Cognitive automation
Automation of Cognitive Work
'Cognitive work' is thinking work: writing, summarising, coding, answering customers. This megatrend is machines taking on more of any task that can be described in words. When customer-support staff were given an AI assistant, they solved about 15% more problems an hour on average, and beginners improved by about 30%. The software keeps getting better and cheaper; what slows it is people and organisations learning to trust it and work with it.
Tasks that can be specified in language trend toward machine execution.
Gradient 2.7 / 5 - weighed down most by the weakest enabler.
The average of these enablers is 3.1. The ramp runs at 2.7.
6 enablers have to mature together before this ramp steepens.
In the network
- Data & Dataficationshared weakest link (with 5 others)
- Digital Transformationshared weakest link (with 5 others)
- Managerial Capabilityshared weakest link (with 5 others)
- Neuromorphic Computeshared weakest link (with 5 others)
- Regulatory Clarityshared weakest link (with 5 others)
- Workforce Trustshared weakest link (with 5 others)
- Adoption Appetite
- AI & Machine Learning
- Capital Availability
- Energy Price
- Deep Learning
- Large Language Models
- Advanced Chips
- Automation
- Cloud Infrastructure
- Brain Interfacesmarginal: not counted as a weakest link
- Quantum Computingmarginal: not counted as a weakest link
- Autonomous Vehiclesmarginal: not counted as a weakest link
Also shaped by The Closing of the Planetary Envelope (1 shared enabler) · The Collapse of Shared Reality (4 shared enablers, 3 couplings) · The Decoupling of Physical and Digital Life (4 shared enablers) · The Inversion of the Population Pyramid (3 shared enablers, 1 coupling) · The Rebuild of the Physical Base (10 shared enablers, 1 coupling) · The Redistribution of Power (9 shared enablers, 1 coupling)
Enablers
| Enabler | Family | Weight | Maturity | Role |
|---|---|---|---|---|
| Data & Datafication | technological | moderate | 2/5 | shared weakest link (with 5 others) |
| Digital Transformation | technological | large | 2/5 | shared weakest link (with 5 others) |
| Managerial Capability | human | large | 2/5 | shared weakest link (with 5 others) |
| Neuromorphic & Alternative Compute | technological | moderate | 2/5 | shared weakest link (with 5 others) |
| Regulatory Clarity | structural | large | 2/5 | shared weakest link (with 5 others) |
| Workforce Trust | human | large | 2/5 | shared weakest link (with 5 others) |
| Adoption Appetite | human | large | 3/5 | - |
| AI & Machine Learning | technological | large | 3/5 | - |
| Capital Availability | structural | large | 3/5 | - |
| Energy Price & Availability | structural | moderate | 3/5 | - |
| Deep Learning & Neural Networks | technological | large | 4/5 | - |
| Large Language Models | technological | large | 4/5 | - |
| Advanced Chips & Semiconductors | technological | large | 5/5 | - |
| Automation | technological | large | 5/5 | - |
| Cloud & Datacentre Infrastructure | technological | large | 5/5 | - |
| Brain-Computer Interfaces | technological | marginal | 1/5 | marginal: not counted as a weakest link |
| Quantum Computing | technological | marginal | 1/5 | marginal: not counted as a weakest link |
| Autonomous Vehicles | technological | marginal | 3/5 | marginal: not counted as a weakest link |
- Data & Dataficationshared weakest link (with 5 others)2/5
- Digital Transformationshared weakest link (with 5 others)2/5
- Managerial Capabilityshared weakest link (with 5 others)2/5
- Neuromorphic & Alternative Computeshared weakest link (with 5 others)2/5
- Regulatory Clarityshared weakest link (with 5 others)2/5
- Workforce Trustshared weakest link (with 5 others)2/5
- Adoption Appetitelarge · human3/5
- AI & Machine Learninglarge · technological3/5
- Capital Availabilitylarge · structural3/5
- Energy Price & Availabilitymoderate · structural3/5
- Deep Learning & Neural Networkslarge · technological4/5
- Large Language Modelslarge · technological4/5
- Advanced Chips & Semiconductorslarge · technological5/5
- Automationlarge · technological5/5
- Cloud & Datacentre Infrastructurelarge · technological5/5
- Brain-Computer Interfacesmarginal: not counted as a weakest link1/5
- Quantum Computingmarginal: not counted as a weakest link1/5
- Autonomous Vehiclesmarginal: not counted as a weakest link3/5
The gradient behaves like a sticky rubber band: it is pulled hardest toward the weakest enabler. Marginal links are shown but do not pull. Everything above 2 is strength this ramp can use only in part until the weakest link moves.
What this pushes on, and what pushes on it
Pushes on
- amplifiesThe Application GapCapability arriving faster than organisations can absorb it widens the gap by definition.
- amplifiesEpistemic FragmentationCheap generated text raises the volume of plausible claims faster than verification capacity grows.
- amplifiesThe Relationship RecessionMachine interlocutors substitute for the low-stakes human contact that seeds relationships. ⚠ Correlational only; the direction is unestablished.
- amplifiesInstitutional Trust DeclineDecisions made by systems nobody can interrogate erode the legitimacy of the institution making them.
- amplifiesPolarisation and RadicalisationClerical displacement concentrates in specific regions, and concentrated displacement turns into political disengagement and radicalisation.
- amplifiesDigital SovereigntyGovernments running on AI they cannot supply buy intelligence as a revocable subscription, turning capability into a sovereignty question.
Bound by
Workforce trust, managerial capability, regulation
NOT compute. Human and institutional absorption.
Momentum. Capability improves continuously and cost per unit falls; neither depends on any institution agreeing to it.
The four clocks
Four systems absorb this at four speeds. The distance between them is the adaptation gap.
Individuals
monthsWeeks. The fastest adoption of any technology measured, because it rides infrastructure people already own.
Companies
yearsOne to five years to adopt, five to forty to reorganise around it. The gap between those is the chapter's subject.
Governments
decadesThree to thirty years. Labour, liability and tax law all assume a human performing the task.
The biosphere
not ours to setIndirect, through the energy and water cost of the infrastructure build.
Impacts by domain
Social & demographic
Freelance writers, coders and other online knowledge workers saw monthly job postings fall 2.0% and earnings fall 5.2% in the months after ChatGPT's public release, relative to freelancers in unexposed trades (Hui, Reshef & Zhou, Organization Science, November 2024).
Environmental
Generative AI's energy draw depends heavily on the task: 1,000 text queries to text-generation models of at most 6.7B parameters burn about 0.047 kWh, but 1,000 image generations burn 2.9 kWh (Luccioni, Jernite & Strubell, ACM FAccT, June 2024).
Economic
Give customer-support staff a generative AI assistant and resolved issues per hour rise about 15% on average, with the gain concentrated among less experienced and lower-skilled workers, while the most experienced see small gains in speed and small declines in quality (Brynjolfsson, Li & Raymond, Quarterly Journal of Economics, 2025).
Geopolitical
Washington has banned the export to China of logic chips with interconnect bandwidth of 600 GB/s or more and the advanced lithography tools used to make them (US Bureau of Industry and Security, Federal Register, October 2022 and October 2023).
What would prove this wrong
Sustained total factor productivity growth attributable to AI above the pre-AI trend, which would show absorption is no longer the constraint. Or the reverse: capability plateauing while the application gap stays open.
Talin, B. (2026). Automation of Cognitive Work. The Future Explorer, dataset v0.4.0 (2026-11-16). https://explorer.talin.digital/m/automation-of-cognitive-work/