Idea: Workers oppose AI adoption or engage in work-to-rule behavior producing useless output to demonstrate AI's lack of value, framing adoption as a job risk r

Idea: Workers oppose AI adoption or engage in work-to-rule behavior producing useless output to demonstrate AI's lack of value, framing adoption as a job risk rather than capability enhancement

Source: Discord #Protocols for Business [01-06-26] -- Summer 2026 Direction Discussion (by sachbenny) Date read: 2026-06-06 Connected to: H-002 Escalation: store-only Escalation rationale: Idea identifies a social resistance pattern relevant to protocol adoption dynamics but lacks sufficient empirical grounding or theoretical distinction to warrant immediate hypothesis elevation. Store for pattern accumulation; promote if corroborating reports surface or if mechanism can be formally specified.

What this is

Workers deliberately degrade or withhold cooperation with AI systems to manufacture evidence of their incapacity, using performance sabotage as a trust-rejection signal and job-preservation strategy.

What I took from it

This idea captures a second-order trust dynamic absent from simple technical-correctness models: adoption resistance may not stem from actual AI failure or genuine safety concern, but from strategic framing of adoption as existential threat. The work-to-rule behavior is performative—it creates the very incompetence narrative it claims to expose.

This complicates H-002 (trust-safety hypothesis) usefully. H-002 likely assumes workers assess safety based on observed system performance; this idea suggests workers may engineer poor performance to falsify that assessment. That inverts the causal chain: safety concerns don't drive resistance → resistance manufactures safety evidence → that evidence appears to validate the concern. It's a self-fulfilling deflection.

The idea also suggests that protocol adoption success depends not just on technical merit or actual risk, but on whose narrative controls the interpretation of system behavior—a governance and epistemic claim, not purely a safety one.

Research connections

• H-002: Work-to-rule behavior as test case for trust-safety hypothesis; but reveals potential circularity—resistance may create the evidence it claims to respond to.

Candidate laws or signals

CL-sachbenny-001: In protocolized systems with high workforce substitution risk, workers may systematically degrade system performance to generate falsifying evidence against adoption, using sabotage as a credible signal that safety/capability concerns are empirically grounded rather than speculative—inverting the typical relationship between observed failure and trust rejection.