How Government and Industry Can Tackle Emerging Cybersecurity and Governance Risks after the Mythos Moment: Highlights from Dynamo AI’s Remarks Before the US House of Representatives Committee

On April 16, 2026, Dynamo AI Co-Founder and President Dr. Christian Lau participated in a roundtable hosted by the U.S. House Committee on Homeland Security focused on AI and national security, alongside leaders from Meta, Cloudflare, and Vannevar.

The discussion centered on how artificial intelligence is rapidly changing the cybersecurity landscape, including the risks posed by Anthropic’s Mythos on critical US infrastructure. As AI capabilities advance, organizations face a dual challenge: adversaries can use AI to automate and scale cyber attacks, while enterprises are simultaneously deploying AI agents that can access data, use tools, and take action across connected business environments.

These developments are creating a new and expanding attack surface that extends beyond models alone to the broader operational environment—including identities, permissions, tools, workflows, infrastructure, and runtime behavior.

“AI security cannot stop at model safety,” said Dr. Christian Lau, Co-Founder and President of Dynamo AI. “The next phase of secure AI adoption requires controls that operate where risk actually emerges: at runtime, across systems, tools, and real-world actions.”

Dynamo AI is focused on building the control layer that agentic AI now demands. This includes capabilities designed to help organizations detect threats earlier, continuously test systems for exploitable weaknesses, and apply runtime guardrails and intervention mechanisms to control AI agent intent, across coding agents and general-purpose agents. 

Dr. Lau highlighted the importance of government and industry investment in this new layer of technical controls to prevent AI-related cyber, information security, and supply chain risks leading to sensitive data loss, operational disruption, or broader mission impact. 

Dr. Lau also emphasized the importance of stronger public-private collaboration to improve resilience across critical infrastructure and the broader economy as AI adoption accelerates.

The path forward is not to halt AI progress or impede its inevitable development. It is to pair innovation with aggressive investment in both defensive and offensive AI cyber capabilities. 

Dynamo’s technology has demonstrated that AI can materially improve cyber defense through stronger detection, faster triage, and greater analyst efficiency. In high-impact environments, leading Fortune 500 companies and public sector organizations such as the U.S. Army trust Dynamo AI to test AI agents, identify vulnerabilities, and secure them in real time.

“The path forward is not to slow innovation,” Dr. Lau underscored. “It is to pair progress with investment in the security, governance, and operational controls required to deploy AI securely.”

‍

About Dynamo AI

Dynamo AI is the trusted GenAI and Agentic AI evaluation, guardrail, and observability platform for enterprises and governments operating in critical sectors. Dynamo AI helps organizations accelerate secure AI adoption with solutions for AI testing, runtime protection, and continuous monitoring.

Learn how Dynamo AI helps agencies and critical industries evaluate, govern, and secure AI agents across the modern cyber environment here.

‍