WeAreDevelopers Logo WeAreDevelopers
Sign in
For You LIVE
World Congress 2025 • Aug 20, 2025 • Session details

Confuse, Obfuscate, Disrupt: Using Adversarial Techniques for Better AI and True Anonymity

David vonThenen

Altering a single pixel can break computer vision, and bad grammar can bypass NLP filters. Discover how to harden your AI pipelines against purposeful adversarial manipulation.

Pause
Mute Enter Fullscreen
#1 about 1 min

Manipulating AI models for privacy and anonymity

How manipulating machine learning inputs can defend personal privacy against automated surveillance systems.

#2 about 3 min

Explainable AI and dataset bias detection

Addressing dataset bias and noise is necessary to build ethical and transparent machine learning models.

#3 about 4 min

Measuring data inconsistencies and annotation errors

Identifying labeling errors and adversarial noise requires qualitative observation tools to validate model performance.

#4 about 2 min

Practical motivations for deploying adversarial machine learning inputs

Intentional adversarial inputs act as edge testing for algorithms and provide a safeguard against unwanted tracking.

#5 about 4 min

Natural language processing obfuscation and confusion techniques

Encoding schemes, code-switching, and metaphorical phrases disrupt text processing systems without breaking human comprehension.

#6 about 5 min

Visualizing machine learning inference decisions with Captum

Highlighting specific token attributions in text and images reveals exactly how models arrive at their classifications.

#7 about 4 min

Disrupting sentiment analysis with intentional linguistic errors

Introducing intentional grammar and spelling errors forcefully alters tokenization logic to artificially flip sentiment scores.

#8 about 4 min

Manipulating vision classifiers using single-pixel data poisoning

Placing a single manipulated pixel across training images trains recognition models to predictably misclassify subjects.

#9 about 2 min

Disrupting real-time object detection with adversarial patterns

Wearing clothing printed with specific algorithmic noise prevents live bounding box detection systems from recognizing a human.

#10 about 3 min

Defending machine learning models against adversarial attacks

Applying input normalization and graphic smoothing counters poisoned data streams before they reach the model.

#11 about 2 min

Mitigating vision classifier attacks using Gaussian blur techniques

Executing targeted blur on incoming images neutralizes microscopic adversarial artifacts to restore accurate programmatic classification.

Matching moments

17:05 min

Navigating data privacy boundaries and adversarial model reliability

Alexandra Waldherr · LIVE

3:08 min

Conducting data audits and adversarial testing on models

Toju Duke · World Congress 2022

2:01 min

Defending visual artificial intelligence models against data poisoning

Mirko Ross · World Congress 2023

4:55 min

Core principles of manipulating artificial intelligence models

Mirko Ross · World Congress 2023

1:44 min

Corrupting AI models and software supply chains

Balázs Kiss · World Congress 2023

2:02 min

Generating adversarial patterns and defending consumer machine learning models

Mirko Ross · World Congress 2023

More from World Congress 2025

GenAI Security: Navigating the Unseen Iceberg thumbnail
24:33
11 months ago

GenAI Security: Navigating the Unseen Iceberg

Maish Saidel-Keesing

Behind the Code: How Women Are Powering the Future of AI thumbnail
30:28
about 1 year ago

Behind the Code: How Women Are Powering the Future of AI

Alexandra Wudel, Laura Moritz, Madalina Florean

The AI Hype Filter: What’s Real, What’s Investable, What’s Noise? thumbnail
28:12
about 1 year ago

The AI Hype Filter: What’s Real, What’s Investable, What’s Noise?

Justus Lumpe, Konstantin Vinogradov, Nils Eiteneyer, Stephan Wirris

AI in Leadership: How Technology is Reshaping Executive Roles thumbnail
28:19
about 1 year ago

AI in Leadership: How Technology is Reshaping Executive Roles

Bernd Datler, Jasmin Kaiser, Jeff Hausmann, Sonja Alvarez

The AI-Ready Stack: Rethinking the Engineering Org of the Future thumbnail
28:09
about 1 year ago

The AI-Ready Stack: Rethinking the Engineering Org of the Future

Alex Laubscher, Jan Oberhauser, Keno Dreßel, Mirko Novakovic

AI Sovereignty: What Does It Take? thumbnail
29:13
about 1 year ago

AI Sovereignty: What Does It Take?

Boris Hecker, Clemens Wasner, Robin Hermann, Sonja Alvarez

Upcoming sessions on this topic

Open session

World Congress 2026 North America

September 25, 2026 · 16:50–17:20

Stage 5

The Things Your AI Isn't Telling You

Desmond Lamptey

Lead Software Engineer at Capital One

Desmond Lamptey
Open session

World Congress 2026 North America

September 25, 2026 · 11:00–11:30

Stage 4

Managing Abuse in the Era of AI Agents

Eli-Shaoul Khedouri

CEO of Intuition Machines

Eli-Shaoul Khedouri
Open session

World Congress 2026 North America

September 24, 2026 · 12:15–12:45

Stage 4

It's Not About the Models

Bob Wambach

Vice President, Market and Customer Insights for Dynatrace

Bob Wambach
Open session

World Congress 2026 North America

September 23, 2026 · 10:00–17:00

Stage 11

Building Stuff with GenAI - The Open Minded Workshop beyond OpenAI

Andreas Erben

CTO for Applied AI and Metaverse at daenet

Andreas Erben
Open session

World Congress 2026 North America

September 24, 2026 · 15:10–15:20

Outdoor Stage

AI That Argues With Itself: Building Self-Debating Systems That Catch Their Own Bugs

Shreya Singhal

AI Applied Scientist at Claritev

Shreya Singhal
Open session

World Congress 2026 North America

September 24, 2026 · 14:50–15:20

Stage 1

The Era of Machine-Driven Defense is Here: Headless Security

Loris Degioanni

Founder & CTO of Sysdig

Loris Degioanni

Related videos

Hacking AI - how attackers impose their will on AI thumbnail
27:02
World Congress 2023 • almost 3 years ago

Hacking AI - how attackers impose their will on AI

Mirko Ross

A hundred ways to wreck your AI - the (in)security of machine learning systems thumbnail
24:23
World Congress 2023 • almost 3 years ago

A hundred ways to wreck your AI - the (in)security of machine learning systems

Balázs Kiss

Explainable machine learning explained thumbnail
56:46
LIVE • over 3 years ago

Explainable machine learning explained

Karol Przystalski

Machine Learning: Promising, but Perilous thumbnail
27:19
World Congress 2023 • about 3 years ago

Machine Learning: Promising, but Perilous

Nura Kawa

AI & Ethics thumbnail
28:27
World Congress 2025 • about 1 year ago

AI & Ethics

PJ Hagerty

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

Related articles

From learning to earning

Jobs that call for the skills explored in this talk.