Reciprocity and the Extractive Practice

In the AI Story

Franklin's reciprocity criterion was one of seven in her technology checklist: Does it promote justice? Does it restore reciprocity? Does it confer divisible or indivisible benefits? Does it favor people over machines? Does it minimize disaster rather than maximize gain? Does it favor conservation over waste? Does it favor the reversible over the irreversible? The reciprocity question is most revealing when applied to AI-augmented cognitive work because the extraction is invisible—no visible depletion, no measurable decline, only the slow narrowing of independent capability that becomes apparent when the tool fails and the worker discovers she cannot function without it.

There are genuinely reciprocal modes of AI use—exchanges building the user's capacity even as they augment it. A practitioner using AI to explore unfamiliar domains, interrogating suggestions against her own judgment, treating interaction as Socratic dialogue rather than production pipeline, may develop understanding through the exchange. These moments exist and are valuable. But they occur within a larger practice structured for extraction. Institutional incentives—metrics rewarding output, evaluation criteria measuring throughput, competitive pressures punishing deliberation—push systematically toward extractive mode and away from reciprocal one.

The design of tools themselves reinforces the extractive pattern. AI systems are optimized for user engagement and output quality, not user development. No mainstream AI tool currently pauses to ask whether the user understands the output she has accepted. No tool tracks whether the user's independent capability is growing or declining over time. No tool distinguishes between a user accepting output after rigorous evaluation and a user accepting output without examination—both interactions look identical from the system's perspective, both contribute equally to engagement metrics the tool's designers optimize for. Franklin would identify this as a design choice, not technical inevitability.

A reciprocal AI tool would be designed differently. It would expose reasoning at critical junctures—not in simplified post-hoc explanation features but in ways genuinely inviting the user to evaluate logic against her own understanding. It would occasionally withhold output, presenting instead a scaffold requiring the user to complete the final step independently, building the muscle of judgment that full automation's convenience allows to atrophy. It would measure the user's growing independence as success metric alongside growing productivity. These features would reduce efficiency, slow output production, introduce friction into a process designed to be frictionless. And that is precisely the point—reciprocity requires friction, the mechanism through which the exchange sustains both parties. The farmer's effort to return nutrients is friction, slowing the harvest cycle, costing time and labor. It is the practice ensuring there will be a harvest next year.

Origin

Franklin's reciprocity framework emerged from her study of ecological systems and her peace activism—both domains where short-term extraction produces long-term collapse. The nuclear testing she helped curtail was extractive in the most literal sense: taking atmospheric stability for weapons development without returning it. The feminist economics she engaged with—Marilyn Waring, ecological economics—all emphasized that market prices systematically undervalue reproductive and maintenance work because these activities produce no commodity for sale, only the conditions under which commodity production remains possible. Reciprocity, in Franklin's framework, is the correction—the insistence that sustainability requires practices giving back what they take, even when the giving is invisible to production metrics.

Key Ideas

Asymmetric exchange. The user gives finite developmental experience; the AI gives infinite output—the structural signature of extraction dressed as reciprocity, sustainable only while accumulated cognitive capital remains.

Design choices encode extractive logic. AI tools optimize for engagement and output quality, not user development—measuring productivity without comprehension, rewarding acceptance without evaluation, training dependency while appearing to provide capability.

Reciprocal design would introduce friction. Tools maintaining cognitive sustainability would expose reasoning, require completion of final steps independently, measure growing independence—reducing efficiency to preserve the conditions for judgment.

Democratization requires reciprocity. The developer in Lagos building what only teams could build before represents genuine capability expansion, but sustainability depends on whether the practice develops her understanding or merely her output—builder versus operator.

Soil depletion is invisible until harvest fails. Cognitive resources keep producing while depleting—the engineer cannot debug without the tool, the lawyer cannot construct arguments from first principles—the foundation gives out when the practice encounters challenges requiring depth no longer there.