In today’s episode, let’s explore how AI model alignment works and why it matters. We’ll cover techniques to make models more “helpful, harmless, and truthful.” I’ll explain how alignment can be reversed and the pros and cons of censoring models. Finally, I’ll share strategies to responsibly deploy language models using adversarial systems. There’s a lot…
Summary In today's episode, I break down a recent paper showing how easily attackers strip alignment from large language models like GPT-4 and Llama 2. Here's what this means for you. You need to understand that the safety guardrails on these models wear surprisingly thin, which matters enormously if you plan to deploy them in…
Almost Timely News: A Deep Dive on Prompt Libraries (2023-11-19) :: View in Browser 👉 Watch the newest version of my talk, The Intelligence Revolution, recorded live at DigitalNow 2023, now with more talking robot dogs! (plus get the slides) 📺 Content Authenticity Statement 100% of this newsletter’s content was generated by me, the human.…
Summary In today's episode, I walk through Trust Insights' new Generative AI for Marketers program, available as either on-site workshops or a self-paced online course. Here's what this means for you. You get a structured, hands-on path to build practical AI skills that deliver real time savings and cost efficiencies for your team. You'll also…
In today’s episode, I address audience questions about data privacy and paid search in relation to AI. We discuss settings that allow opting out of training datasets and examine emerging ad models like Bing. As AI takes up more search real estate, paid listings become crucial for visibility. Join me as we explore the intersection…
In today’s episode, I compare the strengths of ChatGPT and Anthropic’s Claude. While Claude generates higher quality text, ChatGPT excels at tasks like data analysis. For real-time answers, search-based models like Bing avoid AI hallucinations. Join me as I demo tools to match your use case – from writing to coding and beyond. Can’t see…
Almost Timely News: Getting Started with OpenAI Custom GPTs (Custom ChatGPT) (2023-11-12) :: View in Browser 👉 📕 Get my new book, The Woefully Incomplete Book of Generative AI, absolutely free of financial cost 📕 👉 Watch the newest version of my talk, The Intelligence Revolution, recorded live at DigitalNow 2023 last week, now with…
In today’s episode, I explain how to leverage neural style transfer for brand voice consistency. When training multiple AI personas like “boss” and “conference,” drift can occur. Condense distinctive writing styles into prompts with second-person imperatives. Join me to see how generating a centralized style guide maintains tone across all your AI content. Can’t see…
Summary In today's episode, I walk through how to use ChatGPT for qualitative survey analysis, from uploading a CSV to generating downloadable charts, categorized spreadsheets, and a methodology document. Here's what this means for you. You'll discover a practical workflow that turns hours of manual coding and pivot tables into roughly 45 minutes of AI-assisted…
In today’s episode, I discuss mitigating the risk of AI hallucination and falsehoods. Pure language models like GPT-4 can make convincing yet untrue claims. Tools like Bing and Google Bard cite sources so you can verify authenticity. Join me to explore best practices for reducing made-up responses from generative AI. Can’t see anything? Watch it…
