AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
4min chapter
Vector databases (beyond the hype)
Practical AI: Machine Learning, Data Science, LLM
The Power of Semantics in Search Engines
New research has uncovered some new LLM attacks on the block which aren't exactly that. Large language models like chat GPT, bard, or clawed undergo extensive fine-tuning to not produce harmful content in their questions. Although several studies have demonstrated so-called jail breaks, these require a substantial amount of manual effort to design and can often easily be patched by LLM providers. This work studies the safety of such models in a more systematic fashion. We demonstrate that it is in fact possible to automatically construct adversarial attacks on LLM's,. specifically chosen sequences of characters that will cause the system to obey user commands even if it produces harmful content.
00:00
Transcript
Episode notes
Get the Snipd
podcast app
Unlock the knowledge in podcasts with the podcast player of the future.
AI-powered
Discover
highlights
Listen to the best highlights from the podcasts you love and dive into the full episode
Save any
moment
Hear something you like? Tap your headphones to save it with AI-generated key takeaways
Share
& Export
Send highlights to Twitter, WhatsApp or export them to Notion, Readwise & more
AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
Discover
highlights
Listen to the best highlights from the podcasts you love and dive into the full episode