# The Taint We Leave
# Book: Language
# Chapter URL: https://programmer.ie/books/language/the-taint-we-leave/
# Prose fingerprint (sha256, normalized): 4d924877e984fdcb09231688603bc284b1fd94d8531345180e3934aca101032a
# Snapshot generated at build time by Hugo. This is the complete rendered prose
# of the chapter; it is not a transcript of the source file.

Iris asks her proxy to look for someone with whom she could investigate a difficult result. Imagine that she and another person have independently authorised a private search using selected, limited views of their work. The system proposes a connection for an unexpected reason. Both repeatedly ask for the qualification behind a headline. Both reject a polished answer when its evidence cannot be traced. Both turn to a counterexample before accepting a general rule. Their subjects are different, and neither has written those habits into an interest list. The proxy offers Iris a reason to consider a possible collaborator without naming the other person, exposing their history or claiming that the affinity has been established. The other person’s identity and participation remain subject to a separate decision.
This is a constructed scene, not a run. It gives the hypothesis of this chapter its constructive purpose: two people might discover a productive connection that an interest list would miss. A machine might also mistake the habits of their shared assistant, the constraints of their work or the selectivity of the retained record for something about them. The useful signal and the confounder must enter the design together; neither has yet been demonstrated for Taint.
The last chapter asked what a proxy may say, and under which grant. Discovering an unknown recipient creates an earlier boundary. Before the proxy has anyone to contact, it may have to derive and compare representations of interaction histories. That computation itself is a use of information, not a neutral prelude to a later authorisation.
What trace of repeated interaction does it carry, and what may that trace be used for?
A name for the residue The book proposes the technical term Taint for this hypothesis. The capitalisation matters: this is a proposed concept in Language, not a claim to have established a new behavioural fingerprint or a standard representation.
Taint (proposed): a potentially machine-detectable, context-dependent pattern co-produced by a person and the models, prompts, tools, tasks, corrections, accepted outputs, refusals and recording practices present across repeated interactions. Whether the pattern is stable, distinctive or useful must be established for a specified use.
The name carries no moral judgement. It does not mean the person is spoiled or their work suspect. It names a possible residue of interaction that may be easier to describe statistically than from memory. Taint also has an established, different meaning in computer-security information-flow analysis, where labels track how designated data propagate through a program. That established technique can help track provenance; it does not establish the existence of this proposed interaction signature. The two meanings must not be silently equated.
This remains a book hypothesis. No Taint detector has been built or evaluated for this chapter, no participants have been tested, and no matching benefit has been measured. The phrase potentially machine-detectable expresses a falsifiable possibility, not an observation. The candidate pattern must outperform simpler explanations—topic, task, writing style, shared provider, explicit interests or retained-history artefacts—before it earns a distinct role. If it cannot, the term need not become a software object.
Several boundaries matter immediately:
Taint ≠ the person Taint ≠ identity Taint ≠ an explicit profile Taint ≠ memory Taint ≠ a biometric identifier Taint ≠ current intent Taint ≠ truth about the person Taint ≠ permission to infer or match A memory is a retained record or state. Taint, if measurable, would be a derived pattern over selected interactions and their conditions, not the underlying records themselves. A derived vector, cluster or description would be one representation of that candidate pattern, not the pattern’s proof. Keeping a history does not establish that a stable signature exists; deriving a signature does not establish its interpretation. An explicit profile records claims someone chose to declare or infer. A signature may become detectable without such a profile, but turning it into a persistent profile is a separate, permission-dependent act. An old pattern of rejecting brief answers does not establish Iris’s current request.
Even a reproducible association would first be evidence about interaction histories under specified conditions. It would not be a discovery of an essential trait, identity or hidden mental state. Repeatability under one model and one task is not yet transportability to other models, tools or contexts.
Co-produced, not extracted from a person Suppose two people regularly demand citations. That could reflect a shared approach to evidence. It could also reflect a system prompt that inserts citation checks, a tool that makes references easy to retrieve, an institutional review checklist or a task in which unsupported claims are costly. The observable event—citation requests—does not, by itself, attribute their cause to the people. The comparison needs evidence about what each person actually initiated or corrected, and what the machinery supplied.
The candidate pattern has several contributors: what each person requested, the model and its version, system instructions, defaults, tools, task types, interaction modes, feedback opportunities, corrections and refusals, and which records survived. Acceptance is not straightforward approval: Iris may accept an answer because it is sufficient, because review is costly or because she has moved on. A refusal may come from Iris or from the model’s own constraints. Missing provenance must remain missing. It cannot be filled with an invented account of the person’s intention.
pattern in the interaction ≠ property of the person alone Changing the model could change the apparent pattern. Removing correction logs could make agreement look more frequent. A shared organisational template could make unrelated people appear alike. Conversely, two people with related practices might look different because one uses voice input and the other edits long technical drafts. These are competing explanations, not background noise to average away. A defensible study would vary people, models, tasks and recording conditions separately enough to estimate which source accounts for a proposed signal.
Hypothesis illustration; no detector or matching benefit is established. Open full-size diagram. A compiler might eventually add proposed descriptions of communicative acts, qualifications and corrections to an authorised interaction history. Those annotations would remain fallible features of communications, with source and interpreter versions. They are not validated Taint measurements. Shared extractors could manufacture apparent similarity, while changed profiles could manufacture difference. No detector, identity inference, personality assessment, compatibility result or matching advantage follows from making the annotations available.
Birds of a feather, for a stated purpose The proposed positive application is discovering a candidate collaborator for a declared investigation. Two people might share a way of testing claims, preserving assumptions or interrogating counterexamples despite working on different subjects. A purpose-specific system could compare such evidence-handling practices without assuming that similar prose, interests or demographics would identify the better collaborator. Whether these practices predict an actually useful collaboration is precisely what the evaluation must establish.
“Birds of a feather” is a useful image only if it remains a question. A proposed match does not assert that two people are the same kind of person. It proposes that a particular relationship—possible collaborators for this investigation—may be worth considering. Similarity is one candidate mechanism; complementary working methods might be better. A matching study must therefore define the relationship and its outcome in advance rather than assume that the nearest signatures produce the best partners. A pleasant explanation of affinity is not evidence of a productive collaboration.
The chain should therefore remain explicit:
authorised purpose and eligible inputs ↓ permission to derive bounded discovery views ↓ comparison for a typed candidate relationship ↓ private, provisional match with uncertainty ↓ separate authority for disclosure and introduction The order is deliberate. Permission to retain records does not automatically permit computing a signature from them; permission to compute it does not automatically permit comparing it with another person’s view. Each side’s eligible data and purpose must be established before the relevant computation. A later match may be shown only in a form covered by the applicable grants. Neither an attractive score nor the possibility of an introduction supplies retrospective permission for the earlier computation.
detectable ≠ authorised for use similar Taint ≠ established relationship shared pattern ≠ permission to introduce Taint alone grants no match operation, identity reveal, introduction, disclosure or authority. A candidate can be rejected because a shared model produced the apparent affinity, because the proposed relationship is irrelevant, because an approved view has expired or because either participant withheld this use. The source of the pattern, the claimed relationship, the current permission and the decision to make contact must remain independently inspectable. The next chapter takes up the later introduction stages in detail.
What would be permitted to travel The smallest useful object would be a purpose-specific discovery view, not a portable personality dossier. For Iris’s investigation, a proposed view might contain a bounded description of how evidence is handled, the kinds of tasks from which it was derived, the interval covered, known model and tool conditions, an uncertainty statement and the precise uses Iris accepted. This is an illustrative contract, not an implemented schema or an assertion that the pattern can be extracted. A receiver need not learn the underlying examples, private prompts or collaborators’ names merely to evaluate an authorised comparison.
Before any opted-in matching, Iris should be able to inspect a comprehensible description of the proposed view, what evidence categories support it, where its support is weak and what consequences its use can have. Inspection must not require displaying another person’s protected source material. She may correct a misinterpretation, narrow the task scope or exclude a record while leaving legitimately retained history unchanged. She may also reject an accurate observation for this purpose. A view authorised for research collaboration cannot silently be repurposed for recruitment, lending, professional ranking or general identity search.
The controls follow Chapters 23–33: purpose, eligible data, provenance, current validity, authority to derive and compare, limited destinations, inspectable consequences, expiry and correction. A derived view needs its own lifetime and a record of where it was used, bounded by legitimate retention. Revocation stops new authorised uses; it does not prove that remote copies, cached match lists or prior disclosures have disappeared. Such effects need separate invalidation, confirmation or an explicit report of what remains unknown.
Third-party information needs its own boundary. Iris’s history may contain a colleague’s unpublished result, someone else’s refusal or a private conversation. Iris’s permission to search her own work does not settle their rights or the authorised uses of their contributions. Even a derived description may reveal a confidential fact if the pattern is distinctive enough. Inputs without a sound basis for this use must be excluded or the operation refused; summarising them as a vector is not a way around that restriction.
There must be a useful path without Taint. Iris may declare a collaboration goal, choose a small set of public or shareable work samples, or search by explicit criteria. Declining signature derivation cannot be treated as refusal of collaboration itself. The ordinary path should retain comparable privacy, discoverability and correction protections, so that any experimental gain from Taint is measured against a credible alternative rather than an intentionally weak default.
When the signature becomes the mistake The first failure would be covert profiling: deriving a durable representation from traces gathered for unrelated assistance, without permission for that derivation or use. Even a view that omits names might permit linking records across sessions or inferring sensitive attributes. A match score or repeated private query may disclose membership in a candidate pool without ever printing a name. If the system then affects who is offered work, invited into a network or considered credible, discrimination can be hidden inside an apparently helpful discovery service. These are risk hypotheses to investigate, not observed failures of an existing Taint implementation.
Context separation matters because a signature formed under a private task could contaminate a public or professional one. A person’s corrections during a confidential dispute must not silently become evidence offered to a collaboration service. Removing names does not settle the problem when attributes can be linked or reconstructed. The relevant test is whether the new computation and resulting disclosure are permitted for the original contributors and the new purpose—not whether the exported object happens to be called an embedding.
The apparent affinity could be induced or manipulated. Shared models, prompting conventions and organisational templates might make unrelated people look alike. Copied writing, deliberate mimicry or poisoned interaction logs could attract or suppress a match. A model change could erase the apparent similarity without either person changing. A stale pattern could continue influencing results after its subject changes practice or withdraws permission. No single similarity score distinguishes these possibilities; provenance checks, time-scoped evaluation and adversarial controls are needed.
Finally, the system could help create what it later claims to discover. If it continually suggests collaborators, examples and answers resembling the current signature, Iris’s future interactions may converge on those suggestions. A later detector could then score the induced resemblance as fresh evidence of her affinity. Such feedback makes the intervention part of the measurement conditions, as Chapter 30 warned about predictions that shape their own outcomes. A discovery programme should keep an independent or minimally mediated comparison path and examine whether exposure narrows the range of candidates over time. At scale, a shared provider could produce apparent communities that mostly reflect its defaults. The later proxy-ecosystem chapter returns to this concern.
How the hypothesis could lose A proposed evaluation would compare at least three discovery paths: declared interests and selected work samples, ordinary task/topic or embedding similarity, and a structured Taint-derived view. All must operate under comparable, authorised input budgets and the same permitted discovery purpose. The benchmark should define candidate collaboration usefulness before ranking systems are compared, rather than ask raters whether two profiles sound alike. The candidate relationships and their outcomes must be held out from feature construction; comparable task and temporal splits help prevent the same interaction from supplying both the match and its supposed validation.
The evaluation should report ranking quality and coverage alongside false matches, unattributed similarities, unwanted contacts, privacy leakage and correction burden. Where appropriate and voluntarily studied, a later measure of collaboration usefulness should be kept distinct from an initial expression of interest. Exposure to a flattering description may make a suggested match seem attractive without predicting useful work. Privacy and authority failures are separate constraints, not negative points that better ranking can cancel in a weighted score. Legitimate permission for participants and third parties must precede collection or testing; calling an intrusive procedure a privacy experiment does not authorise it.
The hard cases should be deliberately crossed: different people using the same model and template; the same person switching providers; the same task performed with and without an assistant; different people sharing a task; a person changing practices over time; imitated or poisoned histories; and records with missing corrections. The design should measure whether a detector continues to identify an interaction-derived pattern when provider and task artefacts are controlled. Models must not receive identities or held-out examples through metadata, prompts or evaluation feedback. An uncertainty or abstention outcome should be available when evidence is insufficient.
The result could be that Taint adds nothing beyond declared interests or ordinary embeddings, that matches fail when shared-provider artefacts are removed, or that incremental useful matches cost too much in false attribution, profiling or linkability. These are legitimate negative outcomes. The proposal should also be rejected if participants cannot understand or meaningfully limit the derived view, or if matching cannot work under acceptable disclosure limits. In each case the non-Taint route wins. No commitment to this terminology requires deploying the representation.
Proposed evaluation, not results; a non-Taint baseline can win. Open full-size diagram. Research context, not evidence of Taint Research on AI-assisted writing gives reason to treat the shared model as a contributor rather than attribute every repeated pattern to the person. Agarwal, Naaman and Vashistha report in a controlled writing study that AI suggestions shifted aspects of participants’ writing toward the provider’s conventions. This does not demonstrate a Taint detector or collaboration benefit; it motivates controls for model-induced similarity. NIST SP 800-188 treats de-identification as risk reduction that requires evaluation, not a guarantee that derived or name-free data cannot be linked. Neither source validates private matching here. They ground two hazards that the proposed experiment must test.
Where this leaves us Repeated interaction might leave a pattern that a machine can describe more readily than its participants can. Taint names the proposed, co-produced residue and the possibility of using a purpose-specific view of it to suggest collaborators. It is not yet a measured signature, a stable characteristic of Iris, an identity proof or an authorisation. Any evidence of affinity would remain conditional on the model, task, recording history and permission that produced the comparison. Its potential value must be demonstrated against simpler discovery methods without trading away the boundaries that made the search legitimate.
The remaining question is practical and exact. Suppose a private comparison really can help two people discover a worthwhile connection. How can it do so without converting either person’s interaction history into a public profile or revealing an identity before its owner agrees? Computing a candidate relationship, revealing a match, contacting a person and arranging an introduction are different operations, potentially involving different principals and permissions. The next chapter separates those acts.
References Agarwal, D., Naaman, M. & Vashistha, A. (2025). AI Suggestions Homogenize Writing Toward Western Styles and Diminish Cultural Nuances. CHI 2025, Article 1117. https://doi.org/10.1145/3706598.3713564 Garfinkel, S., Guttman, B., Near, J., Dajani, A. & Singer, P. (2023). De-Identifying Government Datasets: Techniques and Governance. NIST SP 800-188. https://doi.org/10.6028/NIST.SP.800-188