How to protect your organization against deepfakes and disinformation

Introduction

Following on the heels of our conference on “Deepfakes, Distrust and Disinformation: The Impact of AI on Elections and Public Perception”, which was held on October 1st 2024, C4DT proposes to shift the spotlight to the strategic and operational implications of deepfakes and disinformation for organizations.

We are hosting two distinct workshops tailored for different audiences – a high-level roundtable for executives, senior managers and project managers on Tuesday, 19th of November, during which strategies to address the challenges posed by deepfakes, and collaboration opportunities and projects to counter them will be discussed. A hands-on workshop aimed at engineers, software developers and cybersecurity experts on Tuesday, 26th of November will allow the participants to develop skills and expertise in identifying and combating cyberattacks through deepfakes.

Objectives:

  1. Increased awareness and understanding of the cybersecurity threats posed by deepfakes and disinformation to organizations.
  2. Knowledge sharing and exchange of best practices in detection, prevention, and mitigation techniques.
  3. Identification of potential collaboration opportunities and projects in the field of deepfakes and disinformation.
  4. Enhanced skills and expertise among participants in identifying and combating cyberattacks through deepfakes.

DeepFake Initiatives

for Executives, Senior Managers & Project Managers

19/11/2024 – Roundtable on Deepfakes

Duration: 10:00-12:00 AM, followed by a standing lunch until 1:00 PM

Location: EPFL, room BC 010

Agenda

09h30 – Welcome Coffee & Coissants

10h00 – Welcome remarks

10h05 – 1. What is the threat to organisations? by Brian Ceccato, NCSC
1.1 Current Swiss threat landscape: What is the latest on cyberattacks (deepfakes and disinformation) on organizations in Switzerland?
1.2 Case study: Online meeting with deepfake boss: CEO fraud 2.0.
1.3 Beyond CH: Threats and risks related to deepfakes and disinformation in the EU.

10h25 – 2. What can / is being done to protect organisations?
2.1 Detection, prevention, and mitigation techniques and best practices: Exploring strategies to identify and combat cyberattacks through deepfakes, by Prof. Touradj Ebrahimi, EPFL
2.2 Prof. Ebrahimi’s lessons learned and pitfalls from 10+ years of experience in the field, by Prof. Touradj Ebrahimi, EPFL
2.3 Case study: Swiss project in banking/finance, by Prof. Touradj Ebrahimi, EPFL
2.4 Preparing employees against phishing attacks via simulations involving deepfakes – awareness / training of employees by Andrey Suzdaltsev, Brightside SA

11h35 – 3. Forward-looking and general discussion
3.1 Discussing the evolution of GenAI security threats and potential future challenges which organizations might face.
3.2 How & in what timeframe will GenAI security threats most likely evolve?
3.3 Addressing any additional areas of interest or concern raised by the participants.

12h00 – Standing lunch

Format
Talks & roundtable discussions

13h00 – End

for Engineers, Software Developers, Cybersecurity Experts

26/11/2024 – Deepfakes Hands-On Workshop

Program

  • 10:00 AM – Introductory talk by Dr. Evgeniy Upenik
  • 11:00 AM – Q&A session
  • 11:30 AM – Lunch break
  • 1:00 PM – Hands-on exercise session
    • Familiarization with the tools
    • Deepfake creation
    • Deepfake detection
  • 4:30 PM – Wrap-up
  • 5:00 PM – End

Location: EPFL, room BC 410

Morning session: Theory Update

Format: Talk, and open discussion.

Discussion points

  • Explain basics of GANs and detections
  • Terminology: training, inference, specific parameters used in the workshop
  • Protection: during generation (watermarks) or after generation (detectors)
  • Open Discussion:
    • What kind of attacks are possible with this technology?
    • How to defend / respond against these attacks?

Afternoon Session – Hands-on Workshop

Format: Exercise session

Discussion points

  • Open Source technologies used to create fake speech and fake images
  • How these tools can be abused: taking an image and changing it. Cloning voice.
  • Overview of the defenses available: watermarks (and metadata) and detectors
    • Pitfalls: Discussing common challenges and limitations in detecting and preventing cyberattacks through deepfakes.
  • Other relevant topics: Addressing any additional areas of interest or concern raised by the participants.


You may register for both sessions if you wish! Please note that the number of participants will be limited to 20 per workshop.

For more information, please contact either David Viollier or Linus Gasser.