displacement.ai
Menu
Start
Glossary

Task exposure

The degree to which current AI systems can materially affect a task, such as drafting, summarizing, classifying, searching, coding, routing, or analyzing information.

What this means in practice

Exposure measures reachability, not outcome. A task is exposed when current AI can plausibly do or assist it — drafting a report, summarizing a call, classifying a document. Exposure says nothing by itself about whether the change helps or hurts the worker.

Interpreters and Translators show why the nuance matters. Both work with language, but written translation is far more exposed to machine translation than real-time spoken interpretation, so the two occupations carry different exposure and pressure scores despite sharing a SOC code.

In the data

See this term in the occupation data

These published occupation pages show task exposure in action — with task-level exposure, automation and augmentation scores, wage context, and transition pathways.

SOC 27-3091

Translators

Written translation is the single most machine-exposed language task: neural and AI translation now produce usable drafts instantly. Post-editing, literary and marketing nuance, certified translation, and terminology governance are where the work is moving.

Exposure
86
Automation
60%
Augment
56%
SOC 27-3091

Interpreters

Machine translation handles written text far better than live speech. Real-time spoken and signed interpretation in courts, hospitals, and schools carries latency, accuracy, confidentiality, and liability demands that keep human interpreters preferred.

Exposure
66
Automation
40%
Augment
46%
SOC 15-1252

Software Developers

Code generation changes the junior task bundle, but architecture, debugging, security, product judgment, and system ownership keep the role augmentation-heavy.

Exposure
63
Automation
29%
Augment
74%