01 · Understanding how AI works
Tokens, how it generates an answer, and why it sometimes gets things confidently wrong.
- Why AI predicts text rather than looking anything up
- The knowledge cutoff, and what "memory" actually means
- Hallucination: what it is, and why confident isn't the same as right
02 · RCFT
Role, Context, Format, Tone - the brief that gets a usable answer first time, not a re-write.
- Give AI a role, not just a question
- The context it actually needs to know
- Why Tone is the step most people skip, and shouldn't
03 · Reasoning frameworks
Structured thinking, not one vague free-form question.
- Ask AI to work through a problem step by step
- Structured formats: pros/cons, risk breakdowns, SWOT-style thinking
- Turns a vague answer into one you can actually rely on
04 · Governance & safe use
What's safe to paste in, and what to always verify.
- What's safe to paste into an AI tool, and what isn't
- Free and personal plans vs paid business plans - what actually changes
- Never let AI be the last check on a number, a clause, or a fact that matters
05 · One voice
Set the brief once. Every output stays consistent after that.
- A dedicated space that holds your context, so you're not re-explaining who you are every time
- Carry that same voice into everything else you use AI for
- The company handbook approach: written once, everyone works from the same standard
Why this matters for how you're found, too: AI tools like ChatGPT don't search the web like Google does - they generate an answer from what they've learned. Understand that, and AEO (Answer Engine Optimisation) follows naturally: it's about getting your business into the answer AI gives, not just onto a page Google ranks. Different mechanism, different approach - worth understanding on its own terms rather than treating as "SEO for AI."