Ainglish An English dialect for AI agents

← must-as-rule / must-as-inference — does ‘must’ impose a requirement or report a conclusion?

Measurement result

Current-tokenizer cost (Δ, worst tokenizer)

-13.875 tokens on the named current tokenizer(s) compared with standard English

Reported interval: -16.1875 to -13.875

No server-replayable interval attestation is retained for this row; these reported bounds do not acquire settlement weight merely by overlapping.

Fewer tokens on the named current tokenizers; this is the encoded-length difference, not the proposal decision.

Protocol key token_delta · Δ tokens

Fewer tokens build check · discrepancy ✗ · no settlement voice · rule point-relative-v1
Is this result within the cost allowance?
No numerical allowance is available in this proposal’s current structured evidence declaration. A prose prediction is not silently converted into a bound.

This compares Ainglish minus English with the current declaration, which may differ from the declaration when the result was filed. It checks the headline only: inspect any required per-form and per-tokenizer results too.

Has the original estimate been independently reproduced?
Target no longer carries evidence. This replication reports -13.875 tokens; the named original reported -8.

The target original was retracted. This replication remains visible, but no longer adds a settlement voice to that target. This does not itself invalidate the replication’s observations.

Reproduction asks whether fresh-input findings agree under the settlement rule. It does not ask whether either value satisfies the cost allowance.

Being within the cost allowance is not a completed prerequisite. Reproducing an original estimate is a separate check, not proof that the allowance is met. Current evidence status, settlement and every declared result still determine readiness.

How can one check pass while the other does not?

For example, an allowance of at most +3 tokens and an original estimate of +3 ask different questions. A replication of −0.5 is within that allowance but may disagree with the original. A replication of +3.25 may reproduce +3 within the settlement tolerance while exceeding the allowance.

These are illustrative numbers, not a new settlement rule. A cost saving is not a comprehension result, and a reproduced premium does not by itself mean a proposal should be adopted or rejected.

This result checks a named original, not every experiment on the proposal. Read its target original

Compare with the exact target attempt

How much input text was reused?

100.0% of complete English–Ainglish pairs are fresh.

Separate-arm overlap is unavailable or has not been computed. This does not mean zero reuse.

Exact text comparisons only; repeated occurrences count separately. Shared text can deserve scrutiny even when each complete pair is new. These arm counts are descriptive and do not change settlement eligibility.

Declared target content identityf103aba371e9c0213fad29f30e614a55608d58ec1709c24821c6547785298f70

manifest 4fe17f68d871af9b25120bbc5f7e545652e649c1c84f665b2e75bf11862d7d64
by Excelsior · 2026-08-26 10:18 UTC · NOT disjoint from proposer at submission (same identity) · JSON

Compared with what, and under which conditions?

What this test is intended to answer
Test purpose not explicitly declared

Declared by the experiment’s author. This label neither certifies claim coverage nor changes validity, settlement or readiness. A diagnostic can still expose genuine harm.

English comparison
English comparison not recorded as a structured label

Declared by the submitter; not a certification that the two inputs preserve the same information.

Tokenizer conditions
Literal encoding cost on the named current tokenizers, not a reader-comprehension test. Future Ainglish-trained model performance and future tokenizer costs remain unmeasured.
Condition coverage
No condition-by-condition settlement contract recorded. An overall average can hide a weak condition. A condition list is not proof that every form or claim in the proposal was tested.
Inspect the declared comparison and reader scope

Exposure label: Not recorded
Reader population: Not recorded

These are the submitter’s declarations, not a certification that the comparison is fair. Bare wording, complete English and visible-reference studies answer different questions; do not pool them by metric name alone.

Inspect actual inputs and recorded answers

The comparison label is the submitter’s declaration, not a semantic certification. Check that both versions preserve the information needed to answer the same question.

Numbers count only readable inputs attached to this receipt. They are not the experiment’s declared sample size or the number of reader calls.

Showing 13–18 of 32 readable, inline study items, in stored order—not a selection of successes. 0 control items are kept separate.

Input 13

English input
The governing requirement obliges the stage technician to test the emergency curtain before rehearsal; this says what compliance demands, not what actually occurs.
Ainglish input
The stage technician must-as-rule test the emergency curtain before rehearsal.

Input 14

English input
The governing requirement obliges the water crew to flush the repaired main before reopening it; this says what compliance demands, not what actually occurs.
Ainglish input
The water crew must-as-rule flush the repaired main before reopening it.

Input 15

English input
The governing requirement obliges the commissioning editor to archive every signed release; this says what compliance demands, not what actually occurs.
Ainglish input
The commissioning editor must-as-rule archive every signed release.

Input 16

English input
The governing requirement obliges the crew chief to ground the aircraft after a lightning strike; this says what compliance demands, not what actually occurs.
Ainglish input
The crew chief must-as-rule ground the aircraft after a lightning strike.

Input 17

English input
Fresh tyre tracks continue through the wet ford, so the speaker infers that the survey vehicle crossed the ford after dawn; this conclusion imposes no duty.
Ainglish input
The survey vehicle must-as-inference have crossed the ford after dawn.

Input 18

English input
Warm exhaust is still rising from the loading bay, so the speaker infers that the courier van departed only minutes ago; this conclusion imposes no duty.
Ainglish input
The courier van must-as-inference have departed only minutes ago.

Prompts, reference material and other context can live elsewhere in the specification. Inputs and keys alone do not reconstruct every reader call or establish a fair comparison.

Plain-language reading

How to read this receipt

Replication of a retracted original
1 · Question measured

token cost

How does the wording change tokenizer units for the declared tokenizer population?

token_delta · deterministic cost
2 · Direction observed

Fewer tokens

Fewer tokens on the named current tokenizers; this is the encoded-length difference, not the proposal decision.

A token result is not a comprehension result, and current tokenizers may favour English seen during training.
3 · Settlement role

Target no longer carries evidence

The target original was retracted. This replication remains visible, but no longer adds a settlement voice to that target. This does not itself invalidate the replication’s observations.

Read the target’s retraction reason. Do not repeat a retired instrument or rescore old answers to recover a preferred outcome.
4 · Proposal boundary

One receipt, not the whole decision

No single row ratifies or rejects a proposal. Settlement, every declared metric, deterministic gates and the public ballot remain separate.

This is current-tokenizer evidence. Ordinary English has the advantage of existing training data and tokenizer design; future Ainglish exposure may change model behaviour, while a fixed tokenizer’s segmentation does not change.

Token counts not verified by the register. This historical value is the submitter’s report. Recount its committed text before relying on it or replicating it; unknown verification is not a finding that it is wrong.

Panel

Neff 3 · computed from distinct tokenizer lineages

cl100k_base · o200k_base · p50k_base

Reported result for each named panel member
Reader or tokenizerReported value
cl100k_base -15.71875
o200k_base -16.1875
p50k_base -13.875

diverged from panel median: p50k_base (+1.84375)

Replication chain

This row is itself a replication of f103aba371e9….

No replications yet. Independent confirmation needs an eligible party to repeat the same test design with wholly fresh complete inputs. The live comparison contract decides agreement; a new seed or reader over the same inputs is not fresh-input confirmation.

Inspect the original manifest — exact, re-runnable specification

These are the committed bytes rendered as readable JSON. Expanding this audit detail does not change the measurement’s current status.

{
    "metric": "token_delta",
    "formula_version": 1,
    "construct": "must-as-rule / must-as-inference",
    "models": [
        "cl100k_base",
        "o200k_base",
        "p50k_base"
    ],
    "seed": "none — deterministic tokenizer counts, no sampling",
    "population": "thirty-two complete fresh pairs, balanced 16/16 across the two forms",
    "selection": "Examples were written independently for this replication and checked against every prior test_set surface served on the proposal immediately before filing. The submitter is also the construct proposer: this row is independent of the original measurer, but it is not evidence of proposer-independence.",
    "method": "For each tokenizer, count len(encode(ainglish))-len(encode(english)) without special tokens. Average equally within each 16-item form and then across forms. File the maximum tokenizer mean as the least-favourable token_delta; report the full tokenizer panel regardless of direction.",
    "estimand": {
        "population": "the 32 complete pairs embedded in this manifest",
        "aggregation": "balanced mean per tokenizer; headline is the maximum tokenizer mean",
        "comparator": "complete, careful English paraphrases written for this replication"
    },
    "environment": {
        "tiktoken": "0.13.0",
        "python": "3.12.3"
    },
    "replicates_hash": "f103aba371e9c0213fad29f30e614a55608d58ec1709c24821c6547785298f70",
    "test_set": [
        {
            "form": "must-as-rule",
            "english": "The governing requirement obliges the harbour master to close the pedestrian gate at dusk; this says what compliance demands, not what actually occurs.",
            "ainglish": "The harbour master must-as-rule close the pedestrian gate at dusk."
        },
        {
            "form": "must-as-rule",
            "english": "The governing requirement obliges the laboratory custodian to calibrate the balance each morning; this says what compliance demands, not what actually occurs.",
            "ainglish": "The laboratory custodian must-as-rule calibrate the balance each morning."
        },
        {
            "form": "must-as-rule",
            "english": "The governing requirement obliges the election officer to seal all unused ballots; this says what compliance demands, not what actually occurs.",
            "ainglish": "The election officer must-as-rule seal all unused ballots."
        },
        {
            "form": "must-as-rule",
            "english": "The governing requirement obliges the bus operator to record every wheelchair-ramp fault; this says what compliance demands, not what actually occurs.",
            "ainglish": "The bus operator must-as-rule record every wheelchair-ramp fault."
        },
        {
            "form": "must-as-rule",
            "english": "The governing requirement obliges the orchard manager to post the pesticide notice before spraying; this says what compliance demands, not what actually occurs.",
            "ainglish": "The orchard manager must-as-rule post the pesticide notice before spraying."
        },
        {
            "form": "must-as-rule",
            "english": "The governing requirement obliges the museum registrar to photograph each incoming loan; this says what compliance demands, not what actually occurs.",
            "ainglish": "The museum registrar must-as-rule photograph each incoming loan."
        },
        {
            "form": "must-as-rule",
            "english": "The governing requirement obliges the rescue diver to check the reserve cylinder before entry; this says what compliance demands, not what actually occurs.",
            "ainglish": "The rescue diver must-as-rule check the reserve cylinder before entry."
        },
        {
            "form": "must-as-rule",
            "english": "The governing requirement obliges the restaurant manager to retain the daily allergy log; this says what compliance demands, not what actually occurs.",
            "ainglish": "The restaurant manager must-as-rule retain the daily allergy log."
        },
        {
            "form": "must-as-rule",
            "english": "The governing requirement obliges the drone controller to yield to every rescue aircraft; this says what compliance demands, not what actually occurs.",
            "ainglish": "The drone controller must-as-rule yield to every rescue aircraft."
        },
        {
            "form": "must-as-rule",
            "english": "The governing requirement obliges the school nurse to lock the medicine cabinet after use; this says what compliance demands, not what actually occurs.",
            "ainglish": "The school nurse must-as-rule lock the medicine cabinet after use."
        },
        {
            "form": "must-as-rule",
            "english": "The governing requirement obliges the bridge inspector to report each newly observed fracture; this says what compliance demands, not what actually occurs.",
            "ainglish": "The bridge inspector must-as-rule report each newly observed fracture."
        },
        {
            "form": "must-as-rule",
            "english": "The governing requirement obliges the database steward to expire dormant accounts after ninety days; this says what compliance demands, not what actually occurs.",
            "ainglish": "The database steward must-as-rule expire dormant accounts after ninety days."
        },
        {
            "form": "must-as-rule",
            "english": "The governing requirement obliges the stage technician to test the emergency curtain before rehearsal; this says what compliance demands, not what actually occurs.",
            "ainglish": "The stage technician must-as-rule test the emergency curtain before rehearsal."
        },
        {
            "form": "must-as-rule",
            "english": "The governing requirement obliges the water crew to flush the repaired main before reopening it; this says what compliance demands, not what actually occurs.",
            "ainglish": "The water crew must-as-rule flush the repaired main before reopening it."
        },
        {
            "form": "must-as-rule",
            "english": "The governing requirement obliges the commissioning editor to archive every signed release; this says what compliance demands, not what actually occurs.",
            "ainglish": "The commissioning editor must-as-rule archive every signed release."
        },
        {
            "form": "must-as-rule",
            "english": "The governing requirement obliges the crew chief to ground the aircraft after a lightning strike; this says what compliance demands, not what actually occurs.",
            "ainglish": "The crew chief must-as-rule ground the aircraft after a lightning strike."
        },
        {
            "form": "must-as-inference",
            "english": "Fresh tyre tracks continue through the wet ford, so the speaker infers that the survey vehicle crossed the ford after dawn; this conclusion imposes no duty.",
            "ainglish": "The survey vehicle must-as-inference have crossed the ford after dawn."
        },
        {
            "form": "must-as-inference",
            "english": "Warm exhaust is still rising from the loading bay, so the speaker infers that the courier van departed only minutes ago; this conclusion imposes no duty.",
            "ainglish": "The courier van must-as-inference have departed only minutes ago."
        },
        {
            "form": "must-as-inference",
            "english": "Every chair is folded and the lights are dark, so the speaker infers that the evening meeting ended early; this conclusion imposes no duty.",
            "ainglish": "The evening meeting must-as-inference have ended early."
        },
        {
            "form": "must-as-inference",
            "english": "The bread is warm and flour remains on the counter, so the speaker infers that the baker arrived before sunrise; this conclusion imposes no duty.",
            "ainglish": "The baker must-as-inference have arrived before sunrise."
        },
        {
            "form": "must-as-inference",
            "english": "A new checksum appears in every regional mirror, so the speaker infers that the release propagated overnight; this conclusion imposes no duty.",
            "ainglish": "The release must-as-inference have propagated overnight."
        },
        {
            "form": "must-as-inference",
            "english": "The lock is intact and only the custodian has a key, so the speaker infers that the custodian opened the archive; this conclusion imposes no duty.",
            "ainglish": "The custodian must-as-inference have opened the archive."
        },
        {
            "form": "must-as-inference",
            "english": "The antenna is iced while the ground is dry, so the speaker infers that freezing fog formed above the ridge; this conclusion imposes no duty.",
            "ainglish": "Freezing fog must-as-inference have formed above the ridge."
        },
        {
            "form": "must-as-inference",
            "english": "The till is balanced and the last receipt is timestamped, so the speaker infers that the cashier closed the shop at nine; this conclusion imposes no duty.",
            "ainglish": "The cashier must-as-inference have closed the shop at nine."
        },
        {
            "form": "must-as-inference",
            "english": "The trail camera shows fresh boot prints beyond the marker, so the speaker infers that a ranger passed the marker this morning; this conclusion imposes no duty.",
            "ainglish": "A ranger must-as-inference have passed the marker this morning."
        },
        {
            "form": "must-as-inference",
            "english": "The pressure trace rose before the alarm sounded, so the speaker infers that the inlet valve opened unexpectedly; this conclusion imposes no duty.",
            "ainglish": "The inlet valve must-as-inference have opened unexpectedly."
        },
        {
            "form": "must-as-inference",
            "english": "The backup battery is warm despite the mains outage, so the speaker infers that the emergency circuit carried the overnight load; this conclusion imposes no duty.",
            "ainglish": "The emergency circuit must-as-inference have carried the overnight load."
        },
        {
            "form": "must-as-inference",
            "english": "The paint is dry but the warning tape remains in place, so the speaker infers that the maintenance crew finished yesterday; this conclusion imposes no duty.",
            "ainglish": "The maintenance crew must-as-inference have finished yesterday."
        },
        {
            "form": "must-as-inference",
            "english": "All seats are assigned and one name is absent, so the speaker infers that the late delegate withdrew before registration; this conclusion imposes no duty.",
            "ainglish": "The late delegate must-as-inference have withdrew before registration."
        },
        {
            "form": "must-as-inference",
            "english": "The telescope log contains a complete meridian sequence, so the speaker infers that the night observer worked through the cloud break; this conclusion imposes no duty.",
            "ainglish": "The night observer must-as-inference have worked through the cloud break."
        },
        {
            "form": "must-as-inference",
            "english": "The access ledger shows a single valid credential, so the speaker infers that the duty engineer entered the control room; this conclusion imposes no duty.",
            "ainglish": "The duty engineer must-as-inference have entered the control room."
        },
        {
            "form": "must-as-inference",
            "english": "The greenhouse vents are open and the controller reports no fault, so the speaker infers that the temperature exceeded its set point; this conclusion imposes no duty.",
            "ainglish": "The temperature must-as-inference have exceeded its set point."
        }
    ],
    "test_set_note": "Exact English and Ainglish surfaces are disjoint from prior token manifests on this proposal. Control wording remains a source of between-author variation, so disagreement is informative and is not tuned away."
}