WeAreDevelopers Logo WeAreDevelopers
Sign in
For You LIVE
World Congress 2023 • Aug 11, 2023

Machine Learning: Promising, but Perilous

Nura Kawa

A simple sticker on a stop sign can completely blind your AI perception model. Uncover the hidden security perils of transfer learning and how to fortify your ML pipelines.

Pause
Mute Enter Fullscreen
#1 about 2 min

Introduction to machine learning security and robustness

The very traits making machine learning accessible and powerful inherently create severe security vulnerabilities.

#2 about 2 min

Driving productivity with complex machine learning models

Massive parameter models enable semantic and instance context recognition to dramatically increase developer productivity.

#3 about 2 min

Overcoming data limitation barriers in automated applications

Organizations lacking resources or sufficient annotated data leverage accessible frameworks and machine learning as a service.

#4 about 3 min

Leveraging transfer learning to train efficient student models

Using feature extraction layers from massive teacher models allows rapid task learning without huge computational burdens.

#5 about 2 min

Uncovering the inherent security risks of transfer learning

Relying on generic teacher architectures allows adversaries to leverage known vulnerabilities against derivative models.

#6 about 2 min

Exploring inference, extraction, and the adversarial threat landscape

Security defenses continually adapt against foundational attack vectors designed to infer private dataset occurrences or steal proprietary algorithms.

#7 about 2 min

Evading detection through targeted adversarial example perturbations

Minimal input optimizations force perception systems to misclassify critical targets like road signs or audio profiles.

#8 about 3 min

Compromising datasets through backdoor and data poisoning attacks

Injecting rare trigger patterns into minimal training instances grants unauthorized system access without affecting regular functionality.

#9 about 3 min

Integrating threat modeling into machine learning security operations

Prioritizing use-case risks and environment impacts prior to deployment establishes standardized machine learning security workflows.

#10 about 2 min

Evaluating transparency across external teacher model resources

Validating outsourced training processes and withholding architecture details minimizes supply chain risks for deployed models.

#11 about 2 min

Developing fail-safes and defenses through adversarial training

Minimizing prediction error across adversarially perturbed datasets inoculates critical applications against straightforward input manipulation.

#12 about 2 min

Stress testing and fortifying dependent student machine models

Fine-tuning foundational layers and tracking deployed model behaviors helps identify domain shifts or emerging adversarial actions.

#13 about 2 min

Security frameworks and resources for developing robust models

Standardizing defenses with established cybersecurity guidelines and robust testing toolboxes operationalizes enterprise machine learning safety.

Matching moments

1:20 min

Utilizing industry threat models for AI security

Balázs Kiss · World Congress 2023

2:29 min

Establishing core machine learning fundamentals and robust guardrails

Ekaterina Sirazitdinova · LIVE

2:02 min

Generating adversarial patterns and defending consumer machine learning models

Mirko Ross · World Congress 2023

1:31 min

Dual usage of machine learning in cybersecurity

Wolfgang Ettlinger +1 · World Congress 2023

2:31 min

Understanding data poisoning and model bias risks

Keno Dreßel Keno Dreßel · World Congress 2025

3:16 min

Unique privacy vulnerabilities in language model architectures

Aditi Godbole · LIVE

More from World Congress 2023

The year 3000, a brief history of Web Development thumbnail
25:30
over 2 years ago

The year 3000, a brief history of Web Development

Lorenzo Pieri

Demografie nutzen um jedem Bewerber das passende Angebot zu machen! thumbnail
8:52
over 2 years ago

Demografie nutzen um jedem Bewerber das passende Angebot zu machen!

Nathalie Gierth

Guided Memory Management: Rust's Ownership Model thumbnail
0:18
almost 3 years ago

Guided Memory Management: Rust's Ownership Model

Stefan Baumgartner

Passwordless future: WebAuthn and Passkeys in practice thumbnail
32:32
almost 3 years ago

Passwordless future: WebAuthn and Passkeys in practice

Clemens Hübner

External Secrets Operator: the secrets management toolbox for self-sufficient teams thumbnail
30:07
almost 3 years ago

External Secrets Operator: the secrets management toolbox for self-sufficient teams

Moritz Johner

Building Reliable Serverless Applications with AWS CDK and Testing thumbnail
25:18
almost 3 years ago

Building Reliable Serverless Applications with AWS CDK and Testing

Raphael Manke

Upcoming sessions on this topic

Open session

World Congress 2026 North America

September 25, 2026 · 09:00–09:30

Stage 6

Red Teaming Your LLM App -- A Hands-On Threat Model You Can Reuse

Saloni Garg

Senior ML Engineer at Adobe

Saloni Garg
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 @ Capital One

Desmond Lamptey
Open session

World Congress 2026 North America

September 25, 2026 · 11:40–12:10

Stage 9

You Can’t Re-Run Sunlight: Designing ML Data Architectures for Physical AI

An Phan

Senior Data Infrastructure Engineer @ Hippo Harvest

An Phan
Open session

World Congress 2026 North America

September 24, 2026 · 17:30–18:00

Stage 6

No Single Model to Rule Them All: Building Resilient AI Agents Across Open & Closed LLMs

Emmanuel Acheampong

Senior Manager Developer Relations at Crusoe AI

Emmanuel Acheampong
Open session

World Congress 2026 North America

September 25, 2026 · 16:10–16:40

Stage 5

Developer Liability in the AI Agent Era: Building Responsibly

Alla Barbalat

Freelancer Trade Show Spokesmodel and Tech Event Host

Alla Barbalat
Open session

World Congress 2026 North America

September 24, 2026 · 16:10–16:40

Stage 7

Your Threat Model Is Lying to You: Why Modeling the Design Isn’t Enough in 2026

Farshad Abasi

CEO/Founder, Eureka DevSecOps + Forward Security

Farshad Abasi

Related videos

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

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

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

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

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

Confuse, Obfuscate, Disrupt: Using Adversarial Techniques for Better AI and True Anonymity thumbnail
27:34
World Congress 2025 • about 1 year ago

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

David vonThenen

Related articles

From learning to earning

Jobs that call for the skills explored in this talk.