Briefs / Brief №027 / Audit Packet
← Return to brief
Autonoma / Intelligence Brief №027 · Public Audit Packet · October 2026

Brief №027 Audit Packet

Using AI More Doesn't Make People Less Afraid of It

This audit packet supports Brief №027: Using AI More Doesn't Make People Less Afraid of It. Read the brief first for the full argument.

Autonoma briefs are designed to be inspectable. This packet shows what the brief claims, how each claim was tested, what it does not claim, and where caveats remain — without exposing raw internal logs, prompts, source-routing mechanics, hashes, local paths, secrets, or unpublished candidate claims.

← Open Brief №027 — Using AI More Doesn't Make People Less Afraid of It

§ 01

Brief Summary and Audit Verdict

Brief 027 asks whether heavier AI use settles fear of displacement or reliably lifts performance. It answers no on both and places the adoption job in how managers frame and support the work, not in more tool exposure.

Audit verdict: Supported at the association layer. Gallup's 9 September 2026 panel holds the fear gap, its costs, and the climate buffer; other Gallup surveys hold the manager-support and productivity shares; the Harvard experiment holds the performance figures. The Brief withholds any causal reading that manager behavior reduces fear, because Gallup's items measure organizational care and respect, not what a manager did. The join from climate and task boundaries to manager framing, and the 12-month timing, are Autonoma synthesis.

§ 02

Claim Register

#Claim in the BriefStatusDomainBound
B027-C01Workers using AI daily or several times a week were more than twice as likely as less frequent users to fear elimination within five years, in most waves; widest gap 6.7 points in 2024; 3% to 4% of all workers say very likely; roughly 19% say somewhat or very likely in Q1 2026. [1]Supportedwww.gallup.comFour waves, nearly 30,000 observations. Expectations, not jobs lost.
B027-C02Within the same worker, seeing the job at risk goes with 0.20 standard deviations lower engagement and 4.1 points more job searching. [1]Supportedwww.gallup.comAbout 30% seen in two or more waves; less precise. Association.
B027-C03Among frequent users, the use-fear link is 11.1 points smaller where workers feel cared for and 6.8 points smaller where they feel respected. [1]Supportedwww.gallup.comClimate items, not manager actions. Little difference among infrequent users.
B027-C04Without clear guidance, frequent use heightens uncertainty about job security; Selenko and colleagues tie identity threat to whether AI complements, replaces, or creates tasks. [1][7]Supportedwww.gallup.com; journals.sagepub.comGallup reading; Selenko is conceptual.
B027-C05Only 28% strongly agree their manager actively supports AI use; those who do are 6.5 and 8.8 times as likely to agree the tools are useful and offer chances to do their best work; 9.3 times as likely to agree AI has transformed work (n=23,717). [2][3]Supportedwww.gallup.comNovember 2025; April 2026. Self-report. Association.
B027-C06Engagement 48% with manager support vs 30% without and 31% nationally; 17% of frequent users give top productivity rating, 50% when plan, support, and engagement combine. [4]Supportedwww.gallup.com21 July 2026. Self-rated. Engagement may lift the 50% on its own.
B027-C07Fifty-two percent of U.S. workers use AI; writing and editing 51%, search 49%; coding and automation each 16% of users; 77% positive productivity rating for coding and automation users. [5]Supportedwww.gallup.com20 July 2026. Self-rated effect.
B027-C08758 knowledge workers, a 2023 language model; 18 tasks inside the frontier, 12.2% more tasks and 25.1% faster; 19% less likely to be correct on a task outside it. [6]Supportedwww.hbs.eduSeptember 2023 working paper. Guidance not tested.
B027-C09A desk review of 48 studies from 489 records says AI "may reduce performance when employees over-rely on inaccurate outputs." [8]Supportedwww.multiresearchjournal.com24 May 2026. Desk review; little independent weight.
B027-C10Manager behavior reduces displacement fear, as Gallup's headline implies.Withheldwww.gallup.comClimate items; association, not intervention.
B027-C11Over the next 12 months, Gallup's panel will keep showing more displacement fear among frequent AI users than occasional ones, and active manager support will stay well below half; seats without manager preparation leave anxiety and flat productivity ratings.Autonoma synthesis-Autonoma join. Labeled forecast.
§ 03

Source Ledger

§ 04

Evidence Boundaries

  • Gallup measured workplace climate (care, respect), not what managers did. Adoption surveys ask about backing the tool, not explaining roles.
  • Fear and engagement results are within-person associations from self-reported expectations, not realized layoffs.
  • The strongest panel estimates rest on the smaller group seen across multiple waves.
  • The Harvard result covers consulting tasks and a 2023 language model, with a single outside-frontier task.
  • The desk review is cited for one stated finding and is not used for any figure.
  • Brief 018 covers role purpose, Brief 025 junior judgment, and Brief 026 workflow redesign. Brief 027 covers fear and performance under frequent use, and how managers frame it.
§ 05

Dissent and Limiting Case

The strong objection is that fear among frequent users may be an accurate reading of exposure. Gallup says workers "are responding to a real concern," and occupational exposure predicts fear. A supportive workplace may ease the feeling without changing the outcome. [1]

The claim survives narrowed: within the same person, fear goes with lower engagement and more job searching, and the fear gap is smaller where people feel cared for and respected. No source tests whether training managers changes either number. [1]

§ 06

Falsification

This Brief is wrong, or has to be rewritten, if:

  • later Gallup panel waves show frequent users no more fearful than occasional ones;
  • a study of what managers say and do finds no effect on displacement fear or engagement; or
  • a test with current models finds no accuracy loss outside the frontier.
§ 07

Forecast Label

The Brief's forecast reads: "Over the next 12 months, Gallup's panel will keep showing more displacement fear among frequent AI users than occasional ones, and active manager support will stay well below half. Organizations that add seats without preparing managers to set task boundaries will see anxiety persist among heavy users and productivity ratings stay flat." It is Autonoma Intelligence synthesis. The timing is not a fact quoted from any source above.

Methodology

This audit checks each figure in Brief 027 against the source that states it. The link from manager framing to lower fear, like the 12-month timing, is Autonoma Intelligence synthesis. This draft was prepared with an AI assistant and reviewed and approved by the editor.