EP35: AI Safety Gone Mad, Stable 3B Cheese Test, GPT4 Vision & DALL-E 3 Diversity + Sydney is BACK!

Not too late to join Discord community, do it here: https://forms.gle/rSx9dYoqc1qxX6sx5. Invites going out today!Thanks for helping us reach 2K subs here on YouTube!This week we dive into the wild world of AI image generation and vision, from racist cartoon captions to heartfelt poetry written by Bing. We discuss the implications of teaching AI to forget unwanted knowledge, and debate whether safety controls are protecting users or limiting creativity. Get ready for philosophical ponderings, hilarious experiments, and our signature irreverent takes as we explore the latest AI advances and absurdities. Whether you're an expert or just fascinated by the future, this episode will challenge your thinking and give you plenty to discuss with friends.CHAPTERS======00:00 - Fooling Bing Vision to Solve Captcha00:26 - Meta's Messenger AI Stickers Out of Control! AI Safety Discussion06:17 - More Safety Nonsense: The Low-Resource Language Jailbreak GPT-4 Paper9:36 - More on Mistral 7B (Safety and Positive Reception)17:31 - Friends and Foes of Open Source AI & Is Anthropic a Crypto-like Scam for Billions? 21:26 - Turnitin Thinks It Can Detect AI, Being a Student in an AI World24:25 - Stable 3B LLM Review and Cheese Test Results38:48 - DALL-E 3 Road Test on ChatGPT & Diversity Prompt Injection Problems48:12 - Using Bing GPT4-Vision to Solve Captchas for Grandma51:01 - The Dawn of LLMs, Explorations with GPT-4Vision Paper + Possibilities of AI Vision1:04:00 - Who's Harry Potter? Making LLMs forget 1:09:00 - Google Assistant with Bard AI1:10:18 - LLaMA Long 32K Initial Thoughts1:12:40 - Sydney Bing is Back BABY!1:15:36 - Comments on Discord Rollout and Survey ResponseSOURCES======https://twitter.com/ibogost/status/1709629850359628211https://twitter.com/paul_rottger/status/1707430998600831424?s=46https://www.theinformation.com/articles/openai-rival-anthropic-in-talks-to-raise-2-billion-from-google-others-as-ai-arms-race-accelerateshttps://twitter.com/abacaj/status/1709455939231772962?s=46https://twitter.com/ylecun/status/1708149902784799121?s=46https://twitter.com/rustykitty_/status/1709316764868153537https://stability.ai/blog/stable-lm-3b-sustainable-high-performance-language-models-smart-deviceshttps://twitter.com/neilkli/status/1709450248186167715/photo/4https://twitter.com/ItakGol/status/1708541450722414798/photo/2https://www.oneusefulthing.org/p/the-shape-of-the-shadow-of-the-thinghttps://www.microsoft.com/en-us/research/project/physics-of-agi/articles/whos-harry-potter-making-llms-forget-2/https://techcrunch.com/2023/10/04/google-assistant-is-getting-ai-capabilities-with-bard/https://venturebeat.com/ai/meta-quietly-releases-llama-2-long-ai-that-outperforms-gpt-3-5-and-claude-2-on-some-tasks/https://twitter.com/lumpenspace/status/1709773644203708527/photo/2PAPERS======https://arxiv.org/pdf/2310.02446.pdfhttps://arxiv.org/pdf/2309.17421.pdf