AI Guides › Playbooks
By Nigel Guy · 7 min read
Most people respond to AI at work by collecting tools and hoping the right skill turns up. That feels like progress, but it skips the prior question: what are you actually trying to win? A promotion, a safe salary, a business of your own and a reputation for deep craft all reward different moves, and AI changes the price of each one differently. Without a choice, you drift into whichever game your employer happens to be playing.
The rule: name the one game you are playing for the next twelve months, write down what counts as winning, and judge every AI habit by whether it moves that score.
Two things are verifiable. First, the mix of skills employers want is shifting. The World Economic Forum's Future of Jobs Report 2025 reports that employers expect a large share of on-the-job skills to change by 2030, and lists AI and big data, cybersecurity, creative thinking and resilience among the skills on the rise. Second, how people use AI is mixed rather than simple replacement. Anthropic's Economic Index, built from analysis of Claude conversations mapped to US occupational task data, found at its first release that usage leaned somewhat towards augmentation (people working with the model) rather than automation (the model doing the task), and that use clustered in particular kinds of task rather than whole jobs. The Index has been updated since (its page shows a June 2026 update), so check the current figures rather than quoting the early ones.
What neither source tells you is which career strategy suits you. That is a judgement, and it is yours. A note on provenance: the six games below are Juliemango's own framework, written for this guide. We could not verify any outside "six games" model, so treat this as a thinking tool, not research.
Six games, each with a different scoreboard. Pick one primary and, at most, one secondary.
| Game | What you are optimising for | What AI does to it | The usual trap |
|---|---|---|---|
| 1. Craft | Being the person whose judgement in a specialism is trusted | Cheapens first drafts, so your taste and error-spotting become the visible part | Refusing to use AI at all, then being slower than peers |
| 2. Leverage | Output per hour: doing the work of a small team | Directly rewards good briefing, checking and workflow design | Producing more of work nobody asked for |
| 3. Ownership | Equity, a business, or products you keep the upside on | Lowers the cost of building and testing an offer | Mistaking building for demand |
| 4. Trust | Relationships, reputation, being the person people call | Makes generic content cheap, so real people and real track record stand out | Using AI to mass-produce contact and spending the goodwill |
| 5. Security | Stable income, predictable hours, protection from shocks | Makes some routine tasks fragile, so you want to be the person who runs the process | Assuming a steady role is the same as a safe one |
| 6. Range | Learning fast and moving across fields | Shortens the time to a competent first pass in a new area | Staying shallow forever and never earning a real specialism |
Use this in Claude, ChatGPT or any assistant. It is built to question you before it recommends anything.
Role: You are a candid career coach who knows how AI is changing white-collar work but does not make predictions you cannot support.
Context: I will describe my situation. You will help me choose one primary "career game" from this list: Craft, Leverage, Ownership, Trust, Security, Range. Definitions are below.
- Craft: trusted judgement in a specialism
- Leverage: high output per hour using AI and workflow design
- Ownership: equity, a business or products I keep upside on
- Trust: relationships and reputation
- Security: stable, predictable income
- Range: fast learning across fields
My inputs:
- Current role and sector: [ROLE_AND_SECTOR]
- Years of experience: [YEARS]
- Fixed constraints (money, family, location, visa, notice period): [CONSTRAINTS]
- What I want my working life to look like in two years: [TWO_YEAR_PICTURE]
- Tasks in my week that AI already helps with: [AI_TASKS_NOW]
- Tasks I most fear AI will absorb: [FEARED_TASKS]
Steps:
1. If any input above is blank or vague, ask me up to five short questions first and wait. Do not guess.
2. For each of the six games, give one sentence of evidence for and one against fit, using only what I told you.
3. Recommend one primary game and at most one secondary. Explain the choice in under 120 words.
4. Write a one-line scoreboard I could check in three months, using a measurable thing I control.
5. Name three AI habits that serve the scoreboard and three things I should deliberately stop or ignore.
6. Suggest what the next game could be and what would trigger the switch.
Format: a six-row table, then the recommendation, the scoreboard, the habits list, and the next-game note. Plain British English.
Guardrails: Do not invent statistics, salaries or job-market claims. If you mention a trend, say it is general and I should check it. Do not promise outcomes. Note where a professional (financial adviser, employment lawyer) would be a better source.
Self-check before answering: confirm that every claim about me came from my inputs, that the scoreboard is measurable, and that you named something to stop.
Fill in the six square-bracket fields; leave a field as "unsure" if you are, and let the assistant ask.
Imagine a mid-career account manager at a regional marketing agency, with a mortgage and two years of strong client relationships. Her instinct is Ownership: going freelance with AI tools. The card pushes back. Her constraint is income stability, and her real evidence sits in Trust (clients ask for her by name). She picks Trust as primary and Leverage as secondary.
Scoreboard: "Three clients would follow me or give a reference by March." AI habits: use it for meeting prep and follow-up drafts, so she has more time on the phone; keep one reusable brief template. What she stops: automating her client emails into sameness, and buying a fourth tool. At the three-month check she reviews the scoreboard. Next game, if the agency lays people off: Ownership, with those clients as the first customers. That is the card working as intended: a decision, an order, and a trigger.