Before you go deep on any single AI tool, learn the prompt patterns and mental models that make every one of them better. Tool-agnostic. Always free.
Most people learn AI tool-first — they pick up ChatGPT or Copilot and collect tricks. Foundations flips that. These fourteen lessons teach the patterns that sit underneath every tool: how to get a specific, non-generic answer on the first try, how to catch a confident-sounding mistake before it costs you, how to pick the right AI for the task in front of you, and how to use AI on sensitive work without giving away anything you shouldn't.
Learn them once and every other track gets easier — you're no longer memorizing tool-specific tricks, you're applying judgment that holds as the models change. Free forever, about 65 minutes total, and the single highest-leverage hour on the whole site.
Each lesson stands alone — but they build on each other if you take them in order.
Watch four AIs attempt the same task, see why their first drafts all sound generic, then learn the two prompt patterns that produce dramatically better output on any AI: specific friction and the constraint stack.
Every AI makes things up sometimes. The skill isn't avoiding hallucinations — it's recognizing them fast. The 4 patterns that should trigger a fact-check, and the 3 prompt tricks that drop hallucination rates dramatically.
ChatGPT, Claude, Gemini, Perplexity, Copilot, Grok — they're not interchangeable. A simple decision tree for picking the right AI by task type, plus the use cases each one wins.
The cases where reaching for AI costs you more than just doing it yourself. Recognizing them saves you from looking like the person who lets AI write everything (including the bad parts).
Customer data, internal docs, NDAs, code — what's safe, what's risky, and the simple workflow that lets you use AI on sensitive work without giving away the farm.
Lock down the right privacy settings, harden your accounts, and recognize the new threats — prompt injection, AI phishing, and deepfakes — that come with putting AI in your workflow.
iOS 27 lets you pick ChatGPT, Claude, Gemini, or Grok for Siri, Writing Tools, and Image Playground. How to set it, which model for which task, and the privacy trade-off.
A vendor-neutral leaderboard ranked by monthly active users: ChatGPT, Gemini, Copilot, Claude, DeepSeek, Perplexity, Grok and more — what each is best at, and how to read the numbers honestly.
The four pricing models behind every AI tool, what a token really is, and what a single prompt costs in real money. The lesson that makes every pricing page make sense.
No math, no hype: what a language model actually does, why it's brilliant and unreliable at the same time, and the one mental model that makes every AI behavior make sense.
The 15 AI terms that actually matter — context window, tokens, RAG, agents, fine-tuning, inference, multimodal and more — each in plain English with the 'why you care' attached.
Designing a personal AI stack: the lane map for choosing which tool owns which job, the two-tool baseline most professionals land on, subscription math, and when to add or drop a tool.
The Pro+ layer above tools: portable voice guides and prompt libraries, per-tool memory management, workflow documentation, and the weekly review that turns AI use into accumulating capability.
A fast framework for the weekly firehose of AI launches: the three-question filter, the 30-minute hands-on protocol, red flags that end evaluations early, and the discipline of 'watch' as a verdict.
The money layer of your AI stack: the spend inventory, hour-saved accounting, usage-billing guardrails for APIs and agents, seat audits for teams, and the quarterly true-up.
Interactive: enter how much you use AI and see pay-per-use vs. subscription vs. Copilot Cowork credits, side by side.
Foundations gives you the patterns; the Tool Tracks teach you to master each AI. Pro membership unlocks every lesson, every track, $9/mo founding pricing.
See Pro membership →