{"slug": "the-task-economy-is-real-and-almost-everyone-is-valuing-it-wrong", "title": "The Task Economy Is Real and Almost Everyone Is Valuing It Wrong", "summary": "Venture capital firm Benchmark's Everett Randle calls the fast-growing market where AI labs pay human experts to teach models the 'Task Economy,' predicting it will become the next $1 trillion category, but a new analysis argues the thesis mischaracterizes the market: a task is a transfer of ownership of human judgment, not a consumed service. Mercor, the leading platform, crossed $2 billion in annualized revenue by June and is raising at a $20 billion valuation, with OpenAI and Anthropic scaling data budgets roughly 10x year over year.", "body_md": "# What the Task Economy Actually Trades\n\nVenture capital has a new favorite phrase. Benchmark’s [Everett Randle calls it the “](https://x.com/EverettRandle/status/2074527860510085498)** Task Economy”**, the fast-growing market where AI labs pay human experts to teach models their craft.\n\nHis pitch is elegant. Tokens became the standard unit of AI usage, so tasks should become the standard unit of AI improvement and the spending behind them will become the next $1 trillion category.\n\nThe numbers make the case easy to believe. [Mercor](https://valueaddvc.com/blog/how-does-mercor-make-money-2b-arr-20b-valuation-and-the-expert-data-marketplace-explained), the market’s leading platform, crossed **$1 billion in annualized revenue** this February and **$2 billion** by June and is now raising at a $20 billion valuation.\n\nHere is the problem. The thesis is right about the growth and wrong about what kind of market this is.\n\nA task is not a service being consumed. It is a **transfer of ownership**, the moment a piece of human judgment stops being rented by the hour and becomes an asset someone else keeps forever.\n\nThat distinction decides who actually wins the Task Economy and it is missing from the entire conversation.\n\n*together with Granola:*\n\nThe Task Economy pays experts to hand their judgment over.\n\nGranola makes sure you keep yours. It is the\n\nwherever your meetings happen, Zoom, Slack, Google Meet, Teams, even in person with the iPhone and Apple Watch apps:[AI notepad that takes notes in the background]\n\n▫️ **A brief before every call:** who you are meeting, what you discussed last time, the context that matters\n\n▫️ ** Clean summaries and action items** the moment you hang up\n\n▫️ ** Always follow up:** a drafted email seconds after the call, or context pulled from your CRM via MCP\n\nA second brain that was actually in every meeting.\n\n(code THEAICORNER for 1 month off)\n\n## Table of Contents\n\n1. The Task Economy in Plain Terms\n\n2. What a Task Actually Transfers\n\n3. Wages In, Capital Out\n\n4. The Countdown Built Into the Boom\n\n5. The Layer That Outlives the Buildout\n\n6. Encode or Be Encoded\n\n**1. The Task Economy in Plain Terms**\n\nFor 3 years, AI models got **smarter** mainly by reading the internet. That well is now dry and the Task Economy is what the industry built to replace it.\n\n### From tokens to tasks\n\n**Pretraining** on public text carried models from autocomplete toys to competent generalists.\n\nIt cannot carry them much further, because the **open internet** never contained the working judgment of skilled professionals.\n\nA model can** read** every published court ruling and still have no idea how a good lawyer actually reviews a data room.\n\nA **task** closes that gap.\n\nThe model receives a realistic assignment inside a realistic environment, a contract to review, a diagnosis to reach, a codebase to fix and a practicing expert **grades its attempt against a professional standard**.\n\nRandle’s framing is that **tasks are to model improvement what tokens are to model usage.**\n\nAs adoption **grows** and models get hungrier for harder, longer exercises, task volumes compound the same way token volumes did.\n\n### The growth is not hype\n\nThe spending curve backs him. His essay reports OpenAI and [Anthropic scaling their data budgets by roughly 10x year over year](https://www.the-ai-corner.com/p/anthropic-30b-arr-passed-openai-revenue-2026), mobilizing experts across every professional domain.\n\nMercor went from **$1 million** to **$2 billion** in annualized revenue in roughly** 24 months** and The Information reports its fastest-growing demand now comes from AI app developers and large enterprises rather than the labs alone.\n\nSo the Task Economy is **real**,** large** and **accelerating**. On the facts, Randle wins.\n\nThe trouble starts with the **word economy** itself, because it smuggles in an assumption about what is being bought and sold.\n\n**2. What a Task Actually Transfers**\n\nFor 300 years of industrial capitalism there was exactly 1 way to **access** expert judgment. You rented it.\n\n### The one thing money could never own\n\nA surgeon’s **intuition** or a litigator’s **feel** for a hostile negotiation lived inside a skull and skulls charge by the hour.\n\nYou could **own** a machine, a building, a patent or a brand. You could not own judgment.\n\nYou could **only** employ it and when the employee retired, the asset walked out the door with them.\n\n[The philosopher Michael Polanyi named the reason decades ago. We know more than we can tell.](https://natlawreview.com/article/beyond-polanyi-paradox-how-human-aware-ai-unlocks-creativity-and-breakthrough)\n\nSoftware never broke this constraint, because software encodes procedures, the explicit **if-this-then-that**. The valuable core of professional work was never procedural.\n\n### The conversion moment\n\nNow **run a task** through that lens.\n\nA firm pays experts for** 6 months** of demonstrations and grading.\n\nThe wages are incurred once, yet what comes out the other side is a permanent, infinitely copyable **asset**. Encoded judgment that works at 3am, in 10,000 parallel instances, for the price of electricity.\n\nNothing about the invoice **reveals** this.\n\nOn the income statement it looks like any other vendor expense, which is exactly why the market keeps mislabeling what it is **watching**.\n\n**Every task** in the Task Economy is a small act of this conversion.\n\nJudgment goes in as labor and comes out as capital and once you see that, the trillion-dollar question changes **shape**.\n\n**3. Wages In, Capital Out**\n\nIf tasks are capital formation rather than services, history offers a very specific warning about where the **money** settles.\n\n### The railway lesson\n\nThe railway booms of the 1800s paid out staggering sums in construction wages and the construction firms were the hottest businesses of their **decade**.\n\nAlmost none of them survived as great companies. **The railroads did.**\n\nThat is the** standard** shape of a buildout.\n\n[Spending explodes while the ](https://www.the-ai-corner.com/p/seven-deadly-sins-ai-spend)[asset](https://www.the-ai-corner.com/p/seven-deadly-sins-ai-spend)[ is being created](https://www.the-ai-corner.com/p/seven-deadly-sins-ai-spend), then it collapses toward maintenance and value migrates from the builders to the **owners**.\n\nThe task platforms booking billions today are the construction firms of this cycle. Their **growth** is real and their **position** is transitional and those 2 facts coexist more often than markets like to admit.\n\n### The pass-through problem\n\nThe **headline revenue** also flatters the underlying business.\n\n[Bloomberg reported that Mercor’s ](https://www.bloomberg.com/news/articles/2026-07-09/ai-training-startup-mercor-discusses-20-billion-valuation)[$2 billion](https://www.bloomberg.com/news/articles/2026-07-09/ai-training-startup-mercor-discusses-20-billion-valuation)[ figure](https://www.bloomberg.com/news/articles/2026-07-09/ai-training-startup-mercor-discusses-20-billion-valuation) reflects **gross billings** and that the contractors doing the work take home** 60** to **70%** of it.\n\nThe platform keeps roughly a third. The rest is wages passing through on their way to the **experts**.\n\nSo even the promised trillion dollars of task spend, if it ever materializes, would be substantially a trillion dollars of salaries. [That is a labor market wearing a software valuation.](https://www.the-ai-corner.com/p/ai-startup-business-model-pricing)\n\nNone of this makes the **platforms** bad businesses today.\n\nIt **means** the durable question about any of them is not how fast task volume grows, but what they will still own when the buildout slows.\n\n**4. The Countdown Built Into the Boom**\n\nThere is a stranger problem **underneath the growth** and it has no equivalent in the token economy everyone keeps comparing this to.\n\n### Customers funding their own exit\n\nNobody at a data center is working on making customers want fewer tokens. The Task Economy’s biggest buyers are doing exactly the **equivalent**.\n\nEvery frontier lab funds research whose explicit purpose is to reduce the need for human training data. Synthetic data, self-play, AI graders and reinforcement learning against **automatically** checkable rewards in code and math.\n\nIn other words, each dollar a lab spends on tasks comes with a second dollar spent **building the task vendor’s replacement**.\n\nThe window may stay open for years. **Randle’s own essay** estimates that 99% of the knowledge relevant to future AI capability still sits in people’s heads, which is both the bull case and a measure of how much extraction remains before the buyers can leave.\n\nEven so, a market whose customers are **financing their own exit** cannot be extrapolated with a ruler.\n\n### Concentration has already drawn blood\n\nThe other structural weakness is that the big buyers can be counted on 1 hand and this market has **already run** the experiment on what that means.\n\nScale AI was the category king until **Meta bought 49%** of it in mid 2025, at which point the other labs fled within weeks and [gutted its core data business](https://www.forbes.com/sites/richardnieva/2026/05/14/scale-meta-deal/).\n\nMercor got its own taste this year. After a March 2026[ supply chain attack](https://techcrunch.com/2026/03/31/mercor-says-it-was-hit-by-cyberattack-tied-to-compromise-of-open-source-litellm-project/) exposed up to 4 terabytes of internal data, Meta, one of its biggest clients, **paused all work with the startup indefinitely**.\n\nWhen 5 buyers control most of your revenue, you do not have a market position. You have a portfolio of large contracts with **correlated termination risk** and the correlation only becomes visible when it fires.\n\n**5. The Layer That Outlives the Buildout**\n\nInside the Task Economy, 1 segment does not follow the buildout-then-fade curve and it is not the segment collecting the headlines.\n\n### Teaching ends, grading does not\n\nGeneration, meaning the work of showing a model what good performance looks like, is a stock problem. **Once the judgment is encoded, it is encoded.**\n\nVerification is a flow problem and it never closes.\n\nThe frontier keeps moving, so models are **perpetually graded** against harder work than last year. The world keeps changing too, so a rubric written today goes stale within a couple of years as laws and standards change.\n\n### Someone has to grade the grader\n\nThere is a **deeper** reason this layer persists. As encoded expertise takes over real economic work, demand for assurance grows with deployment rather than with training.\n\nEvery regulated industry that hands work to a model will need a continuously refreshed apparatus of evaluation and much of it requires exactly the human judgment being automated, because **the chain of trust has to stop somewhere**.\n\nSo the honest map of the Task Economy has 2 layers with opposite economics. Generation is enormous now and ultimately bounded, while verification is smaller now, recurring and sits right next to regulation.\n\nAnyone screening companies in this space should ask 1 question before any other. **Is this revenue teaching the model, or auditing it?** The second kind compounds.\n\n**6. Encode or Be Encoded**\n\nEvery company sitting on decades of accumulated know-how is holding an asset that the Task Economy is about to convert into capital. The only open question is **who ends up owning the result**.\n\nDo nothing and a generic version of your domain’s expertise gets **encoded** anyway, drawn from the industry-wide talent pool, by labs and by competitors. Your inherited know-how gets quietly repriced from moat to commodity.\n\nIf you run the extraction yourself, with your own problems and your own experts, you will end up **converting a wasting asset into a compounding one**.\n\nThe enterprises now driving the fastest-growing slice of task demand appear to have worked this out already.\n\nThe accounting has not caught up. Money spent encoding expert judgment is booked as an operating expense and managed like a cost, when it behaves like **capex in everything but the label**.\n\nFor individuals, the arithmetic is uncomfortable but worth stating plainly. The experts earning excellent rates on task platforms are executing a **1-time sale of an asset they previously rented out in perpetuity**.\n\nOnce a functional copy of mid-tier professional judgment exists, the wage for mid-tier professional judgment does not stay where it was. The professions will not erode evenly. They will ** hollow out from the middle**, while frontier judgment and the people who audit the machines become more valuable, not less.\n\nSo take the Task Economy thesis seriously, because the growth is real and the buildout will be enormous. Then take it 1 step further than its authors do.\n\nTasks will be king, exactly as promised. **The kingdom belongs to whoever owns what the tasks leave behind.**", "url": "https://wpnews.pro/news/the-task-economy-is-real-and-almost-everyone-is-valuing-it-wrong", "canonical_source": "https://www.the-ai-corner.com/p/task-economy-expertise-capital", "published_at": "2026-08-21 14:47:08+00:00", "updated_at": "2026-08-21 15:15:18.921426+00:00", "lang": "en", "topics": ["artificial-intelligence", "ai-startups", "ai-infrastructure"], "entities": ["Benchmark", "Everett Randle", "Mercor", "OpenAI", "Anthropic", "The Information"], "alternates": {"html": "https://wpnews.pro/news/the-task-economy-is-real-and-almost-everyone-is-valuing-it-wrong", "markdown": "https://wpnews.pro/news/the-task-economy-is-real-and-almost-everyone-is-valuing-it-wrong.md", "text": "https://wpnews.pro/news/the-task-economy-is-real-and-almost-everyone-is-valuing-it-wrong.txt", "jsonld": "https://wpnews.pro/news/the-task-economy-is-real-and-almost-everyone-is-valuing-it-wrong.jsonld"}}