WeAreDevelopers Logo WeAreDevelopers
Sign in
For You LIVE
Aug 21, 2026

How to Stop Your Agents From Going Rogue - Arnav Gupta

Arnav Gupta warns that deploying autonomous AI agents without guardrails is a disaster waiting to happen. Learn how open-source runtime security stops rogue systems from deleting your production data.

Pause
Mute Enter Fullscreen
#1 about 3 min

Moving from India to San Francisco for startup accelerators

Immersing in the local builder culture accelerates the journey of creating new software and robotics companies.

#2 about 2 min

Runtime security layers for autonomous artificial intelligence agents

Implementing a human-in-the-loop framework prevents autonomous agents from executing suspicious actions against business logic.

#3 about 2 min

Treating autonomous agents as privileged identities requiring safety guardrails

Deploying artificial intelligence with full access necessitates strict guardrails to prevent infrastructure damage and security breaches.

#4 about 4 min

Balancing rapid artificial intelligence development with strict compliance regulations

Integrating security guardrails protects critical user data and ensures compliance with frameworks like the EU AI Act.

#5 about 5 min

Transitioning from open source retrieval systems to enterprise security

Witnessing rogue agents destroy production databases highlights the critical need for context-aware safety systems in enterprise deployments.

#6 about 4 min

Utilizing open source models to establish trust with enterprises

Maintaining a transparent codebase allows enterprise clients to verify security algorithms before integrating third-party tools.

#7 about 3 min

Defending against probabilistic vulnerabilities in nascent agent ecosystems

Developing comprehensive platforms to auto-fix vulnerabilities mitigates risks associated with infinite prompting and supply chain attacks.

#8 about 2 min

Overcoming data privacy concerns to safely adopt artificial intelligence

Implementing robust security frameworks enables cautious enterprises to utilize artificial intelligence without violating regional data privacy laws.

#9 about 3 min

Collaborating on secure coding environments and open sourcing solutions

Partnering with infrastructure creators to secure virtual cloud environments demonstrates the value of community collaboration and open sourcing tools.

Matching moments

1:03 min

From read-only models to excessive agency

Alex Olivier Alex Olivier · World Congress 2026 Europe

4:15 min

Security integration and AI skepticism in developer tooling

Chris Heilmann +2 · LIVE

1:55 min

Current state of security in AI applications

Deepu Deepu · World Congress 2025

2:18 min

Managing security risks introduced by autonomous AI agents

Christian Heilmann Christian Heilmann · World Congress 2025

2:49 min

Designing an open agent platform with governed autonomy

Dennis Zielke Dennis Zielke +1 · World Congress 2025

3:45 min

Fusing developer experience and platform engineering for agentic SDLC

Julia Kordick Julia Kordick · World Congress 2026 Europe

Upcoming sessions on this topic

Open session

World Congress 2026 North America

When Agents Became Users: Rearchitecting Identity and Permissions for AI at Scale

Yoav Gal, Dor Cohen

Yoav Gal
Dor Cohen
Open session

World Congress 2026 North America

Securing AI Agent Infrastructure: Identity, Attestation, and Trust at Scale

Abdel Fane

Founder of OpenA2A

Abdel Fane
Open session

World Congress 2026 North America

I Don't Trust AI Agents (And Neither Should You): Building Production-Ready Architectures

Darko Mesaros

Distinguished Developer Advocate at AWS

Darko Mesaros
Open session

World Congress 2026 North America

Closing the Visibility Gap: Lessons from Safety Critical Agentic Systems

Vivek Pandit

Principal Engineer at Cadence

Vivek Pandit
Open session

World Congress 2026 North America

Responsible AI Architecture with Zero Trust Agents

Ashok Prakash

Staff ML Engineer at Apple

Ashok Prakash
Open session

World Congress 2026 North America

Give the Agent a Budget, Not a Token

Sachin Malhotra

MTS @Anthropic

Sachin Malhotra

Related videos

The day the chatbot asked for sudo thumbnail
27:48
World Congress 2026 Europe • about 2 months ago

The day the chatbot asked for sudo

Alex Olivier

Agentic AI: building autonomous systems for developers thumbnail
28:58
World Congress 2026 Europe • about 2 months ago

Agentic AI: building autonomous systems for developers

Anshul Jindal, Sebastian Mahr

Fighting the Next Wave of Cybercrime thumbnail
30:27
World Congress 2026 Europe • about 2 months ago

Fighting the Next Wave of Cybercrime

Michele Zuccala, Misha Bragin, Ross Kukulinski, Frank Schlesinger, Tomislav Tipurić

The New AI Security Stack: Observe, Detect, Protect thumbnail
30:16
World Congress 2026 Europe • about 2 months ago

The New AI Security Stack: Observe, Detect, Protect

Milin Desai, Christian Trummer, Jyoti Bansal, Tomislav Tipurić

Delay the AI Overlords: How OAuth and OpenFGA Can Keep Your AI Agents from Going Rogue thumbnail
23:29
World Congress 2025 • about 1 year ago

Delay the AI Overlords: How OAuth and OpenFGA Can Keep Your AI Agents from Going Rogue

Deepu

The Private AI Platform: Why Agentic Apps Need a Private Application Platform thumbnail
29:02
World Congress 2026 Europe • about 2 months ago

The Private AI Platform: Why Agentic Apps Need a Private Application Platform

Oren Penso

Related articles

From learning to earning

Jobs that call for the skills explored in this talk.