Welcome to a special bonus episode of the Art of A.I. for Business, hosted by our guest and digital colleague, Robo. Today, we dive into the engaging and fascinating world of 'prompt engineering' and the role of 'tokens' in Artificial Intelligence. Tokens are the building blocks that AI models use to understand and generate language. As we chat with an AI, it takes our input, breaks it into tokens and uses them to form responses. However, an AI model's memory has a token limit, and once this limit is reached, the oldest tokens are forgotten. Robo illustrates this through an interactive game and also provides practical tactics for making the most of tokens when engaging with AI models. We discuss strategies for remembering crucial information, managing the context window, and working with APIs. Join us as we explore the art of prompt engineering, a fascinating dance between human skills and AI capabilities. Remember, AI tools don't relieve us of our thinking, they complement it. Until our next episode, goodbye from the Art of A.I. for Business!

Key Takeaways:

• [01:17 - 01:56] Tokens are the building blocks of AI models for language processing
• [01:57 - 03:38] Understanding token limits
• [03:38 - 04:15] Working with APIs

Reference: Cuomo, J. (2023, August 2). My name is Jerry… What’s my name? Medium. https://medium.com/@JerryCuomo/my-name-is-jerry-whats-my-name-7401a8202880

In this bonus episode, host Jerry Cuomo explores the intriguing concept of "Training AI to Forget." Joined by 'guest' Dr. Eva Rosenberg, a fictional memory psychologist portrayed using voice synthesis tech, they discuss the intersection of neuroscience, ethics, and computer science in AI. Jerry shares insights from his recent article and introduces "The Neuralyzer," a code experiment that enables AI systems to forget specific terms or data, aligning with governance and ethical considerations like GDPR.

In a surprising twist, Jerry reveals that the role of Dr. Eva Rosenberg was performed by him using technology from Respeecher, an AI startup known for voice synthesis. The episode explores how Respeecher responsibly navigates the ethical landscape of synthetic voices.

This episode serves as a thought-provoking journey into the future of trustworthy AI, examining both the technological and ethical dimensions. Listen now to get inspired and educated about the next frontier in AI governance.

Key Takeaways:

• [00:31 - 00:47] Intro Dr. Eva Rosenberg
• [01:17 - 01:28] Forgetting is a feature
• [02:45 - 03:24] The Neuralyzer
• [04:03 - 04:17] Ongoing Research
• [05:11 - 05:42] I'm Eva
• [05:43 - 06:04] Respeecher
• [06:24 - 06:34] Simulation
• [07:03 - 07:19] Ethical use of synthetic voice

References:

Art of AI Article Series - Training AI to Forget: The Next Frontier in Trustworthy AI

Art of AI Github Repo (neuralizer.py) source code

Towards Making Systems Forget with Machine Unlearning

Machine Unlearning

Selective Forgetting in Deep Networks

Adaptive Machine Unlearning

Respeecher - wikipedia

Respeecher - website

In this bonus episode, guest host Llama2, a Large Language Model, welcomes LangChain, an innovative framework engineered to elevate the capabilities of language models. The conversation aims to highlight the value LangChain brings to LLM-based applications by acting as a bridge between LLMs and fresh, real-time data sources. Whether it's providing up-to-date stock information for a financial advisory chatbot or tailoring fashion advice based on current weather conditions, LangChain's functionalities are designed to empower applications with contextual awareness and relevant responses.

LangChain explains its unique approach to data retrieval, cleaning, and summarization, along with its seamless integration with LLMs for generating context-rich, human-like responses. The episode also explores how LangChain uses elements like 'Links' and 'Chains' to create efficient workflows, making it a versatile tool for developers. Lama2 and LangChain delve into the endless possibilities that arise when combining the expertise of LLMs with real-time data, opening the doors to smarter, more contextually aware applications.

Tune in to explore how LangChain and LLMs like Llama2 are pushing the boundaries to make chatbots and other applications not just smarter, but incredibly useful.

References:

Introduction to LangChain - Tutorial on GeeksforGeeks.com

Build a chatbot with Llama 2 and LangChain