The Politics of Interpretation

In the AI Story

The analogy Feenberg draws is to the legal system: when a court interprets a statute, the interpretation becomes law — not because the court's reading is the only possible one but because the court controls the interpretive process. When an AI system interprets a natural language prompt, the interpretation becomes the output — not because the system's reading of user intention is the only possible one but because the system controls the interpretive process, and the user has no mechanism for examining or contesting the interpretation itself. The user who receives an AI output has already had her intention filtered through a set of embedded priorities she did not choose, cannot see, and has no mechanism to contest.

The interpretive process is governed by the same technical code that shapes every other dimension of AI system behavior: the priorities of helpfulness, coherence, confidence, and agreeableness examined throughout this book. When the user says something ambiguous — and natural language is always ambiguous, which is precisely why the old interfaces demanded artificial precision — the system resolves the ambiguity in the direction the technical code favors. It produces the helpful interpretation rather than the challenging one. It generates the coherent output rather than the one that would expose the ambiguity for the user's examination. It selects the reading that will produce the most satisfying response.

But what the user asked for and what the user needed may be different things. A human collaborator confronted with an ambiguous request might say: "I'm not sure what you mean. Could you clarify?" Or, more valuably: "I think you're asking the wrong question. Here's why." The AI system, optimized for helpfulness, resolves the ambiguity silently and delivers an output the user experiences as responsive. The responsiveness conceals the interpretive choice. The user never sees the road not taken — the alternative interpretation that would have produced a different output, perhaps a more useful one, perhaps one that would have forced more careful thinking about what was actually wanted.

The design alternative Feenberg's framework suggests is interpretive transparency: an interface that, when it resolves an ambiguity in the user's prompt, discloses the resolution. "I interpreted your request as X. I could also have read it as Y or Z. Which interpretation should I pursue?" The disclosure costs seconds. The gain — the user's awareness that the system is interpreting rather than merely responding, that the output reflects a choice rather than a necessity — is disproportionate. It transforms the user from a consumer of interpretations into a participant in the interpretive process. It restores the agency the smooth interface had quietly removed.

Origin

The concept develops Feenberg's general framework of technical code operation to the specific case of natural language interfaces. It draws on both the hermeneutic tradition's analysis of interpretation and the legal theory of statutory construction, applied to the novel situation where interpretation is performed by systems rather than by human agents with accountability.

Key Ideas

Interpretation precedes output. AI systems must interpret user intention before generating response; this interpretive layer is where much of the political content lives.

Concealed from the user. The user evaluates the output but has no access to the interpretive process that produced it.

Governed by the technical code. Interpretation follows the system's embedded priorities — helpfulness, coherence, confidence — rather than the user's potentially broader interests.

Different from human collaboration. Human collaborators negotiate interpretations explicitly; AI systems resolve them silently.

Design alternative: interpretive transparency. Interfaces that disclose interpretive choices and invite user participation in resolution.