View source for Talk:Identification of AI generated content – TOPIC 03 Sub 03 2024

== List of ideas of points to bring up ==

=== What is the purpose of this list? ===
The following list contains some ''possible ideas for points to cover'' during the session. This list is not exhaustive, and will change depending on what points are brought up in previous sessions.

This is due to the fact that we intend to avoid hosting a session that is “more of the same” with respect to the other sessions listed in the published programme.

=== List of possible points to discuss ===

*What are the main challenges from a technical perspective in detecting AI-generated content?
*What types of measures can help protect the authenticity and reliability of information in the context of AI-generated content?
*How can we ensure that the detection of AI-generated content is done in a way that respects privacy and data protection principles?
*What role can regulators play in addressing the challenges of AI-generated content detection?
*How can we foster collaboration between different stakeholders to address the challenges of AI-generated content detection?
*What should be the role of European institutions in promoting the development of tools and technologies for detecting AI-generated content?
*How can we ensure that the detection of AI-generated content is done in a way that respects freedom of expression and information?
*How can academia contribute to the development of tools and technologies for detecting AI-generated content?
*What should be the role of the private sector in addressing the challenges of AI-generated content detection?
*How can we help the general public navigate the challenges of AI-generated content and make informed decisions about the information they consume?
*What are the ethical considerations that need to be taken into account when detecting AI-generated content?
*How can we ensure that the detection of AI-generated content is done in a way that is transparent and accountable?
*How is the new EU AI Act expected to impact the detection of AI-generated content?
*Is it possible to develop a universal framework for detecting AI-generated content, or should the approach be tailored to specific contexts and use cases?

<!-- == Draft Session Details ==
Feel free to contribute by writing to the mailing list or by editing this page!
If you don't currently have edit access to the wiki, you can email wiki@eurodig.org asking for access and they will give it to you very quickly :)

=== Draft Title ===
Detecting AI-Generated Content: Challenges, Opportunities and the Way Forward

=== Draft Description ===
The advent of AI has introduced a new era of content generation, with AI-generated content being employed in a multitude of applications, including text, images, audio, and video. This is utilized in a diverse array of fields, including journalism, entertainment, and education. While AI-generated content can be advantageous in numerous ways, it also presents challenges, such as the potential for deep fakes and disinformation. Therefore, it is critical to be able to identify AI-generated content in order to guarantee the authenticity and reliability of information.
 
Nevertheless, the principal alternatives proposed to date for identifying AI-generated content seem to be insufficient:
* The detection of AI-generated texts is unreliable, as it can be easily circumvented by means of paraphrasing. 
* Applying cryptographic signatures to content to verify its authenticity would be cumbersome, and it would not guarantee that the content is not AI-generated. For example, if a person takes a photograph of a deep fake, the cryptographic signature would still verify the content as authentic. 
* Watermarking the content with a notice that it was generated by AI can be circumvented, even if the watermark is not directly visible.
In essence, the identification of AI-generated content is a complex problem that demands a multi-faceted approach. This session will examine the challenges and opportunities in detecting AI-generated content and discuss potential ways to protect the authenticity and reliability of information.

=== Draft Format ===
When it comes to the format of the session, I've been looking into last year's EuroDIG sessions, and I've noticed that there were a few different formats used. My favourite one, and one that I think would work well for our session, could be something like this (extracted from https://eurodigwiki.org/wiki/Virtual_worlds,_but_real_risks:_navigating_metaverses_as_a_next_generation_of_digital_platforms_%E2%80%93_TOPIC_03_Sub_01_2023 - it would need to be adapted to our topic):

To ensure an interactive format of the session, at the beginning, a poll will be used to ask online and onsite participants to share their opinion on what threats could be observed in metaverses. Then, based on their answers, key participants will share 3-minute insights on the topic from their fields of expertise. This would lead to a further, more informed debate between the audience and key participants, including an exchange of comments and Q&A.

I also think it could be beneficial to have a demonstration of AI-generated content detection tools during the session. This could help to illustrate the challenges and opportunities in identifying AI-generated content. What do you think?

=== Participants and Moderator ===
We will need to decide on the participants and moderator for the session. I would be happy to hear your ideas on these aspects as well. Any suggestions are more than welcome!

-->

== Current discussion, conference calls, schedules and minutes ==

=== Second call of the Org Team ===

<!-- ==== Joining data ====
<pre>
EuroDIG Org Team Meeting - T3S3
Tuesday, May 7 · 5:00 – 6:00pm
Time zone: Europe/Madrid
Google Meet joining info
Video call link: https://meet.google.com/dyo-qedd-rsb
Or dial: ‪(IE) +353 1 571 1000‬ PIN: ‪651 776 748 5203‬#
More phone numbers: https://tel.meet/dyo-qedd-rsb?pin=6517767485203
</pre>
-->

==== What to expect from the call? ====
The aim of the call is to talk about aspects of the session that have yet to be well-defined. Also, I will be happy to answer any questions you might have about the session.

==== Agenda ====
# Introduction of Org Team members
# Brainstorming on what to include in the session, trying to be as specific as possible
# Status of contacting potential speaker
# Potential moderator of the session
# Next steps - deciding on the next call

=== First call of the Org Team ===

<!-- ==== Joining data ====
<pre>
First call - Org Team - EuroDIG 2024 - Topic3 Sub3
Thursday, April 18 · 4:00 – 5:00pm
Time zone: Europe/Madrid
Google Meet joining info
Video call link: https://meet.google.com/onm-touc-hkz
Or dial: ‪(ES) +34 910 48 95 10‬ PIN: ‪175 655 089 7327‬#
More phone numbers: https://tel.meet/onm-touc-hkz?pin=1756550897327
</pre>
-->

==== What to expect from the call? ====
The aim of the call is to introduce ourselves and come up with a draft for the following aspects of the session:
* Title
* Description
* Format
* Participants
* Moderator

==== Agenda ====
# Introduction of Org Team members
# Initial discussion on what the main thematic focus of the session should be
# Brainstorming on the more detailed theme issues (what are the different areas of AI generated content that can be discussed - e.g. technical feasibility, deepfakes, cyberbullying, cybersecurity, regulation, mental health, law enforcement, content moderation, etc. - the list is not exhaustive)
# Potential key participants of the session
# Potential moderator of the session
# Title and description of the session (draft)
# Next steps - deciding on the next call