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-2026-2574
version: 1
kind: certified
reason: Certified into the record
timestamp: 2026-11-12T07:56:10.621144Z
status: published
lens: g_space
sector: labor
headline: Large Language Model-Based Artificial Intelligence as a Relational Leader? Effects of Leadership Style and Stress on Artificial Intelligence Leadership
dek: This study provides an initial investigation into the objective performance, subjective perceptions, and neural correlates of relational leadership behaviors demonstrated by large language model (LLM)-powered artificial intelligence (AI) leaders in the workplace. A 2 × 2 × 2 mixed-design experiment was conducted, with leadership style (relational versus non-relational) and supervisor type (human versus AI) as between-subjects variables, and stress level (high versus low) as a within-subjects variable. Eighty-two…
gain_title: In an experiment with 82 university students, an AI supervisor that used relational leadership behaviors improved workplace well-being, self-perceptions and leadership ratings, and most perceptions were similar to those under a human supervisor.
problem_title: (none)
trace_subject: (none)
gain_reading: In an experiment with 82 university students, an AI supervisor that used relational leadership behaviors improved workplace well-being, self-perceptions and leadership ratings, and most perceptions were similar to those under a human supervisor.
gain_evidence: LLM-based AI leaders could effectively simulate relational leadership behaviors, enhancing workplace well-being, self-perceptions, and leadership evaluations | Most subjective perceptions were similar under human and AI supervision
problem_reading: (none)
problem_evidence: (none)
quick_read: Researchers ran an experiment in which 82 university students performed simulated automotive parts inspections under either a human or an AI supervisor. The AI supervisors were powered by a large language model and behaved in either a relational or a non-relational way.

The study reports that the AI supervisors could convincingly simulate relational leadership, and that this style improved well-being, self-perceptions and leadership evaluations. Most subjective ratings were similar for human and AI supervisors, though a few measures favored human leaders.

High stress worsened participants' subjective perceptions, but relational leadership buffered those effects. Brain scans showed prefrontal activity linked to AI leadership, relational behavior and stress. The authors call this initial evidence and stress careful design to promote well-being.
limitation: The authors describe the work as an initial investigation using 82 university students in a simulated automotive inspection task, so the findings come from a small, non-workplace sample.
tag: Evidence-backed gain
key_points: The experiment used a 2 × 2 × 2 mixed design: leadership style (relational or non-relational) and supervisor type (human or AI) varied between participants, and stress level (high or low) varied within participants. | Eighty-two university students performed simulated automotive parts inspections under human or AI supervision. | Relational AI leaders enhanced workplace well-being, self-perceptions and leadership evaluations. | Most subjective perceptions were similar under human and AI supervision, with a few measures favoring human leaders. | High stress negatively affected subjective perceptions, but relational leadership helped buffer the effect. | Functional near-infrared spectroscopy measured prefrontal cortex activity in areas tied to task switching and emotional regulation.
rundown: The study tested whether a large language model could act as a relational leader in a workplace-style task. Participants inspected automotive parts in a simulation while supervised by either a human or an AI, using a relational or non-relational leadership style.

Stress was manipulated within participants, so each person experienced both high and low stress conditions. High stress lowered subjective perceptions, but relational leadership softened that drop.

Prefrontal cortex activity, recorded with fNIRS, was linked to AI leadership, relational behavior and stress levels, in regions associated with task switching and emotional regulation. The authors say the findings offer initial empirical evidence for LLMs as relational leaders and argue for deliberate design to protect well-being.
sources:
- peer_reviewed | Journal of Leadership & Organizational Studies | https://doi.org/10.1177/15480518261458468 | 2026-06-08
prev: 0000000000000000000000000000000000000000000000000000000000000000

SHA-256 hash

a0becb022942e95557e986dce5c3ef9ac35e2be4d56e3ed734d9fd250dbbf70a

Previous version’s hash

0000000000000000000000000000000000000000000000000000000000000000