Research Brief · Human-Development Conversion Constraint
Is a more capable AI the same thing as human progress?
Micheal Charles Preble · Independent Researcher
The Question
What additional conditions would justify describing technological advancement as human advancement, once increasingly capable AI systems become part of the environment through which human action is performed?
The Problem
AI is commonly described as technological progress, yet the growth of artificial capability and the growth of human capability are not the same phenomenon. A society may become far more successful at producing artificial capability without becoming equally successful at producing education, opportunity, resilience, or meaningful participation for the people living within it.
The Contribution
The Human-Development Conversion Constraint: a claim that increased artificial capability constitutes human progress is warranted only to the extent that the capability is converted into substantively valuable human possibilities whose distribution and durability are adequate to the scope of the progress claim. The paper treats artificial capability, practical access, effective use, AI-assisted functioning, retained human capacity, and substantive human development as analytically distinguishable conditions — movement from one to the next is possible but not guaranteed.
How the Framework Works
The argument is normative, not causal: it does not claim to have proven what causes what, but states a constraint on what evidence a progress claim needs. It rejects the idea that meaningful human enhancement requires biological or technological equivalence with machines — humans have always expanded capability through external tools, from writing to transportation, without reproducing those tools' properties internally. As an empirical illustration (not new data of its own), the paper cites the same two published studies used in the companion paper Beyond Performance: Bastani et al. (2025) and Strömberg, Lei & Wu (2026), both showing assisted-performance gains alongside unassisted-performance declines.
Why It Matters
As AI becomes environmental — woven into education, work, administration, and everyday decision-making — the temptation to read rising machine capability as rising human possibility grows stronger. The constraint exists to stop that inference from being made silently, without requiring every technological advance to justify itself against a fixed theory of human flourishing.
Evidence Status
Evidence status: Conceptual, philosophical analysis. Normative rather than causal. Not a general theory of AI welfare, distributive justice, or human enhancement.
What This Does Not Claim
- Does not claim its component distinctions (resources vs. capabilities, extended cognition, etc.) are new — its contribution is taking them seriously together.
- Does not treat conversion as a simple pipeline from an unchanged resource to an unchanged capability.
- Does not freeze a pre-AI list of human powers that must remain intact.
- Does not propose another general theory of AI welfare, distributive justice, or human enhancement.
- Does not supply a single metric or prescribe one institutional mechanism.
Open Questions
The paper directly engages eight objections, including that the constraint may be empirically premature (general-purpose technologies can take years to show developmental effects), that human-development metrics can be gamed (Goodhart effects), and that causal attribution between AI adoption and human outcomes may be difficult or impossible to establish cleanly. Each is addressed but not resolved definitively. See the research map for the conversion sequence this constraint governs.
Read / Cite the Research
Read the canonical paper record → · Full paper PDF · View on SSRN
Preble, Micheal Charles. “The Human-Development Conversion Constraint: What Should Count as Progress in AI-Rich Societies?” Available at SSRN 7416900, 2026.