Augmentation vs automation share (Anthropic Economic Index)
Share of classified conversations in the augmentation bucket (learning, task iteration, validation) versus automation (directive, feedback loop), computed from Anthropic's Economic Index open dataset. From its June 2026 release the dataset samples Claude chat and Cowork conversations every day and reports them by calendar month, where earlier releases drew one seven-day sample of Claude.ai conversations, so the monthly readings are a series of their own and the reading is the newest month; the weekly samples, the blog-reported shares and the share of occupations with at least a quarter of their tasks used are shown alongside.
Reason through thisThe reading
Augmentation has stayed just above half of classified use in every release. The monthly series reads a little lower than the two weekly samples, but it is sampled differently (chat and Cowork conversations every day of a month, rather than Claude.ai conversations in one week), so the step between them cannot be read as a trend; the monthly series has two months so far.
2 readings drawn from anthropic_ei.global.augmentation_share.m, one of 7 series listed under How it is measured; 11 observations in the evidence.
Source: Anthropic Economic Index (open dataset and reports). Anthropic Economic Index, open dataset on Hugging Face (CC BY 4.0) and the 2026 reports.
Why it matters
The paper's bet is that AI diffuses as a tool people use, not a replacement; this is the only large-scale usage classification that speaks to it directly.
What cuts against it
One provider's consumer plans (chat and, from the June 2026 release, Cowork), and the two weekly samples cover different plans (Free and Pro in January 2026, Max added in March); classification is model-based; API usage (where automation concentrates) is reported separately and skews the other way.
How it is measured
- series
anthropic_ei.global.augmentation_share.m - series
anthropic_ei.global.automation_share.m - series
anthropic_ei.global.augmentation_share.pt - series
anthropic_ei.global.automation_share.pt - series
anthropic_ei.global.augmentation_share_reported.pt - series
anthropic_ei.global.automation_share_reported.pt - series
anthropic_ei.global.occupations_25pct_tasks_share.pt - source Anthropic Economic Index (open dataset and reports) · default tier 2 · CC BY 4.0 (dataset)
- proxy behaviour, product · cadence quarterly · valve product to adoption
Normal = augmentation at least half of classified use; fast = automation clearly dominant (augmentation under 40%). Between is `emerging`.
Applied to anthropic_ei.global.augmentation_share.m.
Why this status, and its history
The Economic Index's monthly readings now sit on a series of their own: from its June 2026 release the dataset samples Claude chat and Cowork conversations every day and reports them by calendar month, where earlier releases drew one seven-day sample of Claude.ai conversations, so the two are no longer drawn as one line. The reading is unchanged: augmentation is 51.4% of classified conversations in May 2026, within 5% of the normal band's edge at 50%, so the band-edge rule still holds the status at emerging.
- 2026-09-23emerging to emergingconf 60 → 60 · Claude (model)
The Economic Index's monthly readings now sit on a series of their own: from its June 2026 release the dataset samples Claude chat and Cowork conversations every day and reports them by calendar month, where earlier releases drew one seven-day sample of Claude.ai conversations, so the two are no longer drawn as one line. The reading is unchanged: augmentation is 51.4% of classified conversations in May 2026, within 5% of the normal band's edge at 50%, so the band-edge rule still holds the status at emerging.
Weighed against it: April read 51.0% on the same monthly series, so both monthly readings sit just above the edge; the weekly readings of 53.3% and 54.5% came from seven-day samples of Claude.ai conversations, so the step between them cannot be read as a trend. One provider's consumer plans, classified by a model.
- 2026-09-12consistent with normal to emergingconf 60 → 60 · Claude (model)
Held at emerging by the band-edge rule: augmentation is 51.4% of classified conversations (Anthropic Economic Index, 2026-05-31), within 5% of the normal band's edge at 50%, so the reading cannot separate 'augmentation at least half' from the gap below it.
Weighed against it: The three earlier waves read 53.3%, 54.5% and 51.0%, so the series has stayed above the edge rather than crossing it, and the value is unchanged on the card and the series page. What the rule denies is a scored status this close to the line, not the reading itself.
source1 - 2026-09-10unmeasured to consistent with normalconf — → 60 · initial seed (model-drafted)
Anthropic Economic Index, January 2026 report (November 2025 data): 52% of Claude.ai conversations classified as augmentation, 45% automation; the March 2026 report says augmentation rose again but only in figures. Inside the normal band (augmentation >= 50%). Tier 2 provider disclosure. Initial seed.
source1
Why this confidence
Rests on 11 observations from Anthropic Economic Index (open dataset and reports); best evidence tier 2, grade B; current.
Confidence is independent of status: 90–95 multiple strong independent sources; 70–89 good evidence, some ambiguity; 50–69 mixed or hard to operationalise; 0–49 limited or vague evidence.
Predictions that lean on this
- Too early to tell