EP39: White House AI Executive Order, The Bletchley Declaration & Adversarial AI Attacks

Join our Discord: https://discord.gg/TRrgAyeMBuy the merch: https://www.thisdayinaimerch.com/ This week the AI guys unpack the White House's sweeping executive order on regulating AI - will this lead to the death of open-source models? They also discuss the vague and fluffy Bletchley Declaration signed by world leaders, why Geoffrey Hinton just won't stop fearmongering, and introduce some hilarious new merch including a life-size shower curtain! Tune in for hot takes on the AI ethics debate, prompt engineering tricks, and key insights on the future of language models.CHAPTERS:=====00:00 - King Charles on AI (Cold Open)00:20 - Thoughts on White House AI Executive Order 23:09 - The Bletchley Declaration & AI Safety Summit38:04 -  LoRA Fine-tuning Efficiently Undoes Safety Training in Llama 2 & They Killed Tay!48:34 - Adversarial Attacks and Defenses in Large Language Models: Old and New Threats Paper51:51 - Mike proposes What The Future of AI Computing Might Look Like55:00 - Leaked: The Secret Prompt Powering ChatGPT's New Multi-Tool Mode (and How to Hack It)1:01:39 - Anthropic Have Raised More Billions & Our Merch Store!SOURCES:======https://www.whitehouse.gov/briefing-room/presidential-actions/2023/10/30/executive-order-on-the-safe-secure-and-trustworthy-development-and-use-of-artificial-intelligence/https://www.aisnakeoil.com/p/what-the-executive-order-means-forhttps://www.gov.uk/government/publications/ai-safety-summit-2023-the-bletchley-declaration/the-bletchley-declaration-by-countries-attending-the-ai-safety-summit-1-2-november-2023https://www.businessinsider.com/sam-altman-and-demis-hassabis-just-want-to-control-ai-2023-10https://twitter.com/ylecun/status/1718263147591573949?s=20https://twitter.com/ldjconfirmed/status/1718456393026490523Leaked Prompt: https://raw.githubusercontent.com/spdustin/ChatGPT-AutoExpert/main/_system-prompts/all_tools.mdhttps://www.cnbc.com/2023/10/27/google-commits-to-invest-2-billion-in-openai-competitor-anthropic.htmlPAPERS:======https://arxiv.org/pdf/2310.17688.pdfhttps://arxiv.org/pdf/2310.20624.pdfhttps://arxiv.org/pdf/2310.19737v1.pdf

Om Podcasten

Join Michael and Chris Sharkey, two proudly average tech enthusiasts, as they stumble through the world of artificial intelligence with all the grace of a robot learning to dance. This (sometimes weekly*) podcast delivers an hour-long conversation about their thoroughly middle-of-the-road adventures with AI. No PhDs. No Silicon Valley insights. Just two guys with enough technical knowledge to be dangerous, sharing their unexceptional yet entertaining experiences with AI tools and technology. Subscribe now to hear: • Mediocre hot takes on AI developments • Stories of AI experiments gone adequately okay • The most average advice you'll ever need • Two Sharkeys trying their best to sound smart about algorithms • Childish AI prank calls that somehow fool everybody • Attempts at using AI for phishing attacks on their mother • "Chart-topping" AI songs according to the brothers Join our perfectly mediocre community where being average at AI is celebrated, questions are encouraged, and learning through mistakes is our specialty. Because let's face it - most of us are figuring this out as we go along. New episodes drop whenever we remember to record them. 🎙️ Proudly supported by Simtheory.ai