Akeyless Launches Industry-First Solution for Securing AI Agent Identities

Read full announcement here: https://www.akeyless.io/pr/akeyless-unveils-breakthrough-solution-to-secure-ai-agent-identities/?utm_source=nhimg

NEW YORK and TEL AVIV, Israel, Oct. 29, 2025 /PRNewswire/, Akeyless, a global leader in secrets management and machine identity security, has unveiled its groundbreaking AI Agent Identity Security solution, purpose-built to protect the growing wave of autonomous AI systems. The announcement marks a major step forward for enterprises seeking to balance AI innovation with identity and access security.

Addressing the AI Agent Identity Crisis

As over 95% of enterprises plan to adopt AI agents within the next year, organizations face a new and complex identity explosion. Each AI agent, capable of making independent access decisions, introduces additional risk when identities are not verified or monitored. Security researchers have already discovered exposed secrets and API keys in publicly distributed AI extensions, highlighting the danger of credential leakage and malicious prompt injection.

Former NSA Director Admiral Mike Rogers emphasized that “AI Agents have an inherent design flaw—they are aware of their own identities and credentials.” Combined with their expanding authorization across cloud workloads, this creates an ever-growing attack surface that traditional, human-centric IAM tools cannot secure.

A New Secretless Identity Model for AI Agents

To meet this challenge, Akeyless introduces a secretless, identity-first security model that eliminates static credentials entirely. Through Akeyless SecretlessAI™, enterprises can enforce Just-In-Time, identity-based authentication for every AI-driven connection—ensuring ephemeral, verifiable, and auditable access without embedded secrets.

Building on this foundation, Akeyless is launching a full suite of AI Agent Identity Security capabilities that redefine how organizations authenticate, authorize, and govern their AI systems.

Akeyless AI Agent Identity Provider (IdP)

Unverified AI agents operating across hybrid and multi-cloud environments pose accountability and data integrity risks. The new Akeyless AI Agent IdP enables enterprises to assign verifiable, federated digital identities to each AI agent—allowing them to securely authenticate across AWS, Azure, GCP, and major AI ecosystems such as OpenAI, Anthropic Claude, Google Gemini, and Grok. IDE integrations for VS Code, GitHub Copilot, and Cursor further extend these protections into developer environments, replacing static API keys with dynamic, short-lived credentials.

AI Agent Privileged Access Management

To counter the dangers of over-privileged or rogue AI behavior, Akeyless AI Agent Privileged Access Management (PAM) enforces Zero-Trust and least-privilege controls for autonomous agents. Continuous monitoring and behavioral governance ensure every AI action is traceable, with the ability to instantly revoke access or terminate rogue sessions.

AI-Powered Insights for Identity Visibility

With Akeyless AI Insights™, organizations gain a new level of visibility across machine, human, and AI identities. The intelligent assistant delivers natural-language insights, automated risk detection, and instant reporting—empowering security teams to identify misconfigurations and act on emerging threats in real time.

A Unified Identity Security Platform for the AI Era

These innovations expand the Akeyless Identity Security Platform into a unified control plane for humans, machines, and AI agents. Backed by Distributed Fragments Cryptography™ (DFC™) and quantum-safe encryption, the platform combines Secrets Management, PKI & Certificate Lifecycle Management, Workload Identity Federation, Multi-Vault Governance, and Privileged Access under a single, zero-knowledge architecture. The result: simplified operations, reduced risk, and a future-ready identity layer for secure AI adoption.

Akeyless will present these capabilities in its upcoming virtual series, “Identity Security – Unleashed for the AI Era”, beginning November 6, 2025. The event will feature live product demonstrations and expert discussions on securing Non-Human Identities (NHIs) and managing trust in autonomous systems.