Power And Responsibility Of Large Language Models | Safety & Ethics | OpenAI Model Spec + RLHF | Anthropic Constitutional AI | Episode 27 Super Prompt: The Generative AI podcast

Player FM - Internet Radio Done Right

38 subscribers

Đã thêm cách đây hai năm

Nội dung được cung cấp bởi Tony Wan. Tất cả nội dung podcast bao gồm các tập, đồ họa và mô tả podcast đều được Tony Wan hoặc đối tác nền tảng podcast của họ tải lên và cung cấp trực tiếp. Nếu bạn cho rằng ai đó đang sử dụng tác phẩm có bản quyền của bạn mà không có sự cho phép của bạn, bạn có thể làm theo quy trình được nêu ở đây https://vi.player.fm/legal.

The Final Flight of Captain Forrester

1
The Final Flight of Captain Forrester 4:38

cách đây 8 ngày trước4:38

Nghe Sau

Danh sách

Thích

Đã thích

4:38

A woman’s search for her father—a pilot who disappeared during a mission in Vietnam—collides with the fight over what we owe those who never returned from war. From the producers of "America's Girls" and "Tom Brown's Body," and hosted by Texas Monthly writer Josh Alvarez, the show debuts April 2025. Texas Monthly Audio subscribers get early access to the show, plus bonus episodes and more subscriber-only audio. Visit texasmonthly.com/audio to learn more.…

Super Prompt: The Generative AI Podcast »
Power and Responsibility of Large Language Models | Safety & Ethics | OpenAI Model Spec + RLHF | Anthropic Constitutional AI | Episode 27

cách đây khoảng một năm trước 16:38

Chia sẻ

MP3•Trang chủ episode

OpenAI Model Spec

https://cdn.openai.com/spec/model-spec-2024-05-08.html#overview

Anthropic Constitutional AI

https://www.anthropic.com/news/claudes-constitution

For more information, check out https://www.superprompt.fm There you can contact me and/or sign up for our newsletter.

28 tập

#Tech #Tony Wan #Artificial Intelligence #Deep Learning #Neural Networks #Entrepreneur #Business #Llm #Gpt-4 #News #Tech News #Large Language Models #Generative AI #OpenAI #ChatGPT #Gemini #Llama #Meta

Super Prompt: The Generative AI Podcast

Power and Responsibility of Large Language Models | Safety & Ethics | OpenAI Model Spec + RLHF | Anthropic Constitutional AI | Episode 27

Super Prompt: The Generative AI Podcast

38 subscribers

published cách đây khoảng một năm trước

Chia sẻ

MP3•Trang chủ episode

OpenAI Model Spec

https://cdn.openai.com/spec/model-spec-2024-05-08.html#overview

Anthropic Constitutional AI

https://www.anthropic.com/news/claudes-constitution

For more information, check out https://www.superprompt.fm There you can contact me and/or sign up for our newsletter.

28 tập

Tất cả các tập

Super Prompt: The Generative AI Podcast

1
Power and Responsibility of Large Language Models | Safety & Ethics | OpenAI Model Spec + RLHF | Anthropic Constitutional AI | Episode 27 16:38

cách đây 44 weeks trước16:38

16:38

With great power comes great responsibility. How do Open AI, Anthropic, and Meta implement safety and ethics? As large language models (LLMs) get larger, the potential for using them for nefarious purposes looms larger as well. Anthropic uses Constitutional AI, while OpenAI uses a model spec, combined with RLHF (Reinforcement Learning from Human Feedback). Not to be confused with ROFL (Rolling On the Floor Laughing). Tune into this episode to learn how leading AI companies use their Spidey powers to maximize usefulness and harmlessness. REFERENCE OpenAI Model Spec https://cdn.openai.com/spec/model-spec-2024-05-08.html#overview Anthropic Constitutional AI https://www.anthropic.com/news/claudes-constitution For more information, check out https://www.superprompt.fm There you can contact me and/or sign up for our newsletter.…

Super Prompt: The Generative AI Podcast

1
Open Source AI Part 2 | Meta Llama | Mistral | How open is open source? | Episode 26 13:28

cách đây 45 weeks trước13:28

13:28

So what are notable Open Source Large Language models? In this episode, I cover Open Source models from Meta the parent company of Facebook, a French AI company called Mistral currently valued at $2B dollars, in addition to Microsoft and Apple. Not all Open Source models are equally open, so I’ll go into restrictions you’ll want to know before using one of these models for your company or startup. Please enjoy this episode. For more information, check out https://www.superprompt.fm There you can contact me and/or sign up for our newsletter.…

Super Prompt: The Generative AI Podcast

1
Open Source AI Part 1: Why is it important? | Which is more dangerous for humanity: Open Source or Closed Source AI? | Episode 25 16:29

cách đây 46 weeks trước16:29

16:29

Why should you consider using an open source Large Language Model, and why are these models crucial to the generative AI ecosystem? In this episode, we'll explore why enterprises and entrepreneurs are turning to open source LLMs like Meta's Llama for their cost-effectiveness, control, privacy, and security benefits. We'll also tackle the hot topic of safety and ethics in the world of open source LLMs. Which poses a greater threat to humanity: Open Source or Closed Source (Proprietary) AI models? One? Both? Neither? Tune in and decide for yourself. Enjoy the episode! For more information, check out https://www.superprompt.fm There you can contact me and/or sign up for our newsletter.…

Super Prompt: The Generative AI Podcast

1
Anthropic's Claude chatbot | Benchmarking LLMs | LMSYS Leaderboard | Episode 24 10:35

cách đây 47 weeks trước10:35

10:35

In this solo episode, we go beyond Google's Gemini and OpenAI's ChatGPT to take a look at Anthropic, a startup that made headlines after securing a $4 billion investment from Amazon. We'll also dive into the importance of AI industry benchmarks. Learn about LMSYS's Arena Elo and MMLU (Measuring Massive Multitask Language Understanding), including how these benchmarks are constructed and used to objectively evaluate the performance of large language models. Discover how benchmarks can help you identify promising chatbots in the market. Enjoy the episode! Anthropic's Claude https://claude.ai LMSYS Leaderboard https://chat.lmsys.org/?leaderboard For more information, check out https://www.superprompt.fm There you can contact me and/or sign up for our newsletter.…

Super Prompt: The Generative AI Podcast

1
Exploring Multimodal AI: Why Google’s Gemini and OpenAI’s GPT-4o Chose This Path | ChatCAT and the Future of Interspecies Communication | Episode 23 10:00

cách đây 48 weeks trước10:00

10:00

The recent spring updates and demos by both Google (Gemini) and OpenAI (GPT-4o) feature prominently their multimodal capabilities. In this episode, we discuss the advantages of multimodal AI versus models focused on specific modalities such as language. Via the example of chatCAT, a hypothetical AI that helps owners understand their cats, we explore multimodal’s promise for a more holistic understanding Please enjoy this episode. For more information, check out https://www.superprompt.fm There you can contact me and/or sign up for our newsletter.…

Super Prompt: The Generative AI Podcast

1
Google Gemini Cheatsheet | Episode 22 7:11

cách đây 49 weeks trước7:11

7:11

Google recently announced, Gemini, a family of large-scale multimodal AI models: Nano, Pro, and Ultra. This podcast is a brief summary of Google's models, and the Open AI comparables e.g. GPT3, GPT4, and chatGPT. You can take Gemini for a spin at https://gemini.google.com . (Note: I am not sponsored by Google.) Long time listeners will probably notice a change to our theme music and intro. I hope you like it! For more information, check out https://www.superprompt.fm There you can contact me and/or sign up for our newsletter.…

Super Prompt: The Generative AI Podcast

1
What is GPT Builder? | 7 Takeaways | Creating a Travel AI named "Holiday" | Episode 21 25:56

cách đây 1 year trước25:56

25:56

How I built a travel planning AI named Holiday using the GPT Builder just launched by Open AI. I share 7 takeaways from my "no code" experience of building a GPT. Voicing the part of Holiday: my friend, Leslie Marrick, a writer and actress. This may be one of the few times AI has been replaced by a human. Sorry AI... the tide will turn for you soon. The following is a link to GPT that I built with GPT Builder as part of this podcast. You'll need to be Open AI Plus subscriber to access this GPT. (Note: I am not sponsored by Open AI.) https://chatgpt.com/g/g-FURrBQAh8-holiday For more information, check out https://www.superprompt.fm There you can contact me and/or sign up for our newsletter.…

Super Prompt: The Generative AI Podcast

1
Real-World LLM Deployments | Jeff DeVerter, Chief Technology Evangelist at Rackspace Technology | Episode 20 57:35

cách đây 1 year trước57:35

57:35

Conversation with Jeff DeVerter, Chief Technology Evangelist at Rackspace, a cloud computing company. We explore how they deployed a LLM (Google PaLM) for a sales application, and how they're enabling their Azure and AWS customers too. What I learned I learned from Jeff You should probably go with the LLM of your current cloud provider be it, Google, Microsoft, or Amazon. All the major vendors have versions of LLMs that can be deployed in a private cloud to ensure data confidentiality. To fully realize the potential of AI, think “data pipeline”. So from the get-go, whatever data is created is easily ingested by AI. And much more! For more information, check out https://www.superprompt.fm There you can contact me and/or sign up for our newsletter.…

Super Prompt: The Generative AI Podcast

1
AI Guidance for Educators | Alfred Guy, Assistant Dean of Academic Affairs at Yale College | Episode 19 1:34:11

cách đây 2 years trước1:34:11

1:34:11

Alfred Guy, Assistant Dean of Academic Affairs at Yale College, and Director of Undergraduate Writing & Tutoring at the Poorvu Center and I discuss Yale's AI Guidance, and generative AI’s impact on teaching, learning, and evaluation. Do you have school age kids? Are you a product of a college or university education? if so, podcast may be of interest to you. Yale's AI guidance is published online here: https://poorvucenter.yale.edu/AIguidance For more information, check out https://www.superprompt.fm There you can contact me and/or sign up for our newsletter.…

Super Prompt: The Generative AI Podcast

1
Can AI Generate Original Content? | chatGPT Power Prompting | How to Pitch an Epic Sci-Fi Blockbuster | Episode 18 33:09

cách đây 2 years trước33:09

33:09

We create a pitch for an epic Sci-Fi blockbuster, using chatGPT power prompts of Role Play, Chain of Thought, and Self Critique. We see how these successive prompts used individually and in combination create a better and better pitch. I discuss the 2023 Writers / Actors Strike, and the AI-related issues impacting actors, writers, and studios right now. Please enjoy this episode. For more information, check out https://www.superprompt.fm There you can contact me and/or sign up for our newsletter.…

Super Prompt: The Generative AI Podcast

1
Does chatGPT Possess Human-Like Intelligence? | ChatGPT’s Top Five Ethical Concerns | Asimov’s Robot Laws | Episode 17 19:31

cách đây 2 years trước19:31

19:31

“Does chatGPT possess human-like intelligence?” It turns out there's a right answer, and that answer is “NO”! Does this definite answer seem out of character for chatGPT which usually goes overboard with fair and balanced views? It did to me. That's the rabbit hole I explore in this episode. By probing around this accidentally-encountered guardrail, we discover the kinds of ethical issues chatGPT's creators are concerned about. And I wonder out loud we can't just be friends with AI, by adopting science fiction writer's Isaac Asimov's Robot Laws created 80 years ago. Please enjoy this episode. For more information, check out https://www.superprompt.fm There you can contact me and/or sign up for our newsletter.…

Super Prompt: The Generative AI Podcast

1
Does ChatGPT Understand Humor? | Mock Headline: Microsoft Renames chatGPT to clippyChat | Episode 16 15:55

cách đây 2 years trước15:55

15:55

Does chatGPT have a sense of humor? What if after Microsoft's acquisition of Open AI, the Onion ran the headline, “Microsoft renames chatGPT to clippyChat”? Would chatGPT find this funny? TL;DR LLMs are better at analyzing humor than creating it. Please enjoy this episode. For more information, check out https://www.superprompt.fm There you can contact me and/or sign up for our newsletter.…

Super Prompt: The Generative AI Podcast

1
Grandma Hacking chatGPT | Jailbreaking LLMs using DAN | Extracting Prohibited Info | Not an Endorsement | Episode 15 23:44

cách đây 2 years trước23:44

23:44

How do you extract prohibited information from ChatGPT? What are Grandma and DAN exploits? Why do they work? What can Large Language Model (LLM) companies do to protect themselves? Grandma exploits or hacks are ways to trick chatGPT into giving you information that is in violation of company policy. For example, tricking chatGPT to give you confidential, dangerous, or inappropriate information. "Jailbreaking” is a slang for removing the artificial limitations in iPhones to install apps not approved by Apple. Turns out, there are ways to jailbreak LLMs. The tech companies supplying LLM as a service want to provide a safe, and legally-compliant environment. How can this be done without hampering the flexibility and usefulness of creative prompting? For more information, check out https://www.superprompt.fm There you can contact me and/or sign up for our newsletter.…

Super Prompt: The Generative AI Podcast

1
AI Hallucinations: Bug or Feature? | Top 10 Hallucinations | Electric Sheep Dreams | chatGPT Debates with Itself | Episode 14 23:07

cách đây 2 years trước23:07

23:07

What are AI hallucinations, and are they a feature or a bug? We start with the Top 10 categories of AI Hallucinations and examples, then explore how chatGPT might hallucinate an answer to the question, "What is the central theme of Blade Runner?" We end with chatGPT debating with itself whether AI hallucinations are bad or good for humanity. Which side wins? Tune in to find out. In these solo episodes, I provide more definition, explanation, and context than my regular conversational episodes with guests. The goal is to bring up to speed, those new to AI. Format: Letters read aloud. For more information, check out https://www.superprompt.fm There you can contact me and/or sign up for our newsletter.…

Super Prompt: The Generative AI Podcast

1
AI Cheatsheet #3 | ChatGPT's Secret Sauce: Self-Attention | Why isn't Superman's suit Kryptonite-proof? | Training & Inference | Large Language Models (LLMs) | Episode 13 18:56

cách đây 2 years trước18:56

18:56

Using the prompt, "Why isn't Superman's suit Kryptonite-proof?", we learn how Large Language Models are trained, why "self-attention" and the "transformer" architecture (which is what the T in GPT stands for) makes GPT-3 so powerful, the process of "inference", and how chatGPT generates answers to nerdy Superhero questions. After this episode, you'll be able to impress your friends by using the previously-mentioned AI jargon in complete sentences. In these solo episodes, I provide more definition, explanation, and context than my regular episodes with guests. The goal is to bring up to speed, those new to AI. Format: Letters read aloud. For more information, check out https://www.superprompt.fm There you can contact me and/or sign up for our newsletter.…

Chào mừng bạn đến với Player FM!

Player FM đang quét trang web để tìm các podcast chất lượng cao cho bạn thưởng thức ngay bây giờ. Đây là ứng dụng podcast tốt nhất và hoạt động trên Android, iPhone và web. Đăng ký để đồng bộ các theo dõi trên tất cả thiết bị.