This week we cover the Center for AI Safety's Statement and ask, "Will AI Really Make Humans Extinct?" We cover the discussion on GPT-4 Deteriorating: is it really getting worse? And discuss some likely causes. We also use the Internet Archive to discuss OpenAI's GPT-4 roadmap including a stateful API, giant context sizes, plugins having no product market fit and why OpenAI is constrained by GPUs.We also cover the lawyer who got caught using ChatGPT and the problems with hallucinations. Will step-by-step rewards help solve these problems? Do we need better warnings for stupid lawyers?If you like this episode please consider subscribing, comment and liking to help others find our podcast. We appreciate your support.----Chapters:----00:00 - Is GPT-4 Getting Worse?00:20 - Center for AI Safety Statement: Will AI Make Humans Extinct?12:09 - Is GPT-4 Getting Worse? Is Alignment the Problem?29:27 - GPT-4 Roadmap: 1M Context? Multi-modal?36:11 - Vision of ChatGPT: Have Plugins Failed?38:02 - Will OpenAI API's Have Too Much Competition?40:21 - Giant AI Models Aren't Over: Is GPT-4 Much More Capable?44:39 - ChatGPT for Lawyers: What Could Possibly go Wrong?48:17 - The Hallucination Problem53:22 - Process Supervision: Solution to Hallucination?55:59 - Japan Removes Copyright Laws for AI Progress57:53 - UAE's Falcon 40B Open Source & Royalty Free!1:01:04 - Will We Value Written Language?----Sources:----https://www.safe.ai/statement-on-ai-riskhttps://time.com/6283609/artificial-intelligence-race-existential-threat/https://www.nytimes.com/2023/05/30/technology/ai-threat-warning.htmlhttps://twitter.com/ylecun/status/1663616081582252032?s=46&t=uXHUN4Glah4CaV-g2czc6Qhttps://twitter.com/chrmanning/status/1663920862439915524?s=20https://news.ycombinator.com/item?id=36134249https://twitter.com/OfficialLoganK/status/1663934947931897857https://www.reddit.com/r/MachineLearning/comments/13tqvdn/uncensored_models_finetuned_without_artificial/?utm_source=share&utm_medium=ios_app&utm_name=iossmfhttps://web.archive.org/web/20230531203946/https://humanloop.com/blog/openai-planshttps://www.nytimes.com/2023/05/27/nyregion/avianca-airline-lawsuit-chatgpt.htmlhttps://simonwillison.net/2023/May/27/lawyer-chatgpt/https://twitter.com/drjimfan/status/1663972818160332800?s=46&t=uXHUN4Glah4CaV-g2czc6Qhttps://technomancers.ai/japan-goes-all-in-copyright-doesnt-apply-to-ai-training/https://twitter.com/emollick/status/1663960017697898497/photo/2https://www.tii.ae/news/uaes-falcon-40b-now-royalty-free

Neurolink's Brain Chip Gets FDA Approval, NVIDIA Stock Skyrockets, and OpenAI wants to regulate everyone but themselves.  We discuss LLM specialization and fine-tuned models like LIMA, Gorilla and ask is less training data more? We also discuss whether we would get the Neuralink brain implant. The future is arriving fast in AI this week and we try our best to keep you up to date.----If you like this podcast help us spread the word. Please consider leaving a comment, liking and subscribing if you haven't already.----00:00 - NVIDIA, Trillion Dollar AI Company?08:42 - Neuralink gets FDA approval for Human Trials15:30 - Can AI Read Your Thoughts? We Discuss Mind-Video20:12 - Sam Altman's World Regulation Tour23:25 - Governance of Superintelligence and Democratic AI29:16 - Meta's MEGABYTE: The Future of AI?34:36 - Specialized LLMs: Gorilla, LIMA48:34 - Adobe's Generative Fill in Photoshop54:23 - Microsoft CoPilot for Windows 11: Will it be Safe from Prompt Injection?57:42 - AI Training use Game Mechanics & More of AI Gaming----Sources:https://www.google.com/finance/quote/AMD:NASDAQ?window=5Yhttps://twitter.com/neuralink/status/1661857379460468736https://neuralink.com/https://mind-video.com/https://www.reuters.com/technology/openai-may-leave-eu-if-regulations-bite-ceo-2023-05-24/?utm_source=reddit.comhttps://openai.com/blog/governance-of-superintelligencehttps://openai.com/blog/democratic-inputs-to-aihttps://twitter.com/marvinvonhagen/status/1661735558065057798?s=20https://twitter.com/ylecun/status/1661853891690987523?s=20https://openai.com/research/universehttps://twitter.com/yusuf_i_mehdi/status/1661029509716676609?s=20https://blogs.windows.com/windowsdeveloper/2023/05/23/bringing-the-power-of-ai-to-windows-11-unlocking-a-new-era-of-productivity-for-customers-and-developers-with-windows-copilot-and-dev-home/https://arxiv.org/pdf/2305.15334.pdfhttps://arxiv.org/pdf/2305.11206.pdfhttps://twitter.com/Tom__Quan/status/1661326971207557122?s=20https://www.adobe.com/products/photoshop/generative-fill.html#watch-videohttps://www.artisana.ai/articles/meta-ai-unleashes-megabyte-a-revolutionary-scalable-model-architecturehttps://github.com/fmaclen/julia-sanfrancisco/https://julia.strictoaster.com/game/

In Episode 15 we explore AI regulation, new job-stealing capabilities for AI, and the dangers of autonomous drones, while pondering how to advise students on future careers. We discuss OpenAI's call for regulation amid fears of losing their monopoly, the rise of AI in gaming creating interactive stories, and their anxiety over autonomous killer drones. We also cover the latest news including ChatGPT for iOS and the wide release of Plugins for ChatGPT PRO users.Please consider leaving a review where you get your podcasts to help spread the word.CHAPTERS===00:00 - Cold open00:25 - Sam Altman Tries to Regulate AI & AI Regulation Senate Hearing03:29 - OpenAI's Lack of Moat to Open Source LLMs & Regulation 13:32 - Spear Phishing Attacks with AI LLMs & Bad Actors Using AI22:38 - ChatGPT for iOS now available in USA26:40 - The Future of AI: AI will be Everywhere29:30 - Which Jobs will be Displaced First and How?34:47 - As AI Chain of Thought Reasoning Improves Will More Jobs Go?40:32 - Connecting AI to Our World as a New Interface44:37 - Excitement on the Future of AI Gaming 48:51 - Will Startup Teams Be Smaller Due to AI & AI Regulation52:10 - Building Things with AI: Custom Software, Movies, Games55:35 - Palmer Luckey Interview: Autonomous Decision-Making AI DronesSOURCES:===https://www.theguardian.com/technology/2023/may/16/ceo-openai-chatgpt-ai-tech-regulationshttps://www.theguardian.com/technology/2023/may/16/ceo-openai-chatgpt-ai-tech-regulationshttps://openai.com/https://apps.apple.com/app/openai-chatgpt/id6448311069https://www.searchenginejournal.com/chatgpt-plugins-web-browsing-beta/486785/#closehttps://arxiv.org/ftp/arxiv/papers/2305/2305.06972.pdfhttps://twitter.com/CNBCOvertime/status/1657120760513679375https://arxiv.org/pdf/2305.04388.pdfhttps://www.theguardian.com/technology/2023/may/01/ai-makes-non-invasive-mind-reading-possible-by-turning-thoughts-into-texthttps://www.reddit.com/r/ArtificialInteligence/comments/13kfajq/im_excited_af_for_the_possibilities_of_ai_in/?utm_source=share&utm_medium=ios_app&utm_name=ioscss&utm_content=1&utm_term=1

Disclaimer: this description was written 100% by Anthropic's AI including the video title and text in the thumbnail. We have not modified it. Thanks Anthropic!If you like this podcast please consider subscribing, liking the video and leaving us a comment. We really appreciate the support.-----This week in AI is always packed with the latest news in artificial intelligence, but this episode delivers breakthroughs that are firmly in sci-fi territory. Chris and Michael discuss Anthropic releasing a 100,000-token context window for their language model—enabling it to understand entire novels or every customer record in your database at once. They explore what might be possible now, like AI systems writing convincing news articles, handling legal documents, or providing personalized healthcare recommendations with full medical histories.We are also awed by Google's new AI capabilities, including models that can translate between any languages, handle complex reasoning or generate code. However, they're concerned by the company forcing it into their search engine and email—noting most users won't appreciate or understand interacting with an AI.Finally, we dive into an eyebrow-raising use of AI: an influencer selling time with her "virtual girlfriend" chatbot for $1/minute. While concerning, the technology may point to AI companions becoming common for the elderly or lonely. Overall, this mind-blowing episode highlights how AI continues to shape the future at an exponential pace.CHAPTERS:====00:00 - Introduction00:35 - Anthropic’s 100k context size announcement21:58 - Google I/O announcements: PaLm 2, Bison & Makersuite30:04 - Google Generative AI Search and Workspace Updates39:37 - Google Search Perspectives42:01 - Google Bard Vs ChatGPT45:33 - Hugging Face Transformers Agent & Multimodal models54:09 - AI model memory and learning56:41 - The AI girlfriend chatbotSOURCES:====https://caryn.ai/https://www.anthropic.com/index/100k-context-windowshttps://www.mosaicml.com/blog/mpt-7bhttps://ai.google/discover/palm2/https://developers.generativeai.google/guidehttps://www.theverge.com/23718158/google-io-2023-biggest-announcements-ai-pixel-fold-tablet-android-14https://twitter.com/cutiecaryn/status/1653308852689928192/video/1https://workspace.google.com/blog/product-announcements/generative-aihttps://blog.google/products/search/generative-ai-search/https://huggingface.co/docs/transformers/transformers_agentshttps://twitter.com/DrJimFan/status/1642563455298473986/photo/1#AI #ArtificialIntelligence #MachineLearning #GPT3 #GenerativeAI  #Anthropic #Google #GoogleIO #HuggingFace #Chatbots #Snapchat #VirtualGirlfriends #ThisDayInAI #Bard #ArtificialIntelligence #MachineLearning #DeepLearning #NeuralNetworks #Chatbots #NLP #LanguageModels #GPT3 #BERT #ConversationalAI #AIAssistant #VirtualAssistant #Robotics  #TechNews #Innovation #Podcast #Technology #AI

In Episode 13 of This Day in AI We Discuss the Leaked Internal Google Document, "We Have No Moat, and Neither Does OpenAI", Discuss "The Godfather" of AI leaving Google, Discuss The Implications of Open Source LLM Advancement, Cover Open Assistant's Comparable Model to ChatGPT, OpenAI's Code Interrupter and Touch of Democratizing LLM Alignment.If you like this episode please consider subscribing and liking the episode to help others discover the podcast.TIMESTAMPS:====00:00 - Godfather of AI Quits Google Cold Open00:17 - Google & OpenAI Have No Moat: Advancement of Open Source LLMs21:13 - 'Godfather of AI' Quits Google30:54 - Open Assistant & Training LLMs using AI40:52 - What Increased Prompt Sizes Mean & Long Range Transformers50:49 - AI Agents, AI Disruption & Mark Zuckerberg's Agent Plan for MetaSOURCES====https://www.semianalysis.com/p/google-we-have-no-moat-and-neitherhttps://www.nytimes.com/2023/05/01/technology/ai-google-chatbot-engineer-quits-hinton.htmlhttps://www.theverge.com/2023/4/26/23699633/mark-zuckerberg-meta-generative-ai-chatbots-instagram-facebook-whatsapphttps://arxiv.org/pdf/2305.01625.pdfhttps://twitter.com/tanayj/status/1653243500354322436?s=46&t=uXHUN4Glah4CaV-g2czc6Qhttps://restofworld.org/2023/chatgpt-taking-kenya-ghostwriters-jobs/https://www.oneusefulthing.org/p/it-is-starting-to-get-strangehttps://www.nvidia.com/en-au/autonomous-machines/embedded-systems/https://open-assistant.io/chat