IGF 2023 Town Hall #13 Countering Terrorist Use of Generative AI

Terrorists and violent extremists often experiment and adapt their techniques to make use of emerging technologies. Technological innovation has enabled terrorist and violent extremists to exploit new tools - such as end-to-end encryption, the decentralised web, and cryptocurrencies - to further their agenda. Much like these new technologies, generative artificial intelligence (AI) is at risk of being exploited by terrorist and violent extremists to accelerate their agenda.

Generative AI’s potential to revolutionise aspects of our society, its increasing integration into the broader online landscape, and recent rapid developments have sparked a significant increase in public attention around the opportunities and risks associated with AI.

Given its transformative potential, it is imperative that tech platforms begin to design policies and processes to prevent terrorist exploitation of generative AI, whilst ensuring the safeguarding of human rights and freedom of expression online. Governments also have a responsibility to regulate and set the parameters for legal use of generative AI, considering how terrorists and violent extremists might abuse emerging technologies.

At Tech Against Terrorism, we specialise in countering terrorist use of the internet. This necessitates staying up to date with emerging technologies like AI, how these can be exploited by terrorist and violent extremist actors, and how they can be utilised to counter terrorist use of the internet.

For this edition of Internet Governance Forum, Tech Against Terrorism will discuss both the risks and opportunities posed by generative AI for countering terrorist use of the internet, and how both should be considered equally to positively harvest the potential of this technology.

First, we will provide an overview of the threat posed by terrorist use of generative AI, including its potential to enable and accelerate propaganda generation, its potential to radicalise, and its ability to upskill terrorist and violent extremist actors by lowering technical barriers.

We will outline mitigation strategies for platforms to disrupt terrorist use of AI, including by highlighting content moderation avoidance strategies that could be used to bypass action taken by generative AI companies’ policies. We will also discuss how governments plan to regulate AI, drawing on the expertise from our Online Regulation Series, and considering AI as the next frontier of online regulation.

Lastly, we will discuss the opportunities AI provides for improving responses to terrorist use of the internet, including for content moderation, content detection, creating classifiers, supporting with behavioural analysis, and providing more rapid responses in crisis situations.

For this session, Tech Against Terrorism will draw on our expertise of terrorist use of the internet, as can be seen in our State of Play report, as well as our expertise of terrorist exploitation of emerging technologies, for instance our work on End-to-End Encryption. We will also utilise our extensive experience of mentoring tech companies from varying sectors on how to improve their policies and processes to disrupt terrorist use of their services.

To facilitate interaction between onsite and online speakers and attendees, we will include a Q&A session at the end of the presentation, to enable discussion and questions from the attendees. We will use a powerpoint presentation to ensure the best possible experience for online and onsite participants.