5D Library › One-Person Enterprise Automation, Avatar Livestream Funnels and Task-Exposure Arbitrage · 中文

Chapter 5. Task Exposure and the Acemoglu-Autor Task Model

Jobs are bundles of tasks

Labour economists stopped treating an occupation as one thing when they began counting its tasks: routine tasks are easier to automate than non-routine ones (Autor et al. 2003), and a technology affects wages through the tasks it takes over and the tasks it leaves or creates (Acemoglu and Autor 2011). Exposure measures for large language models follow the same logic, scoring each task by whether such a model could cut its time substantially (Eloundou et al. 2024). Below, a marketing manager’s 40-hour week is split into eight tasks with illustrative exposures, and arrows show which task’s output feeds which.

Calculation

# Chapter 5. p09: An occupation as a graph of tasks. A marketing
# manager's 40-hour week is split into eight tasks, each with an
# illustrative exposure e (the share of its time AI tools can take over)
# and with arrows "this task's output feeds that one". The exposure score
# is the time-weighted mean of e; the freed hours are sum(hours * e).
# Judgement tasks (budget, strategy, relationships) take a larger share
# of the remaining week. A task's "downstream hours" are the hours of
# all tasks that depend on it, directly or through others.
TASKS = {   # name: (hours a week, exposure)
    "copy and posts": (8, 0.7), "visual briefs": (4, 0.5),
    "reporting": (5, 0.8), "research summaries": (4, 0.7),
    "coordination": (6, 0.4), "budget and vendors": (3, 0.2),
    "strategy": (4, 0.1), "relationships": (6, 0.1),
}
FEEDS = {"strategy": ["copy and posts", "visual briefs"],
         "research summaries": ["strategy"],
         "relationships": ["strategy", "budget and vendors"],
         "reporting": ["budget and vendors", "strategy"]}
JUDGEMENT = ("budget and vendors", "strategy", "relationships")

def downstream(t, seen=None):
    seen = set() if seen is None else seen
    for u in FEEDS.get(t, []):
        if u not in seen:
            seen.add(u)
            downstream(u, seen)
    return seen

if __name__ == "__main__":
    total = sum(h for h, _ in TASKS.values())
    freed = sum(h * e for h, e in TASKS.values())
    print("   exposure score {:.2f}; hours freed {:.1f} of {}".format(
        freed / total, freed, total))
    j0 = sum(TASKS[t][0] for t in JUDGEMENT)
    j1 = sum(TASKS[t][0] * (1 - TASKS[t][1]) for t in JUDGEMENT)
    print("   judgement share of the week: {:.3f} before, {:.3f} after"
          .format(j0 / total, j1 / (total - freed)))
    for t in sorted(TASKS, key=lambda x: -sum(TASKS[u][0]
                                               for u in downstream(x))):
        d = sum(TASKS[u][0] for u in downstream(t))
        if d:
            print("   {:19s} e={:.1f}: {:2d} downstream hours".format(
                t, TASKS[t][1], d))
#    exposure score 0.46; hours freed 18.4 of 40
#    judgement share of the week: 0.325 before, 0.528 after
#    reporting           e=0.8: 19 downstream hours
#    relationships       e=0.1: 19 downstream hours
#    research summaries  e=0.7: 16 downstream hours
#    strategy            e=0.1: 12 downstream hours

The exposure score is 0.46: 18.4 of 40 hours could be taken over. The tasks left are mostly judgement: budget and vendors, strategy and relationships rise from 0.325 of the week to 0.528 of what remains. The graph adds what a single score hides. Reporting is highly exposed, yet 19 hours of work downstream depend on it, so errors in automated reports spread. Relationships are barely exposed and also feed 19 downstream hours: whoever owns them owns the inputs to strategy. A person planning a move should look for tasks with low exposure and many downstream hours; they are where complementarity, and pay, will concentrate.

Displacement against productivity

In the task model, raising the share II of tasks done by machines has two effects on wages: people lose those tasks (displacement), and output rises (productivity). Acemoglu and Restrepo stress that automation which replaces people while adding little productivity, “so-so automation”, lowers wages, and that new tasks for people are what restore them (Acemoglu and Restrepo 2019). With a Cobb-Douglas task aggregate the wage is w=(1−I)Yw = (1-I)Y.

Calculation

# Chapter 5. p10: Displacement against productivity in a task model.
# Tasks fill [0, 1]; those below I are done by AI, the rest by people.
# With a Cobb-Douglas task aggregate and AI capital K = 1, labour L = 1,
# ln Y = I ln(A / I) + (1 - I) ln(1 / (1 - I)), and the wage is
# w = (1 - I) Y. A is how much better AI does an automated task than the
# capital it replaces. When I rises, people lose tasks (displacement)
# while output rises (productivity); the wage rises only if A is large.
import math

def wage(I, A):
    lny = I * math.log(A / I) + (1 - I) * math.log(1 / (1 - I))
    return (1 - I) * math.exp(lny)

if __name__ == "__main__":
    Is = (0.3, 0.4, 0.5, 0.6, 0.7)
    print("   automated share I: " + "".join("{:8.1f}".format(I)
                                            for I in Is))
    for A in (1.0, 1.5, 2.0, 4.0):
        base = wage(0.3, A)
        print("   A = {:.1f}, wage index:".format(A) +
              "".join("{:8.3f}".format(wage(I, A) / base) for I in Is))
#    automated share I:      0.3     0.4     0.5     0.6     0.7
#    A = 1.0, wage index:   1.000   0.912   0.776   0.608   0.429
#    A = 1.5, wage index:   1.000   0.950   0.841   0.687   0.504
#    A = 2.0, wage index:   1.000   0.978   0.891   0.749   0.566
#    A = 4.0, wage index:   1.000   1.048   1.023   0.922   0.746

When AI does automated tasks only as well as the capital it replaces (A=1.0A = 1.0), raising the automated share from 0.3 to 0.5 lowers the wage index to 0.776, and to 0.7 lowers it to 0.429. With A=4.0A = 4.0, the first step raises wages (1.048) before displacement wins again (0.746 at 0.7). In this simple model, without new tasks, wages keep falling once automation goes far enough; what reverses that in history is the creation of new kinds of work, which the model leaves out. For one person the lesson is narrower: income is safest in tasks that are hard to automate and that automated tasks feed.

Note

Exposure is not loss. An exposure score says how much of a job’s time a technology could take over, not how much will be taken, how fast, or whether the job then disappears or changes. Using it to rank one’s own tasks is reasonable; using it to forecast employment is not.

Summary

A marketing manager’s week in the illustrative example has an exposure score of 0.46, and judgement tasks rise from 0.325 to 0.528 of the remaining week. In the task model, “so-so” automation lowers wages as it spreads; only large productivity gains, or new tasks, offset displacement.

References

Acemoglu, Daron, and David Autor. 2011. “Skills, Tasks and Technologies: Implications for Employment and Earnings.” In Handbook of Labor Economics, edited by David Card and Orley Ashenfelter, vol. 4B. Elsevier.
Acemoglu, Daron, and Pascual Restrepo. 2019. “Automation and New Tasks: How Technology Displaces and Reinstates Labor.” Journal of Economic Perspectives 33 (2): 3 to 30.
Autor, David H., Frank Levy, and Richard J. Murnane. 2003. “The Skill Content of Recent Technological Change: An Empirical Exploration.” Quarterly Journal of Economics 118 (4): 1279 to 1333.
Eloundou, Tyna, Sam Manning, Pamela Mishkin, and Daniel Rock. 2024. “GPTs Are GPTs: Labor Market Impact Potential of LLMs.” Science 384 (6702): 1306 to 1308.

Fan, Y. (2026). One-Person Enterprise Automation, Avatar Livestream Funnels and Task-Exposure Arbitrage: Coasean Transaction-Cost Collapse, Cyclic Script Retention and High-Margin Solopreneur Portfolios [一人公司自动化架构、数字人直播漏斗与任务级人工智能套利] (Version 1.0.0). 5D Library. https://doi.org/10.5281/zenodo.24090235 · Licensed under Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0).