Page 11 - Detecting deepfakes and generative AI: Report on standards for AI watermarking and multimedia authenticity workshop

Detecting deepfakes and generative AI: Report on standards for AI
watermarking and multimedia authenticity workshop

3 Session 1: Setting the scene – The challenges and risks of

deepfakes and generative AI multimedia

The main objectives of this session were to introduce the risks posed by deepfakes and AI-
generated content and the policy and legal measures being introduced to address those
risks. The panel discussion explored challenges including misinformation and disinformation,
consumer protection, copyright protection, and what policy and technological measures would
be needed to address them.

Opening Remarks: Bilel Jamoussi, Deputy Director, Telecommunication Standardization
Bureau, ITU

Moderator: Alessandra Sala, Senior Director of AI and Data Science, Shutterstock

Speakers

• Sam Gregory, Director, WITNESS.
• Nick Thompson, CEO, The Atlantic.
• Robin Raskin, Founder, Virtual Events Group.
• Harry Yeff, the voice artist Reeps One.
• Helena Leurent, Director General, Consumers International.
• Tobias Bednarz, Legal Counsellor, Copyright Law Division, Copyright and Creative
Industries Sector, World Intellectual Property Organization (WIPO).

Bilel Jamoussi, Deputy Director of ITU's Telecommunication Standardization Bureau, provided
the opening remarks for the session and highlighted that governments are assigning high
priority to the need to address risks posed by generative AI and deepfakes.

The workshop discussed technical solutions and standards already available to address those
risks and gaps still to be covered, with the aim of making recommendations on how best to
coordinate relevant technical work and provide related information to governments in support
of their work on relevant policies and regulations.

Deepfake technology allows people to swap faces in videos and images, change voices, and
alter texts in documents. Deceptive videos and images created using generative AI can be used
for identity theft, the impersonation of popular figures, disinformation, and the bypassing of
identity verification methods to commit fraud.

3