This episode covers practical applications of AI in everyday life, from simple email drafting to sophisticated company research. We discuss different prompting techniques such as few-shot prompting and chain of thought prompting to improve the quality and accuracy of AI-generated responses. We also discuss custom instructions, custom GPTs, and explore the powerful potential of AI in automating repetitive tasks. Tune in to discover how you can harness the power of AI to enhance productivity and streamline your workflows. === ⏰ CHAPTERS === 00:00: Introduction 08:02: Getting Started with AI and Identifying Tasks and Workflows 13:51: General-Purpose AI vs. Purpose-Built AI 27:30: Examples of ChatGPT usage 31:20: Prompt engineering 48:10: Custom instructions and custom GPTs === 🧠 KEY CONCEPTS === - Identifying AI Use Cases: Look for tasks in your life that are repeatable, time-consuming, and frustrating to see where AI can help. - General Purpose AI vs. Purpose-Built AI: General purpose AI (e.g. ChatGPT) is great for a wide range of tasks, while purpose-built AI (e.g. Adobe Acrobat AI) can be more efficient for specific tasks and workflows. - Prompt Engineering: The quality of AI output is significantly improved with well-crafted prompts. Use the few-shot prompting technique by giving examples to guide the AI model's response. - Chain of Thought Prompting: Encouraging AI to think out loud and show its reasoning can lead to more accurate and thoughtful responses. - Custom GPTs: Creating custom GPTs tailored to specific tasks can save time and increase efficiency. This involves setting custom instructions, uploading relevant files, and enabling specific capabilities. === 🔗 LINKS === - Homepage: https://artscienceai.substack.com/about - YouTube: https://www.youtube.com/@ArtScience-AI - Spotify: https://podcasters.spotify.com/pod/show/art-science-ai - Linkedin: https://www.linkedin.com/company/art-science-ai - Facebook: https://www.facebook.com/ArtScienceAI - Instagram: http://instagram.com/artscienceai/ - Threads: https://www.threads.net/@artscienceai - Twitter / X: https://x.com/ArtScienceAI === 💬 KEYWORDS === #AI #ArtificialIntelligence #GenerativeAI #GenAI #PromptEngineering #LLM #MachineLearning #ML #tech #podcast
Your one-stop-shop for all things Apple Intelligence! In this episode, we discuss the Apple Intelligence announcements from WWDC 2024. We discuss new AI capabilities in Apple’s apps and in Siri, the all-new AI assistant that can perform actions across multiple apps. We deep-dive into the architecture of Apple Intelligence, with a focus on privacy, and we discuss the implications of Apple Intelligence for users, developers, and the overall AI ecosystem.
In this episode we discuss the hype around AI and the challenges in achieving its full potential in 2024. The last 10% of solving problems with AI has proven to be difficult due to LLM hallucinations and reliability challenges. We discuss how this problem can be addressed by grounding LLMs with a knowledge base via the paradigm of Retrieval Augmented Generation (RAG). We discuss the different approaches to working with language models, including training from scratch, fine-tuning, and using RAG, and the opportunities for entrepreneurs in the AI space. Takeaways Generative AI may be the next major platform since the internet and mobile, but we are coming down from the peak of inflated expectations of the Gen AI hype cycle LLMs are general purpose models, and when asked domain-specific questions, LLMs tend to “hallucinate” (i.e. generate plausible-sounding answers) rather than admit ignorance Grounding in facts and providing relevant context can help mitigate the hallucination problem. Retrieval Augmented Generation (RAG) is a common paradigm for grounding LLMs in facts. As AI models and agents become commoditized and democratized, competitive moats will be built around proprietary data and tailored user experiences
Welcome to season 2 of Art and Science of AI! In this episode we reflect on our journey since season 1 last year. We discuss the impact of understanding AI on Piyush’s career in ad sales at Google. It enabled him to reimagine his job as selling AI rather than just selling ads. We also discuss the increasing importance of AI for Nikhil’s work as a Product Manager at Meta, and reflect on whether “AI is eating the world”! === 🔗 REFERENCES === - Art and Science of AI: https://ArtScienceAI.substack.com - Nikhil Maddirala: https://www.linkedin.com/in/nikhilmaddirala/ - Piyush Agarwal: https://www.linkedin.com/in/piyush5/ - “Why software is eating the world” by Marc Andreessen: https://a16z.com/why-software-is-eating-the-world/ === 💬 KEYWORDS === #AI #ArtificialIntelligence #GenerativeAI #GenAI #LLM #MachineLearning #ML #tech
In this episode we consider whether AI is having its "iPhone moment" as proclaimed by Nvidia’s CEO. The launch of the iPhone revolutionized the ways in which people interact with business and society, paving the way for innovative new businesses and social experiences such as Uber and Instagram. Will the rise of AI assistants and agents lead to a similar revolution? We discuss the evolution of AI, the dominance of big tech companies, and opportunities for entrepreneurs in this rapidly changing landscape. === ⏰ CHAPTERS === 00:00: Preview and intro 01:25: AI’s iPhone moment 08:00: ChatGPT plugins and entrepreneurship opportunities 16:17: Dominance of big tech in AI 23:17: AI industry landscape === 🔗 REFERENCES === Art and Science of AI: ArtScienceAI.substack.com (detailed show notes and full episode transcripts) Nikhil Maddirala: https://www.linkedin.com/in/nikhilmaddirala/ Piyush Agarwal: https://www.linkedin.com/in/piyush5/ === 💬 KEYWORDS === #AI #ArtificialIntelligence #GenerativeAI #GenAI #LLM #MachineLearning #ML #tech #podcast
In this episode we discuss the limitations of ChatGPT such as its lack of up-to-date knowledge, lack of private / domain specific knowledge, and lack of tool use abilities. We then discuss innovative new solutions to these problems such as vector database retrieval, ChatGPT plugins, and AI agents with access to external tools that can orchestrate complex tasks and workflows autonomously. What are the implications for business and society? We also touch upon some criticisms of OpenAI’s convoluted governance structure and lack of transparency into their model training and data sources.
In this episode we go beyond classical machine learning into the fascinating world of neural networks. We discuss how neural networks, inspired by the human brain, revolutionize our ability to process unstructured data like images and text. Using a detailed example of handwriting digit recognition, we break down how neural networks learn patterns, make predictions, and transform raw data into valuable insights. Tune in to explore the magic of hidden layers, the significance of activation functions, and the trade-offs between model power and interpretability in modern AI systems.
In this episode, we dive into the fundamentals of AI, starting with its basic definition and exploring how it has evolved over time. We discuss the differences between two approaches to AI — rule based systems and machine learning — using examples such as classification for spam filtering and linear regression for house price prediction. We dive deeper into classical machine learning, introducing the concepts of model architecture, features, parameters, objective functions, and training algorithms. We conclude by considering some of the limitations of classical machine learning, and the rise of neural networks.
In this introductory episode of Art and Science of AI we discuss our background, expertise, and motivation for embarking on this journey to demystify AI! Piyush shares his fascination with seemingly magical AI technologies like ChatGPT and MidJourney, and his desire to understand the underlying mechanics. Nikhil shares his background in AI, and his plan for demystifying ChatGPT in this season: starting with the basics of Artificial Intelligence (AI) and Machine Learning (ML), we will demystify key concepts such as neural networks, deep learning, and Large Language Models (LLMs). We will also explore the exciting potential of ChatGPT and its implications for business and society. Note: Season 1 was originally recorded back in May 2023 as a continuous 3-hour long conversation. Season 2 is now live, so please subscribe for new episodes every week! artscienceai.substack.com === 🔗 REFERENCES === Art and Science of AI: ArtScienceAI.substack.com Nikhil Maddirala: https://www.linkedin.com/in/nikhilmaddirala/ Piyush Agarwal: https://www.linkedin.com/in/piyush5/ === 💬 KEYWORDS === #ChatGPT #AI #ML #ArtificialIntelligence #MachineLearning #LLM #tech
In this episode we delve into the world of generative AI with a focus on language modeling. We explore how text is transformed into semantically meaningful data through the use of vector embeddings, providing an in-depth look at the mechanics behind AI models like Open AI’s GPT-3. Discover the complexities of semantic relationships in language, the role of mathematical concepts in AI, and the advancements that make generative AI both powerful and conversational.