Everyday AI Podcast – An AI and ChatGPT Podcast AI Without the Jargon: The Language Every Business Leader Needs in 2026 (Start Here Series Vol 2)
123 snips
Jan 16, 2026 Explore the alarming AI gap stunting company growth as leaders struggle to communicate effectively about new technologies. Discover how generative AI has grown rapidly, widening the understanding gap. Learn the essential framework for transforming human prompts into actionable AI outcomes and how businesses can leverage training for ROI. Delve into concepts like tokenization, RAG, and agentic models, while understanding the risks of AI, such as hallucinations and prompt injection. Finally, empower leaders to ask the right questions for successful AI implementation.
AI Snips
Chapters
Transcript
Episode notes
Models Think In Tokens Not Words
- Jordan explains LLMs operate on tokens not words, converting text into token sequences for prediction.
- He notes tokens average ~4 characters and token context, not raw words, drives understanding.
Keep Important Instructions First
- Jordan explains context windows behave like rolling hard drives that drop oldest info when full.
- He advises keeping critical instructions early because models forget earlier context as tokens cycle.
Parameters: Capability Versus Cost
- Parameters are neural connections and roughly equate to a model's capability and computational cost.
- Jordan predicts many tasks will move to smaller specialized models over time.