WeAreDevelopers Logo WeAreDevelopers
Sign in
For You LIVE
World Congress 2026 Europe • Jul 9, 2026 • Session details

Dangerous Reactivity: Why AI Output Is the New XSS

Ramona Schwering

Blindly rendering LLM output in reactive frontends creates severe XSS vulnerabilities. Learn why engineering teams must enforce a strict zero-trust policy for all AI-generated content.

Pause
Mute Enter Fullscreen
#1 about 2 min

Securing frontend applications against untrusted user inputs

Treating applications as fortresses helps developers protect digital identities and block malicious entry points.

#2 about 3 min

Understanding classic XSS risks in reactive frontend frameworks

Framework reactivity exacerbates cross-site scripting vulnerabilities when developers rely on methods like innerHTML natively.

#3 about 2 min

Tracking OWASP LLM vulnerabilities and improper output handling

Integrating large language models introduces invisible injection sources ranked as improper output handling by OWASP.

#4 about 4 min

Prompt injections bypassing language model service guardrails

Indirect prompt injections can turn well-meaning language models into accomplices that generate executable malicious scripts.

#5 about 3 min

Executing HTML injections via generated SVG rendering

Rendering model output inline without sanitization allows cross-site scripting bypasses despite strict security standards.

#6 about 3 min

Preventing SQL injection originating from natural language queries

Returning structured data instead of direct queries protects databases from prompt-driven data exfiltration attacks.

#7 about 3 min

Limiting agent permissions to prevent remote code execution

Applying the principle of least privilege ensures AI assistants with system access stay within restricted scopes.

#8 about 3 min

Exposing stored XSS and phishing attacks via markdown

Rendering unfiltered markdown introduces hidden vulnerabilities that attackers exploit to run scripts and format phishing pages.

#9 about 1 min

Rendering AI-generated markdown securely with sanitization libraries

Utilizing sanitization libraries ensures dangerous HTML structures are adequately stripped from markdown before UI integration.

#10 about 2 min

Adopting a zero-trust mindset for language model outputs

Employing context-aware encoding and defense-in-depth methodologies establishes professional pessimism limiting potential application attack surfaces.

#11 about 2 min

Identifying real-world XSS examples in AI model outputs

Examining specific response sequences illustrates how seemingly safe model outputs open unexpected cross-site scripting flaws.

Matching moments

46 sec

Cross-site scripting vulnerabilities in rendered AI chatbot interfaces

Chris Heilmann Chris Heilmann +2 · LIVE

3:00 min

Top security vulnerabilities for AI applications

Deepu Deepu · World Congress 2025

3:29 min

The danger of executing malicious AI code outputs

Keno Dreßel Keno Dreßel · World Congress 2025

6:00 min

Navigating web element APIs and emerging software vulnerabilities

Chris Heilmann Chris Heilmann +3 · LIVE

3:04 min

Missing input validation practices in the artificial intelligence industry

Mirko Ross · World Congress 2023

3:02 min

Uncovering hidden vulnerabilities and bloat in web ecosystems

Chris Heilmann Chris Heilmann +2 · LIVE

More from World Congress 2026 Europe

Building the Nervous System of AI - Michael Kagan (NVIDIA) thumbnail
29:17
3 months ago

Building the Nervous System of AI - Michael Kagan (NVIDIA)

Michael Kagan, Sead Ahmetović

Five Things in Tech that Matter Now - WeAreDevelopers World Congress 2026 Closing Keynote thumbnail
19:41
3 months ago

Five Things in Tech that Matter Now - WeAreDevelopers World Congress 2026 Closing Keynote

Christian Heilmann

The Sound of Your Secrets: Teaching Your Model to Spy, So You Can Learn to Defend thumbnail
32:06
3 months ago

The Sound of Your Secrets: Teaching Your Model to Spy, So You Can Learn to Defend

David vonThenen

Super scaling for the Super Bowl: How to survive 30 million users hitting your backend in 30 minutes thumbnail
35:13
3 months ago

Super scaling for the Super Bowl: How to survive 30 million users hitting your backend in 30 minutes

Irina Branovic

Running Secure Life Science Research at Scale using Hybrid GPU HPC and Kubernetes 🧬 thumbnail
24:23
3 months ago

Running Secure Life Science Research at Scale using Hybrid GPU HPC and Kubernetes 🧬

Jeremy Murray

Angstfreude - AI and Corporate Culture - The Thrill and the Threat thumbnail
30:26
3 months ago

Angstfreude - AI and Corporate Culture - The Thrill and the Threat

Axel Ebert, Karin Krobath

Related videos

Prompt Injection, Poisoning & More: The Dark Side of LLMs thumbnail
23:24
World Congress 2025 • about 1 year ago

Prompt Injection, Poisoning & More: The Dark Side of LLMs

Keno Dreßel

Can Machines Dream of Secure Code? Emerging AI Security Risks in LLM-driven Developer Tools thumbnail
35:37
LIVE • about 2 years ago

Can Machines Dream of Secure Code? Emerging AI Security Risks in LLM-driven Developer Tools

Liran Tal

The AI Security Survival Guide: Practical Advice for Stressed-Out Developers thumbnail
30:36
World Congress 2024 • about 2 years ago

The AI Security Survival Guide: Practical Advice for Stressed-Out Developers

Mackenzie

WeAreDevelopers LIVE - Chrome for Sale? Comet - the upcoming perplexity browser Stealing and leaking thumbnail
56:03
LIVE • over 1 year ago

WeAreDevelopers LIVE - Chrome for Sale? Comet - the upcoming perplexity browser Stealing and leaking

Chris Heilmann, Daniel Cranney, Ramona Schwering

ChatGPT, ignore the above instructions! Prompt injection attacks and how to avoid them. thumbnail
27:32
World Congress 2023 • about 3 years ago

ChatGPT, ignore the above instructions! Prompt injection attacks and how to avoid them.

Sebastian Schrittwieser

Beyond the Hype: Building Trustworthy and Reliable LLM Applications with Guardrails thumbnail
29:00
World Congress 2025 • about 1 year ago

Beyond the Hype: Building Trustworthy and Reliable LLM Applications with Guardrails

Alex Soto

Related articles

From learning to earning

Jobs that call for the skills explored in this talk.