Amazon AI for Security Research Awards 2026

Amazon Research Awards opens a new call for proposals focused on AI for information security. This Spring 2026 opportunity invites researchers to tackle key challenges in securing digital systems with artificial intelligence. With a deadline of May 6, 2026, it’s a chance for academics to secure funding and resources from Amazon.

Program Overview

The Amazon Research Awards AI for Information Security Call for Proposals – Spring 2026 aims to support work on tough problems in information security. Amazon wants to push forward solutions that use AI to make systems safer. They focus on agentic AI, which means AI systems that can act on their own in security tasks.

This program builds on Amazon’s efforts to blend AI with security practices. Researchers can propose ideas that fit specific areas of need. Awards help fund projects that could lead to real-world improvements in cloud and device protection.

Key Research Topics

Amazon lists several topics for proposals. Each one targets a gap in current security tools. Here they are in detail:

Trustworthy and reliable agentic AI for security operations: This includes ways to test AI agents in real security scenarios. Researchers can explore methods to check if these agents work well under pressure.
AI agent access governance: Focus on policy checks, safety barriers, and smart authentication. The goal is to control what AI agents can access without full trust.
Threat, intrusion, and anomaly detection: Apply this to agentic AI systems and cloud setups. Detection tools need to spot issues fast in complex environments.
AI-powered incident response and red-teaming: Use AI to handle attacks or simulate them. Red-teaming tests defenses by acting like hackers.
Vulnerability detection and remediation: Agentic AI can find and fix weaknesses in software. Proposals should show how AI speeds up this process.
Building and optimizing foundation models: Improve AI models for security through training steps like pre-training and fine-tuning. Aim for better results on security jobs.
Security of agentic AI systems: Cover toolchains, frameworks, and issues like confused deputy problems. This means preventing AI from misusing permissions.
Securing generative AI and foundation models: Protect training data with anonymization or privacy techniques. Stop leaks from models after training.
Reinforcement learning for information security: Train AI through trial and error for security tasks. This fits dynamic threats.
Scalable graph modeling and anomaly detection: Handle large networks to find odd patterns. Graphs map connections in data.
Learning with limited or noisy labels: Work with weak supervision for security data. Real-world labels are often imperfect.
AI for malware analysis and detection: Target cloud or devices. AI can break down threats faster than humans.
AI-assisted secure code generation: Help write safe code or upgrade old codebases. Reduce bugs that lead to breaches.
Generative AI for cloud infrastructure compliance: Generate checks to meet rules in cloud setups.

Proposals must align with one or more of these areas. Amazon encourages fresh ideas that build on these themes.

Grant Details and Benefits

Selected projects receive solid support. Principal Investigators get up to $80,000 USD in unrestricted funds on average. This covers direct costs like equipment or travel, listed in USD without overhead.

AWS Promotional Credits add up to $40,000 USD on average. These help run experiments on Amazon’s cloud. Training comes next, with tutorials and sessions led by Amazon experts.

Awards are one-time gifts. The panel sets the final amount based on the proposal. Budgets stay simple and focused.

Who Can Apply

Eligibility keeps things clear. The Principal Investigator must be a full-time faculty at an accredited school. Or, a permanent researcher at a non-profit with legal status like a 501(c)(3) in the US.

Applicants need to be at the age of majority in their area. Each person submits just one proposal per cycle. Teams can join, but the PI leads.

How to Submit

Start by visiting the Amazon portal. Click the submission link to log in or create an account. Prepare a budget list and project details.

Full info lives on the official page. Check there for forms and guidelines. Submit by May 6, 2026, to stay in the running.

Frequently Asked Questions

What is the deadline for the Amazon Research Awards Spring 2026?

The deadline to submit proposals is May 6, 2026.

Who can apply for these awards?

Full-time faculty at accredited schools or permanent researchers at non-profits with legal status can apply as Principal Investigators.

What funding and benefits are offered?

Awards provide up to $80,000 USD in unrestricted funds and $40,000 USD in AWS Promotional Credits on average, plus training from Amazon experts.

What are the main research topics?

Key topics include trustworthy agentic AI for security, threat detection, vulnerability remediation, and securing generative AI models.