HoneyPod - The Blue Side of Cybersecurity

Are LLMs Immune to Obliviate Spells?

8 min · 16. okt. 2024
episode Are LLMs Immune to Obliviate Spells? cover

Beskrivelse

Welcome to another episode of HoneyPod! Today, we're having a look into the world of Artificial Intelligence, and we're exploring how it interacts with the beloved universe of Harry Potter. Yes, you heard that right — magic and machines, working together, or perhaps, at odds with each other. Episode links: * The Boy Who Survived: Removing Harry Potter from an LLM is harder than reported [https://arxiv.org/abs/2403.12082v1] (Arxiv) * Who's Harry Potter? Approximate Unlearning in LLMs [https://arxiv.org/abs/2310.02238] (Arxiv) * ConfAIde: Can LLMs Keep a Secret? Testing Privacy Implications of Language Models [https://confaide.github.io/] (Website) SUPPORT THE PODCAST: Share this episode with your friends and colleagues if you liked it, and help us spread the buzz of cybersecurity intrigue! And.. If you really like the show, you can buy me a beer [https://buymeacoffee.com/tsumarios] 😊🍻 ABOUT THE HOST: Website: tsumarios.github.io [https://tsumarios.github.io/] Blog: tsumarios.github.io/blog [https://tsumarios.github.io/blog] GitHub: @tsumarios [https://github.com/tsumarios] Twitter: @tsumarios [https://twitter.com/tsumarios] Thank you for listening!

Kommentarer

0

Vær den første til å kommentere

Registrer deg nå og bli medlem av HoneyPod - The Blue Side of Cybersecurity sitt community!

Prøv gratis

Prøv gratis i 14 dager

99 kr / Måned etter prøveperioden. · Avslutt når som helst.

  • Eksklusive podkaster
  • 20 timer lydbøker i måneden
  • Gratis podkaster

Alle episoder

2 Episoder

episode Are LLMs Immune to Obliviate Spells? cover

Are LLMs Immune to Obliviate Spells?

Welcome to another episode of HoneyPod! Today, we're having a look into the world of Artificial Intelligence, and we're exploring how it interacts with the beloved universe of Harry Potter. Yes, you heard that right — magic and machines, working together, or perhaps, at odds with each other. Episode links: * The Boy Who Survived: Removing Harry Potter from an LLM is harder than reported [https://arxiv.org/abs/2403.12082v1] (Arxiv) * Who's Harry Potter? Approximate Unlearning in LLMs [https://arxiv.org/abs/2310.02238] (Arxiv) * ConfAIde: Can LLMs Keep a Secret? Testing Privacy Implications of Language Models [https://confaide.github.io/] (Website) SUPPORT THE PODCAST: Share this episode with your friends and colleagues if you liked it, and help us spread the buzz of cybersecurity intrigue! And.. If you really like the show, you can buy me a beer [https://buymeacoffee.com/tsumarios] 😊🍻 ABOUT THE HOST: Website: tsumarios.github.io [https://tsumarios.github.io/] Blog: tsumarios.github.io/blog [https://tsumarios.github.io/blog] GitHub: @tsumarios [https://github.com/tsumarios] Twitter: @tsumarios [https://twitter.com/tsumarios] Thank you for listening!

16. okt. 20248 min
episode Behind the (Digital) Crime Scenes cover

Behind the (Digital) Crime Scenes

Welcome to the first episode of HoneyPod! Today, we’re diving into the captivating world of digital forensics – think Sherlock Holmes meets the Matrix! We’re going beyond the tools and tech, having a look at how digital forensics fits into criminal investigations, from the first hint of suspicion all the way to the courtroom showdown. Episode links: * Behind the (Digital Crime) Scenes: An MSC Model [https://doi.org/10.1109/ISDFS60797.2024.10527316] (also here in Arxiv preprint [https://arxiv.org/abs/2403.16196]) * Digital Forensics [https://en.wikipedia.org/wiki/Digital_forensics] (Wikipedia) * Computer Forensics [ https://en.wikipedia.org/wiki/Anti%E2%80%93computer_forensics] (Wikipedia) * Message Sequence Chart [https://en.wikipedia.org/wiki/Message_sequence_chart] (Wikipedia) SUPPORT THE PODCAST: Share this episode with your friends and colleagues if you liked it, and help us spread the buzz of cybersecurity intrigue! And.. If you really like the show, you can ⁠buy me a beer⁠ [https://buymeacoffee.com/tsumarios] 😊🍻 ABOUT THE HOST: Website: ⁠tsumarios.github.io⁠ [https://tsumarios.github.io/] Blog: ⁠tsumarios.github.io/blog⁠ [https://tsumarios.github.io/blog] GitHub: ⁠@tsumarios⁠ [https://github.com/tsumarios] Twitter: ⁠@tsumarios⁠ [https://twitter.com/tsumarios] Thank you for listening!

29. sep. 20248 min