---
id: automation-of-cognitive-work
name: Automation of Cognitive Work
force: optimisation-of-everything
enablerSet: complete
url: https://explorer.talin.digital/m/automation-of-cognitive-work/
author: Benjamin Talin
citation: Talin, B. (2026). The Future Explorer. https://explorer.talin.digital (dataset v0.4.0, 2026-11-16)
license: CC-BY-NC-4.0
---

# 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.

## Direction
tasks that can be specified in language trend toward machine execution

## Momentum
Capability improves continuously and cost per unit falls; neither depends on any institution agreeing to it.

## Gradient
gradient 2.7/5 (a sticky rubber band: pulled toward the weakest enabler)
minimum 2/5 — status: tie
**Shared weakest link:** Data & Datafication, Digital Transformation, Managerial Capability, Neuromorphic & Alternative Compute, Regulatory Clarity, Workforce Trust

## Enablers
- Brain-Computer Interfaces — 1/5, marginal (marginal: excluded from the minimum)
- Quantum Computing — 1/5, marginal (marginal: excluded from the minimum)
- Data & Datafication — 2/5, moderate
- Digital Transformation — 2/5, large
- Managerial Capability — 2/5, large
- Neuromorphic & Alternative Compute — 2/5, moderate
- Regulatory Clarity — 2/5, large
- Workforce Trust — 2/5, large
- Adoption Appetite — 3/5, large
- AI & Machine Learning — 3/5, large
- Autonomous Vehicles — 3/5, marginal (marginal: excluded from the minimum)
- Capital Availability — 3/5, large
- Energy Price & Availability — 3/5, moderate
- Deep Learning & Neural Networks — 4/5, large
- Large Language Models — 4/5, large
- Advanced Chips & Semiconductors — 5/5, large
- Automation — 5/5, large
- Cloud & Datacentre Infrastructure — 5/5, large

## Clock reading
- **individuals:** Weeks. The fastest adoption of any technology measured, because it rides infrastructure people already own.
- **companies:** One to five years to adopt, five to forty to reorganise around it. The gap between those is the chapter's subject.
- **governments:** Three to thirty years. Labour, liability and tax law all assume a human performing the task.
- **biosphere:** Indirect, through the energy and water cost of the infrastructure build.

## Falsifier
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.

## Couples
- amplifies → The Application Gap: Capability arriving faster than organisations can absorb it widens the gap by definition.
- amplifies → Epistemic Fragmentation: Cheap generated text raises the volume of plausible claims faster than verification capacity grows.
- amplifies → The Relationship Recession: Machine interlocutors substitute for the low-stakes human contact that seeds relationships. ⚠ Correlational only; the direction is unestablished.
- amplifies → Institutional Trust Decline: Decisions made by systems nobody can interrogate erode the legitimacy of the institution making them.
- amplifies → Polarisation and Radicalisation: Clerical displacement concentrates in specific regions, and concentrated displacement turns into political disengagement and radicalisation.
- amplifies → Digital Sovereignty: Governments running on AI they cannot supply buy intelligence as a revocable subscription, turning capability into a sovereignty question.
