TRV-2026-2526Version 1 · Published

Saved · see the current record

Why this version exists

Certified into the record

The exact text we saved

This is the version exactly as it was saved, character for character. The SHA-256 hash below is a 64-character code worked out from that text. Run the same text through SHA-256 and you get the same code; change a single letter and the code comes out completely different. Each version also stores the previous version’s hash, which chains them together so no version can be swapped out quietly.

TRUVACE RECORD VERSION
record: TRV-2026-2526
version: 1
kind: certified
reason: Certified into the record
timestamp: 2026-11-12T07:48:23.696131Z
status: published
lens: p_space
sector: labor
headline: How rejection letters from AI recruiters shape recruiter's prestige and experienced respect
dek: Introduction Organizations increasingly rely on artificial intelligence (AI) in recruitment, yet little is known about how AI systems should communicate unfavorable outcomes such as rejection. The present research examined how type of rejection letters of an AI recruiter shape applicants' perceptions of the recruiter and their experienced respect. Methods Two experiments were conducted. In Study 1, the type of AI-generated rejection letter (considerate vs. typified) and applicants' employment status (unemployed…
gain_title: (none)
problem_title: In a scenario experiment with 232 adults, a generic AI rejection email made the recruiter seem more machine-like than a personalized, explained one, and that perception was linked to lower recruiter prestige and less felt respect.
trace_subject: (none)
gain_reading: (none)
gain_evidence: (none)
problem_reading: In a scenario experiment with 232 adults, a generic AI rejection email made the recruiter seem more machine-like than a personalized, explained one, and that perception was linked to lower recruiter prestige and less felt respect.
problem_evidence: Mechanistic dehumanization significantly predicted lower experienced respect
quick_read: Two online experiments tested what happens when an AI recruiter turns down a job applicant. Participants imagined applying for a Project Coordinator job and then received either a short generic rejection email or a longer one with personalization and an explanation. In the second study, with 232 adults, the generic email led people to see the recruiter as colder and more machine-like, which was tied to lower perceived prestige and less felt respect.

The first study, with 197 adults, gave a narrower result. Generic letters made the recruiter seem more machine-like only among participants who were told their application showed a 2-year unemployment gap. Among those without a gap, letter type did not change how mechanical the recruiter seemed.

The second study tried to replicate this with real employment status: 150 employed people and 82 unemployed job seekers. Unemployment status did not change the effect of letter type there. Exploratory analyses suggested attitudes toward unemployed people played a role, so who feels the sting of a generic AI rejection is still unsettled.
limitation: Both studies were scenario-based experiments in which participants imagined being rejected, and the vignette did not specify any particular AI system, so real applicants' reactions may differ.
tag: Evidence-backed problem
key_points: Study 2 included 232 adults from the UK, US and elsewhere: 150 employed and 82 unemployed and actively seeking a job. | Study 1 included 197 adults and used a 2 (unemployed gap vs. control) by 2 (generic vs. considerate letter) design. | A pilot of 80 employed UK adults confirmed the letters differed as intended; the full letter was rated most considerate (mean 5.10) and the mandatory-only letter least (mean 3.16). | The considerate letter added personalization and an explanation to the standard thanks, decision, refusal and best wishes. | In Study 1, generic letters raised perceived machine-likeness among participants framed as unemployed (effect 0.66, p = 0.01) but not among controls. | In Study 2, generic letters raised machine-like ratings of the recruiter, and that rating was tied to lower recruiter prestige and lower experienced respect. | In Study 2, the direct and total effects of letter type on experienced respect were both significant.
rundown: Participants recruited through Prolific in October and November 2025 imagined applying to a fictitious firm, IntraCOM UK, and were told an AI recruiter reviewed applications, assessed the interview and sent the decision. The vignette named no specific AI system, and the researchers did not measure participants' AI literacy.

The considerate letter used in the main studies added personalization and explanation to the standard components. Measures covered how machine-like the recruiter seemed (six items), perceived recruiter prestige (three items) and experienced respect (ten items). Study 1 also controlled for attitudes toward AI, belief that decision-making is uniquely human, and organizational conspiracy beliefs.

In Study 1, letter type alone did not significantly predict how machine-like the recruiter seemed. The interaction with employment status was significant (p = 0.04). Among participants framed as unemployed, the indirect effects of letter type on prestige (−0.35) and respect (−0.45) through machine-likeness were significant. Among controls they were not.

Study 2 was preregistered and recruited actual employed and unemployed participants. Letter type affected machine-likeness, and the indirect path to prestige and respect was significant. Employment status did not moderate the effect, which was an unexpected result. Exploratory analyses suggested attitudes toward unemployment shaped the interaction between letter type and employment status.
sources:
- peer_reviewed | Frontiers in Organizational Psychology | https://doi.org/10.3389/forgp.2026.1933320 | 2026-09-07
prev: 0000000000000000000000000000000000000000000000000000000000000000
SHA-256 hash
bedacac52f59fcb57bad147751a4aea4765e20dea8fc7484faa1762077f5bcfa
Previous version’s hash
0000000000000000000000000000000000000000000000000000000000000000
Check this record hasn’t been altered
How to check it yourself, without trusting this page

Download any version from /api/record/TRV-2026-2526 and hash its canonical field yourself, for example with shasum -a 256. You should get its content_hash. Each version’s text also ends with prev:and the hash of the version before it (version 1 points to 64 zeros). If anyone had changed a single character of any version after it was saved, those hashes wouldn’t line up.