AI, which needs minimal supervision.

During the company’s event, AWS re:Invent 2025, Amazon unveiled three new AI “frontier agents,” which are supposed to handle security and development tasks.

One of three agents is Kiro AI, which “works independently on development tasks, maintaining context and learning from every interaction,” notes the company.

Kiro AI can work autonomously for days with minimal human intervention.

The AI software coding agent is created according to Amazon’s previously released AI tool for code generation.

The frontier agent is autonomous, meaning that it can work on tasks without requiring human supervision.

According to the company, “AI can take a high-level task description, figure out the implementation plan, write code across multiple repositories, run tests, and create pull requests, all while you work on other things.”

What makes it stand out from other agents is that it can work on a task for days without needing a person to remind it of what it needs to do.

The Kiro agent can also watch and learn from existing code, enabling it to understand how to accomplish tasks in a similar manner to what has already been done.

After the agent learns how a team performs certain tasks and is asked to take on a task, it is then able to figure out how to do it independently.

One of the recently unveiled frontier agents is the AWS Security Agent, which can identify security problems during code development, test after it's complete, and provide possible fix solutions.

The third AWS agent, the DevOps Agent, can then be used to “resolve and proactively prevent incidents, continuously improving reliability and performance.”

Unlock more exclusive Cybernews content on YouTube.