Sander Schulhoff

Sander Schulhoff, author of The Prompt Report and creator of LearnPrompting.org, dives deep into the intricacies of prompt engineering and AI safety. He discusses the extensive research landscape with over 1,600 papers on prompting, clarifying popular methodologies like chain-of-thought and zero-shot prompting. Schulhoff also tackles ethical dilemmas surrounding AI-generated academic work and introduces innovative AI tools to streamline research methodologies. Further, he covers advancements in multimodal prompting, illustrating the complexities and potential of AI interactions.

Sander Schulhoff, creator of LearnPrompting.org and author of the Prompt Report, dives deep into innovative prompting techniques for AI. He discusses best practices and the challenges faced in ensuring effective prompts. The conversation highlights new developments in generative AI courses and automated systems, along with insights into advanced methods like few-shot prompting. Schulhoff also addresses misconceptions in prompt optimization and the critical role of training methodologies, revealing how these elements shape the future of AI performance and collaboration.

Sander Schulhoff, CEO of Learnprompting.org and a researcher in prompt engineering, shares insights on navigating the complexities of AI interaction. He emphasizes essential techniques for effective language modeling, revealing the power of few-shot prompting. Schulhoff highlights challenges in evaluating AI models and discusses vulnerabilities, including user manipulation and creative input modifications. The conversation also touches on the role of self-generated examples in problem-solving and the future of AI skills in professional environments.

An interview with Sander Schulhoff, creator of learnprompting.org, the largest prompting resource online. Watch the video version on YouTube: https://youtu.be/BNrPs3ANz1E References: ►Listen on Spotify: ►Listen on Apple podcasts: ►Learn prompting: https://learnprompting.org/ ►Twitter: https://twitter.com/learnprompting, https://twitter.com/Whats_AI ►My Newsletter (A new AI application explained weekly to your emails!): https://www.louisbouchard.ai/newsletter/ ►Support me on Patreon: https://www.patreon.com/whatsai ►Support me through wearing Merch: https://whatsai.myshopify.com/ ►Join Our AI Discord: https://discord.gg/learnaitogether Chapters: 0:00 Hey! I hope you enjoy this interview. Follow to learn more about AI and prompting with amazing upcoming guests! 00:30 Who are you? What's your background? 01:40 How where you involved with reinforcement learning? 03:10 How was AI used within Minecraft? 06:04 What made you transition into prompting? 07:03 What is learnprompting.org? 08:57 How does prompting work? What are common types of prompting? Is it only in English? 10:48 What can you improve with prompting vs. just natural language and talking with another human being? 13:49 How do LLM work? Why are they able to understand and generate text? Why can’t it be able to understand us like other humans do? 15:51 Is the main thing that LLM’s lack emotion? Or is there another big difference? 17:27 Can you give an example of where you used prompting? 19:11 Can you give more details on what is Lang Chain and how it’s used? 20:17 How can one benefit from using prompting vs. Googling the answer? 24:38 Can you get paid by being a prompt engineer? 25:55 What do prompt engineers do? What are their tasks? 30:40 Who can become a prompt engineer? 32:48 Is there any way to measure that you are improving and going in the right direction for improving prompts for a specific task? 35:24 How would you recommend people to learn prompting? 37:57 Is prompt engineering a job for the future? 39:00 Do you think prompt engineering might get better and better as to in the end replace a human? 40:40 What are the differences in prompting between two separate models and a model over time? 42:24 Is there a way to be sure you’re not specializing on only one model? 43:54 Is it important for prompt engineers to better understand models? 45:49 Could you explain the difference between fine tuning and prompting? 49:29 Are there any misconceptions that you often need to clarify to those new to prompting? 51:46 Do you see any clear limitations of prompting? 55:12 What are the biggest challenges you see for prompting? 56:23 Was GPT-3 trained to be unbiased? 58:43 Did you solve the issue of ChatGPT not having only one word? 01:00:38 Why do you think it’s so difficult to output only one word? 01:01:37 Do you think the model doesn’t listen or it doesn't understand? 01:02:55 Can you think of any ethical problems that may arise because of this new technology? 01:04:09 What are you currently working on? What’s your next project?

Join Sander Schulhoff, a specialist in prompt engineering, Philip Resnik, a computational linguistics professor, and Dennis Peskoff from Princeton as they delve into the cutting-edge world of AI. They explore the security risks of prompt hacking and its implications for military use. Discussion highlights include the evolving role of generative AI across various fields, innovative techniques for improving AI self-criticism, and the pressing need for energy-efficient large language models. Their insights offer a fascinating glimpse into the future of AI research.