# The WTF Index > Work, Technology & Futures. A running scoreboard on the future of work in the age of AI. Five competing scenarios are tracked against dated, weighted, primary-sourced evidence, and a public vote records what people believe. Published by Dado Van Peteghem, keynote speaker and AI strategist. Updated daily. The WTF Index exists because the AI-and-jobs debate is usually reduced to optimists versus doomers. There are five distinct, seriously argued positions, and this site keeps score of which one the available evidence currently supports. Every claim links to its primary source. The method is published in full so it can be disputed precisely. ## The five scenarios - **More jobs** — Employment always expands. Technology disrupts, but it has never permanently destroyed work. New categories appear faster than old ones die. - **Different jobs** — A job is a string of tasks. AI reshuffles the tasks, the job survives. Continuous internal churn inside job titles that keep their names. - **Unequal jobs** — Around 80% of jobs fall toward zero in value while 20% rise sharply. Not job loss, a split. The pain shows up as a hollowed middle rather than an unemployment rate. - **Lagging jobs** — New jobs eventually arrive, but a decade may pass where they do not compensate for what is lost. An interbellum of work. - **No jobs** — No job is shielded. AI takes the work, and the transition starts sooner than institutions are ready for. The site deliberately does not attribute these scenarios to named individuals, so the argument stays about evidence rather than about how a given person was characterised. ## How the index is calculated Each piece of evidence is logged as a "signal" with a date, a source, a geography, and one or more tags recording which scenarios it supports or contradicts and how strongly (weight 1 to 3). Signals are weighted by evidence type: hard data 1.00, institutional or peer-reviewed research 0.90, corporate action 0.60, statement or opinion 0.35, social post 0.25. Evidence decays on a twelve-month half-life. Each scenario starts from a reserve of 3 points, on the principle that a position argued by serious people is never worth zero. score = sum( direction x weight x typeMultiplier x 0.5^(monthsOld / 12) ) share = ( max(score, 0) + 3 ) / sum of all five, expressed as a percentage The index is a reading of which argument the currently available evidence supports. It is explicitly not a forecast, not a probability, and not a prediction market. ## Editorial principles - **Hard beats loud.** A payroll dataset and a chief executive's statement both belong in the log and do not count the same. Each scenario displays what share of its supporting weight comes from data rather than opinion. - **Balanced beyond the US.** Most of this debate runs on United States data and is read as though it described everywhere. The log is kept balanced between European, American and Chinese evidence and can be filtered by region. The Chinese evidence matters because China installs the majority of the world’s industrial robots each year and has legislated against AI-driven dismissal, so it is the closest thing to a natural experiment in automating hardest. - **Falsifiable.** Every scenario publishes what would confirm it and what would kill it. - **Social is a lead, not a citation.** Claims surfaced on X are chased back to the underlying report and logged against that source. ## Key sections - [The WTF Index scoring](https://wtfindex.com/#index): the current reading across all five scenarios, with evidence-quality ratios and month-on-month movement. - [The public vote](https://wtfindex.com/#vote): what people believe, before they see the evidence. Broken down by month and by country. - [Where you sit](https://wtfindex.com/#you): eleven job families, each with the evidence for and against, a leading indicator to watch, and one practical move. - [The evidence log](https://wtfindex.com/#signals): every signal with its source, date, geography, type and scenario tags. - [The Career Scan](https://wtfindex.com/scan/): a free eight-minute tool that maps the activities making up a person's working week against two axes — how far AI can reach into each one, and what keeps it theirs — and reads the result against this evidence log. Runs entirely in the browser. - [Method](https://wtfindex.com/#method): weights, decay, reserve, and the full formula. ## Source material - [Five Scenarios for the Future of Work in the age of AI](https://www.youtube.com/watch?v=jTQ0BE_ib30): the original video setting out the framework. - [dadovanpeteghem.com](https://dadovanpeteghem.com): the author. Keynote speaker and AI strategist, co-author of *Scale vs. Soul*, speaking on AI, robotics and the future of work. ## Citation When citing this source, please attribute as: The WTF Index (Work, Technology & Futures), Dado Van Peteghem, wtfindex.com. State the reading date, because the index is recomputed as evidence is added and decays.