Deepfake detectors can be defeated, computer scientists show for the first time

#1
C C Offline
https://www.newswise.com/articles/deepfa...first-time

INTRO: Systems designed to detect deepfakes --videos that manipulate real-life footage via artificial intelligence--can be deceived, computer scientists showed for the first time at the WACV 2021 conference which took place online Jan. 5 to 9, 2021.

Researchers showed detectors can be defeated by inserting inputs called adversarial examples into every video frame. The adversarial examples are slightly manipulated inputs which cause artificial intelligence systems such as machine learning models to make a mistake. In addition, the team showed that the attack still works after videos are compressed.

“Our work shows that attacks on deepfake detectors could be a real-world threat,” said Shehzeen Hussain, a UC San Diego computer engineering Ph.D. student and first co-author on the WACV paper. “More alarmingly, we demonstrate that it’s possible to craft robust adversarial deepfakes in even when an adversary may not be aware of the inner workings of the machine learning model used by the detector.”

In deepfakes, a subject’s face is modified in order to create convincingly realistic footage of events that never actually happened. As a result, typical deepfake detectors focus on the face in videos: first tracking it and then passing on the cropped face data to a neural network that determines whether it is real or fake. For example, eye blinking is not reproduced well in deepfakes, so detectors focus on eye movements as one way to make that determination. State-of-the-art Deepfake detectors rely on machine learning models for identifying fake videos.

The extensive spread of fake videos through social media platforms has raised significant concerns worldwide, particularly hampering the credibility of digital media, the researchers point out. “"If the attackers have some knowledge of the detection system, they can design inputs to target the blind spots of the detector and bypass it,” ” said Paarth Neekhara, the paper’s other first coauthor and a UC San Diego computer science student... (MORE)
Reply


Possibly Related Threads…
Thread Author Replies Views Last Post
  Call for tech providers to take action on AI non-consensual 'deepfake pornography' C C 0 854 Sep 30, 2026 11:40 PM
Last Post: C C
  Article Iran-linked hackers blamed for taking down U.K. power plant for first time C C 0 1,321 Aug 24, 2026 07:34 PM
Last Post: C C
  Research World’s first non-silicon 2D computer developed C C 0 4,381 Jun 12, 2025 06:37 PM
Last Post: C C
  Article Melbourne start-up launches 'biological computer' made of human brain cells C C 0 4,698 Mar 5, 2025 07:02 PM
Last Post: C C
  Research Scientists make first mechanical qubit (steampunk quantum computers) C C 0 4,371 Nov 15, 2024 08:08 PM
Last Post: C C
  Research Computer scientists invent an efficient new way to count C C 0 4,447 May 17, 2024 03:35 PM
Last Post: C C
  Article Should we stop AI hallucinations? + Why overtrain AI? + AI utterly defeated CAPTCHA C C 0 4,355 Apr 15, 2024 11:53 PM
Last Post: C C
  Research Method identified to double computer processing speeds C C 0 4,273 Feb 23, 2024 05:40 PM
Last Post: C C
  Article AI outperforms conventional weather forecasting for the first time: Google study C C 0 4,307 Nov 16, 2023 01:23 AM
Last Post: C C
  Software engineering or computer science? Ostronomos 10 6,416 Mar 8, 2023 12:51 AM
Last Post: Syne



Users browsing this thread: 1 Guest(s)