From Knowledge to Purpose: what happens to human value when intelligence is cheap
Dixon, R.J., 2026d. From Knowledge to Purpose: The Shifting Bottleneck of Value Creation and Five Domains of Human Value as We Approach AGI. SSRN Working Paper, 33 pages, posted 23 March 2026. Available at dx.doi.org/10.2139/ssrn.6396778.
In brief. The second paper in my SSRN series, and the academic version of a LinkedIn article I published in February (Dixon, 2026a). The claim in one sentence: economic history is a story of shifting bottlenecks, generative AI is making cognitive execution abundant, and so the scarce resource that determines where value concentrates is moving from human intelligence to purpose and judgement.
Why I wrote it
The paper starts with a conversation I had with a senior partner in a professional services firm. She had spent twenty-five years getting good at the bit the machine now does in four seconds, and she asked me what exactly she was for now. Her job wasn't at risk; her firm was busier than ever. The public debate, "will AI take my job?", doesn't reach her case. It produces estimates that swing from 47 per cent of US employment at high risk (Frey and Osborne, 2017) to 9 per cent of jobs in OECD countries when you count tasks rather than titles (Arntz, Gregory and Zierahn, 2016). Autor's point that the unit of analysis should be tasks, not jobs, has held up well (Autor, 2015). But none of it answers her question. I wrote the paper to answer it properly, with the model and the propositions the LinkedIn article didn't have room for.
What it argues
Four points carry the paper.
First, economic history is a sequence of bottleneck shifts. In agrarian economies the constraint was land and labour. In industrial economies it was capital and machinery. In the knowledge economy it has been human intelligence: the ability to analyse, synthesise, advise, and decide. Each transition made the previous scarce resource abundant and moved the constraint. Mechanisation made physical labour abundant. Information technology made data processing abundant. Generative AI is making cognitive execution abundant. That's the Theory of Constraints applied to economic eras (Goldratt, 1990): address the binding constraint and a new one appears somewhere else.
Second, this shift is different in kind. In the first three eras, value came from execution: physical, then mechanical, then cognitive. For the first time, execution stops being the primary source of value at all. Value migrates from the doing to the deciding. An economy that valued execution rewarded speed, volume, and efficiency. An economy that values purpose and judgement will reward clarity of thought, ethical accountability, strategic imagination, and the courage to choose wisely under uncertainty.
Third, AI unbundles the knowledge worker. Before generative AI, most knowledge workers were what the paper calls integrated PEJ practitioners: they bundled Purpose, Execution, and Judgement in one identity, because cognitive execution was scarce and couldn't be separated from the direction that gave it meaning. Generative AI dissolves the glue. Execution automates. Purpose is exposed as a distinct and previously undervalued activity, and the cost of solving the wrong problem escalates because AI executes a flawed purpose with extraordinary speed. Judgement becomes the primary value filter, because when quantity is free, value flows to selection (Dixon, 2026b). And the professional's role changes from executor to director: from producing outputs to deciding what ships. The framework underneath this is the one I set out in the first paper: Purpose, Execution, Judgement (PEJ) proposes that every productive human-AI activity comprises three domains, Purpose (why are we doing this?), Execution (how will we do it?), and Judgement (which output is the one that matters?).
Fourth, purpose and judgement stay human by choice, not by technical limitation. The paper is explicit about this assumption. Machines may well approximate both. Ceding them would be an irreversible transfer of agency from human to machine, an argument I first made in the Playbook (Dixon, 2025d), and the decision to keep human primacy over purpose and judgement is itself a purpose-setting decision, arguably the most consequential one of the era.
Five domains of human value
The paper identifies five domains of human economic value that endure as we approach AGI. Purpose-setting and judgement follow from the framework and from the agency-retention assumption. The bit I think is new is the argument about the next two. Empathy and authenticity are different: their value doesn't reside in the behaviour, which a machine may replicate, but in the knowledge that a fellow human being chose to care, or chose to create. An AI can produce a ceramic mug indistinguishable from a hand-thrown one. The value of the hand-thrown mug is knowing someone chose the clay, felt it turn, and decided when it was done. That knowledge is the product. Authenticity, the paper argues, is an emerging economic category, analogous to organic or fair-trade certification. The fifth domain, continuous learning, isn't one among five; it's the engine that powers the other four. In a purpose-based economy, learning is not preparation for work. Learning is work.
What it means in practice
For leaders, three things to start: audit every AI initiative through a PEJ lens (who sets the purpose, who exercises judgement, who is accountable?); treat AI literacy as a universal, continuous capability rather than a specialist, one-off skill; and redesign performance management around purpose, judgement, and empathy rather than volume of output. Three things to stop: measuring productivity by execution volume; deploying AI without a named human accountable for its outputs; and treating learning as a cost centre. If AI governance sits with the CIO because AI is a technology matter, the paper's argument is that it belongs at board level, because the bottleneck has moved from execution to purpose.
For educators, rebalance curricula across all three PEJ domains and assess the quality of purpose-setting, the sophistication of AI collaboration, and the rigour of judgement, rather than the output alone. For policymakers, universal AI literacy, social protection built for transitions between eras, and regulatory frameworks for human-made certification and AI accountability.
The paper presents no new data of its own. The empirical evidence it reviews is other people's, and the limitations section says as much: the model is conceptual and the propositions need testing. The evidence includes the 14 per cent productivity gain for customer support agents, concentrated among novices (Brynjolfsson, Li and Raymond, 2023), and the jagged frontier, where consultants using GPT-4 were faster and better inside the model's capability boundary and worse outside it (Dell'Acqua et al., 2023). The six propositions are offered for testing. The one I most want tested is the last: that this is value migration, not value destruction. The aggregate demand for human contribution persists; it concentrates in different activities. Intelligence is becoming free. The question that remains is what we choose to do with it, and who gets to decide.
How to cite it
Dixon, R.J., 2026d. From Knowledge to Purpose: The Shifting Bottleneck of Value Creation and Five Domains of Human Value as We Approach AGI. SSRN Working Paper. Available at: https://dx.doi.org/10.2139/ssrn.6396778
References
Arntz, M., Gregory, T. and Zierahn, U., 2016. The Risk of Automation for Jobs in OECD Countries: A Comparative Analysis. OECD Social, Employment and Migration Working Papers No. 189. Paris: OECD Publishing. Available at: https://doi.org/10.1787/5jlz9h56dvq7-en.
Autor, D.H., 2015. Why are there still so many jobs? The history and future of workplace automation. Journal of Economic Perspectives, 29(3), pp.3–30. Available at: https://doi.org/10.1257/jep.29.3.3.
Brynjolfsson, E., Li, D. and Raymond, L.R., 2023. Generative AI at Work. NBER Working Paper 31161, April. Available at: https://www.nber.org/papers/w31161 [Accessed 4 September 2026].
Dell'Acqua, F., McFowland, E., Mollick, E.R., Lifshitz-Assaf, H., Kellogg, K., Rajendran, S., Krayer, L., Candelon, F. and Lakhani, K.R., 2023. Navigating the Jagged Technological Frontier: Field Experimental Evidence of the Effects of AI on Knowledge Worker Productivity and Quality. Harvard Business School Working Paper 24-013, 15 September. Available at: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4573321 [Accessed 4 September 2026].
Dixon, R.J., 2025. The AI Transformation Playbook: my first book, and why I wrote it. [online] robdixon.ai, 19 December. Available at: https://www.robdixon.ai/post/the-ai-transformation-playbook [Accessed 4 September 2026].
Dixon, R.J., 2025d. The AI Transformation Playbook: A Step-by-Step Guide to Becoming an AI-Enabled Organisation. Manchester: Dixon AI Press. Available at: https://www.amazon.co.uk/dp/1919400508.
Dixon, R.J., 2026a. From Knowledge to Purpose: What Happens to Jobs as We Approach AGI. [online] LinkedIn, 8 February. Available at: https://www.linkedin.com/pulse/from-knowledge-purpose-what-happens-jobs-we-approach-agi-rob-dixon-vnqze/ [Accessed 4 September 2026].
Dixon, R.J., 2026b. Purpose, Execution, Judgement: A Framework for Human-AI Collaboration and Organisational AI Capability in the Post-ChatGPT Era. SSRN Working Paper. Available at: https://dx.doi.org/10.2139/ssrn.6282200 [Accessed 4 September 2026].
Dixon, R.J., 2026d. From Knowledge to Purpose: The Shifting Bottleneck of Value Creation and Five Domains of Human Value as We Approach AGI. SSRN Working Paper. Available at: https://dx.doi.org/10.2139/ssrn.6396778 [Accessed 4 September 2026].
Dixon, R.J., 2026. Purpose, Execution, Judgement: the paper behind the framework. [online] robdixon.ai, 10 March. Available at: https://www.robdixon.ai/post/purpose-execution-judgement-pej-framework-paper [Accessed 4 September 2026].
Frey, C.B. and Osborne, M.A., 2017. The future of employment: how susceptible are jobs to computerisation? Technological Forecasting and Social Change, 114, pp.254–280. Available at: https://doi.org/10.1016/j.techfore.2016.08.019.
Goldratt, E.M., 1990. Theory of Constraints. Great Barrington, MA: North River Press. Available at: https://northriverpress.com/the-theory-of-constraints/.
Further reading. Added September 2026: later work that builds on this paper.
Dixon, R.J., 2026e. AI Innovation Architecture: Nested Purpose, Closed-Loop Dynamics, and the Chief Agency Officer. SSRN Working Paper. Available at: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=6546178 [Accessed 4 September 2026].
Dixon, R.J., 2026. AI Innovation Architecture: who sets purpose when your systems become agents? [online] robdixon.ai, 23 April. Available at: https://www.robdixon.ai/post/ai-innovation-architecture-chief-agency-officer [Accessed 4 September 2026].
Dixon, R.J., 2026. The CEO's Guide to AI: what I told a room full of CEOs in Chicago. [online] robdixon.ai, 3 September. Available at: https://www.robdixon.ai/post/ceos-guide-to-ai-chicago-keynote [Accessed 4 September 2026].
Rob Dixon is Founder and CEO of Dixon AI Ltd and lectures at Manchester Metropolitan University. He writes and speaks on AI capability, strategy, and human-AI collaboration. Bibliography · Google Scholar · ORCID · LinkedIn · dixonai.com
Comments