Why are current security tools inadequate for AI agents?

Existing security infrastructure like firewalls and endpoint protection systems are not agent-aware—they cannot distinguish between actions taken by humans versus autonomous agents. This creates a fundamental visibility problem as agents and human activity mesh together into an indistinguishable network of actions.

What is the autonomy slider concept for AI agents?

The autonomy slider represents the balance between granting AI agents independence to operate and implementing appropriate guardrails and controls. Similar to how parents incrementally allow children more freedom with boundaries, organizations must determine how much autonomy to grant agents while maintaining human oversight and risk management controls.

AI Agents and the Future of Enterprise Security

Name: AI Agents and the Future of Enterprise Security
Uploaded: 2026-05-11T18:05:30-04:00
Duration: 3 min 19 s
Description: AI agents are rapidly reshaping enterprises and soon we will not be able to tell whether a human or an agent is behind the action. Cyera Chief Strategy Officer, Jason Clark explores the AI agent inflection point and why today’s security and control mod...

Cyera

05/11/2026

0 (0%)

Report Like Favorite

Transcript

enterprises in a massive way is agents. Agents will outnumber everything else. It will be interconnected. It will be accessing things everywhere. We're not going to be able to discern the difference between the two, especially with the tools we use today. It's all going to look like one mesh. Think about your firewalls. Is your firewall going to be agentic aware at all? There's no way. Your endpoint security today is not agentic aware. It doesn't know whether you're doing something versus an agent's doing something. This becomes a really, really scary thing because it's all going to mesh together and agents are going to exponentially change our enterprises and our life. Very rapidly. Like I said, a lot of people are sitting in the room and are like, I don't know about that. Who's scared? Who's scared of agents? Who's scared? Who's scared of agents right now? All right. I think that just like I sat in the room and with those CIOs and those CISOs and said, they said that cloud thing's not happening. I'm seeing the same thing on agents. Not AI, but agents. Now look at this like an autonomy slider and think about if you have kids. You let them. You're like, all right. You can drive the house. Okay. You can ride your bike around the neighborhood without me watching you, but you can't leave the neighborhood, right? You put these controls in place, these guardrails, but it's an autonomy slider and it's the same thing with AI. How much do I allow you to do things at the same time trust you to do that and what guardrails I'm going to put in place? Now the beginning here is just autocomplete of code. If you talk to most developers, they'll say I'm getting somewhere between 10 and 30% benefit to programming. I'm autocompleting. I'm looking at the one that provided me. I'm getting some value. I'm moving on. Now, if I ask our marketing team, we're getting 50, 60% benefit of ChatGDP writing articles from a content creation, but they can't go to 95. If they went to 95%, it would look sloppy. It would look like AI wrote it. There'd be a lot of em dashes and emojis and it would look like AI. You're sitting here, so you have to find this balance of how do you learn to use this technology with controls. The best example to me of really 99.99% autonomy is the Waymo car. You're riding in this thing, you're trusting it with your life. Other people are trusting it with their life. The pedestrians are trusting it with their life, but every once in a while, a person sitting in an operation center has to engage and activate and manage this car. There's still a human in the loop to manage the risk, but all the guardrails. This is 15 years of maturity and development in place, but ultimately, if we can do this with cars, how can we not automate most of the things that you do every day in managing data and managing security and managing risk?

TL;DR

AI agents will soon outnumber and operate indistinguishably from humans in enterprise environments, creating a security blind spot that current tools cannot address.
Existing security infrastructure like firewalls and endpoint protection is not agent-aware and cannot differentiate between human and agent actions, creating significant risk.
The autonomy slider concept—balancing trust with guardrails—will be critical for managing AI agents, similar to how parents incrementally grant independence to children or how Waymo manages self-driving cars with human oversight.

Summary

Cyera Chief Strategy Officer Jason Clark delivers a provocative warning about the coming wave of AI agents and their implications for enterprise security. Clark argues that current security infrastructure—from firewalls to endpoint protection—is fundamentally unprepared for a world where autonomous agents operate indistinguishably from humans. Drawing parallels to early cloud skepticism, he challenges the audience's complacency about agent adoption, emphasizing that agents will exponentially transform enterprises far more rapidly than most leaders anticipate. Using the autonomy slider concept and the Waymo self-driving car as a maturity benchmark, Clark frames the central challenge: how to implement appropriate guardrails and human-in-the-loop controls while enabling the productivity gains that AI agents promise. The presentation positions data security as a critical enabler for safe agent deployment, suggesting that if autonomous vehicles can achieve 99.99% autonomy with proper controls, enterprises should be able to automate most data management and security operations with similar rigor.

Chapters

0:00 - The Agent Inflection Point
0:25 - Security Infrastructure Gaps
1:16 - The Autonomy Slider Concept
2:33 - Waymo as the Maturity Model

Key Quotes

0:10 "Agents will outnumber everything else. It will be interconnected. It will be accessing things everywhere. We're not going to be able to discern the difference between the two, especially with the tools we use today."
0:31 "Your endpoint security today is not agentic aware. It doesn't know whether you're doing something versus an agent's doing something."
1:10 "I sat in the room and with those CIOs and those CISOs and said, they said that cloud thing's not happening. I'm seeing the same thing on agents."

Categories:

Tags:

Show more Show less

Browse videos

Upcoming Webinar Calendar

11/10/2026

01:00 PM

11/10/2026

Zendesk: From AI Readiness to Results: A Framework for Transforming IT Service

https://www.truthinit.com/index.php/channel/2191/zendesk-from-ai-readiness-to-results-a-framework-for-transforming-it-service/
11/10/2026

01:00 PM

11/10/2026

Agentic Shopping Trends: Insights for the 2026 Holiday Season

https://www.truthinit.com/index.php/channel/2219/agentic-shopping-trends-insights-for-the-2026-holiday-season/
11/12/2026

01:00 PM

11/12/2026

Leaving VMware Means Leaving vSAN: Your VMware Exit Is Really a Storage Decision

https://www.truthinit.com/index.php/channel/2192/leaving-vmware-means-leaving-vsan-your-vmware-exit-is-really-a-storage-decision/
11/12/2026

01:00 PM

11/12/2026

Practical Session: Connecting On-Prem Data to Databricks via MinIO AIStor (Americas)

https://www.truthinit.com/index.php/channel/2202/practical-session-connecting-on-prem-data-to-databricks-via-minio-aistor-americas/
11/18/2026

08:00 AM

11/18/2026

Hands-On Workshop: Connecting On-Prem Data to Databricks via MinIO AIStor (International)

https://www.truthinit.com/index.php/channel/2194/hands-on-workshop-connecting-on-prem-data-to-databricks-via-minio-aistor-international/
11/19/2026

01:00 PM

11/19/2026

Securing the Microsoft Enterprise: Strengthening Configuration Management, Identity Security, and Data Protection

https://www.truthinit.com/index.php/channel/2076/securing-the-microsoft-enterprise-strengthening-configuration-management-identity-security-and-data-protection/
11/19/2026

01:00 PM

11/19/2026

Zendesk Deep Dive: Get the work done faster with Employee Service AI Agents

https://www.truthinit.com/index.php/channel/2181/zendesk-deep-dive-get-the-work-done-faster-with-employee-service-ai-agents/
11/19/2026

01:00 PM

11/19/2026

A Deep Dive into Muse, Dots, and Duet's Impact on Your Website

https://www.truthinit.com/index.php/channel/2228/a-deep-dive-into-muse-dots-and-duets-impact-on-your-website/
12/03/2026

01:00 PM

12/03/2026

CISO Insights: Steering Cybersecurity Strategies in an AI-Driven Era

https://www.truthinit.com/index.php/channel/2231/ciso-insights-steering-cybersecurity-strategies-in-an-ai-driven-era/
01/19/2027

02:00 PM

01/19/2027

HUMAN Dialogue: Embracing the Rise of the Empowered AI Consumer.

https://www.truthinit.com/index.php/channel/2160/human-dialogue-embracing-the-rise-of-the-empowered-ai-consumer/
03/18/2027

01:00 PM

03/18/2027

360View: Securing Access for People, Partners and AI Agents

https://www.truthinit.com/index.php/channel/2251/securing-access-for-people-partners-and-ai-agents/
06/17/2027

01:00 PM

06/17/2027

360View: Stop the Leak: Data Loss Prevention for the AI Era

https://www.truthinit.com/index.php/channel/955/360view-stop-the-leak-data-loss-prevention-for-the-ai-era/
08/19/2027

01:00 PM

08/19/2027

Who Has Access to What...and Should They?

https://www.truthinit.com/index.php/channel/2252/who-has-access-to-what-and-should-they/
10/21/2027

01:00 PM

10/21/2027

After VMware: Building the Private Cloud You Actually Want

https://www.truthinit.com/index.php/channel/2253/after-vmware-building-the-private-cloud-you-actually-want/
11/18/2027

01:00 PM

11/18/2027

Assume Breach, Prove Recovery

https://www.truthinit.com/index.php/channel/2254/assume-breach-prove-recovery/
12/16/2027

01:00 PM

12/16/2027

The Self-Running Service Desk

https://www.truthinit.com/index.php/channel/2255/the-self-running-service-desk/