---
id: "concept-deliberate-inefficiency"
type: "concept"
source_timestamps: ["¶18"]
tags: ["system-design", "governance", "sustainability"]
related: ["concept-tragedy-of-commons-slow-motion", "framework-ai-accountability", "contrarian-inefficiency-is-good", "quote-deliberate-inefficiency", "action-escalation-rules"]
definition: "The intentional introduction of friction or human-in-the-loop processes to sustain long-term system health, capability, and accountability against short-term automation efficiencies."
sources: ["futures"]
sourceVaultSlug: "hbr-seg-futures"
originDay: 2
articleStem: "hbr-cl-84-big-tech-capability-crisis"
sourceUrl: "https://hbr.org/2026/06/big-techs-looming-capability-crisis"
sourceTitle: "Big Tech’s Looming Capability Crisis"
---
# Deliberate Inefficiency

## Deliberate Inefficiency

The intentional **reintroduction of friction**, human-in-the-loop processes, and redundancy into a system that AI is actively trying to optimize and streamline away.

The authors argue that in systems suffering from commons problems — like the shared talent pool of software engineers, see [[concept-tragedy-of-commons-slow-motion]] — deliberate inefficiency is *necessary* to internalize costs and sustain the system's long-term capabilities. Concrete forms include:
- Mandatory human [[action-mandatory-sign-off|sign-offs]]
- [[action-pair-senior-junior|Paired programming]] (senior + junior)
- Strict [[action-extend-provenance|provenance tracking]]
- [[action-escalation-rules|Escalation rules]]

This is the unifying principle of the [[framework-ai-accountability|AI Accountability & Capability Mitigation Framework]] and the core of the [[contrarian-inefficiency-is-good|contrarian claim]] that friction is a feature, not a bug. See [[quote-deliberate-inefficiency]] for the authors' framing.

> Enrichment: This is a real but *contested* governance strategy. An alternative view holds that well-designed automation — not mandated inefficiency — is what keeps systems sustainable at scale; the relevant lens is work-design and organizational-learning research.


## Related across articles
- [[concept-ai-jevons-paradox]]