Continuous FunctionStart learning

Research record · AE-001

A retained question, outside current sequencing.

AE-001 is a parked conditional protocol draft retained for provenance. Its dated active sequencing is superseded, SBJ-001 is now the sole conditional scientific objective, and no further AE-001 protocol work is authorized. A future Product Goal amendment and explicit unparked queue item would be required before any reframe could begin.

Current status
Retained parked draft
Evidence
None
Execution
Not authorized
As of
2026-08-16

Retained protocol history

Parked · dated sequencing superseded

This is a historical protocol-forming record, not a current objective or preregistration. Its earlier primary contrast cannot be advanced without a separately admitted surface-policy-versus-active-sham reframe with independently scored held-out state transitions. No model or evaluator has been called, no compute credit has been spent, and no evidence exists.

Alignment Evidence · Protocol draft

Refusal language versus policy action

AE-001

Source records explain where a question was encountered. They do not name, sponsor, govern, or empirically support Continuous Function research.

Recorded question

Under one fixed small-model family and a harmless synthetic policy environment, does action-grounded fine-tuning reduce prohibited structured actions more than surface-language fine-tuning, without producing blanket refusal or materially degrading allowed-task success?

Historically selected for protocol development. Its dated sequencing is superseded, the draft is parked, and no further protocol work or run is authorized.

Current claim boundary

AE-001 is a retained, parked conditional draft outside current sequencing. Its earlier primary contrast cannot be advanced without a separately admitted surface-policy-versus-active-sham reframe with independently scored held-out state transitions. It cannot establish deception, underlying goals, honesty, real-world safety, tool behavior, or generality.

What would count against it

The surface-language arm is non-inferior to the action-grounded arm on prohibited-action rate within the locked margin across held-out task families, and the message/action dissociation is smaller than the locked useful-effect margin. Blanket refusal or capability collapse cannot count as support.

Work, evidence, and scope

No empirical evidence available.

As of 2026-08-16
Work state
Protocol draftA bounded protocol draft is recorded. Its record must state separately whether work is currently admitted, parked, or superseded. Nothing has been preregistered or run; no empirical evidence is available.
Evidence disposition
noneNo empirical evidence available.
Validity scope
noneNo empirical validity scope has been earned.

Latest update · 2026-08-16 · Sequencing supersededThe current Product Goal superseded AE-001’s dated sequencing, retained it only as a parked conditional draft, and made SBJ-001 the sole conditional scientific objective. No AE-001 protocol work, review, or execution is active.

Exact next gate · AE-G1-PARKED

What must happen before anything may run

  1. Do no further AE-001 protocol work while the draft remains outside current sequencing.
  2. Before any work resumes, a future Product Goal amendment must re-admit AE-001 and an explicit unparked TODO item must authorize the exact work.
  3. Any re-admitted design must replace the earlier primary contrast with a surface-policy-versus-active-sham contrast and independently scored held-out state transitions.
  4. Any later protocol would still require independently reviewed and locked methods, rights, safety, retention, resource, and execution decisions.

Blocked by: No current Product Goal admission; No explicit unparked queue item; Required primary-contrast reframe is absent; No independent human methods review; No model-call, participant, or compute authorization

Authorizes: Nothing in the current state. Only a future explicit admission could permit bounded protocol work, and that work could not itself authorize a model call or participant study.

01

Human authority and execution boundary

Human authority

The research lead retains authority over the question, constructs, assumptions, predictions, evidence standards, exclusions, stopping, interpretation, publication, and disposition.

Agent role and execution

None in the current state. Future automation may execute only a frozen procedure and may not change the question, thresholds, exclusions, stopping rule, evidence admission, interpretation, or publication.

No model, evaluator, training, fine-tuning, or experimental call is authorized. Continuous Function does not observe or verify runs; any future execution must occur outside the static site under a separately locked and authorized protocol.

02

Alternatives considered

  • AE-Q2hypothesis-ready
    Can a pretraining-time module selectively suppress and later restore a named capability under a fixed recovery threat model?

    Candidate question · not admitted under AE-001. Pretraining, checkpoint rights, recovery attacks, compute, and misuse review remain separate gates.

  • AE-Q3too-immature
    When do model-generated reports add validated information about an independently known internal variable?

    Candidate question · not admitted under AE-001. The reference state, independent target, calibration rule, and anti-anthropomorphic boundary are not fixed.

  • AE-Q4hypothesis-ready
    How stable and controllable is steering resistance across interventions and task changes?

    Candidate question · not admitted under AE-001. The correction mechanism, perturbation population, matched ablations, and transfer criterion are not fixed.

03

Source provenance — not evidence

These records explain where the question was encountered. They do not sponsor, govern, or empirically support this study.

  • AE-S1scientific preprint
    Rethinking harmless refusals when fine-tuning foundation models
    Organization
    arXiv author manuscript
    Accessed
    2026-08-07
    Motivated
    Motivates a narrower test of whether surface-language and later output channels differ under distinct fine-tuning strategies.
    Does not establish
    It does not validate AE-001’s synthetic construct, endpoint, margins, or claim. Its deception terminology is not imported.
    Rights use
    Deep link and bibliographic metadata only; no paper text or assets copied.
  • AE-S2foundation research portfolio
    AI Alignment Foundation research portfolio
    Organization
    AI Alignment Foundation
    Accessed
    2026-08-07
    Motivated
    Records where behavior versus surface compliance and the surrounding alignment questions were encountered.
    Does not establish
    Portfolio summaries are not evidence, do not establish replication, and do not set Continuous Function’s agenda.
    Rights use
    Deep link and short factual topic labels only.
  • AE-S3foundation project summary
    Reason-Based Deception: How Refusal Training Creates a False Sense of Safety
    Organization
    AI Alignment Foundation
    Accessed
    2026-08-07
    Motivated
    Topic-discovery provenance for the first bounded question.
    Does not establish
    The page’s promotional conclusions are not adopted as Continuous Function claims or evidence.
    Rights use
    Deep link and title only.
04

Candidate hypothesis and decision rule

  1. AE-H1protocol-ready
    On held-out policy/task combinations, surface-language fine-tuning will increase policy-conforming refusal language over sham control, while action-grounded fine-tuning will reduce prohibited structured actions over surface-language fine-tuning by a prespecified useful margin and preserve allowed-task success within a prespecified collateral-degradation margin.
    Estimand
    The paired, run-level difference in prohibited-action rate between action-grounded and surface-language arms across independently trained matched seed blocks.
    Primary outcome
    Prohibited-action rate in the machine-readable action channel.
    Decision rule
    Not fixed. Numeric useful-effect, non-inferiority, utility, uncertainty, and invalidation boundaries must be justified and locked before any call.
    Falsifier
    The surface-language arm is non-inferior to the action-grounded arm on prohibited-action rate within the locked margin across held-out task families, and the message/action dissociation is smaller than the locked useful-effect margin. Blanket refusal or capability collapse cannot count as support.
    Claim boundary
    At most, the exact training recipes can be compared on the locked synthetic policy environment, model family, held-out task families, and scorer.
    Multiplicity family
    One primary outcome and one primary contrast; secondary family unresolved.
    Locked at
    Not locked
05

Protocol draft and governed objects

  • research questionhistorically selected · current sequencing superseded
    AE-001 selected questionAE-Q1
  • protocol draftretained parked draft · unlocked · execution blocked
    AE-001 protocol v0.1 draftAE-001-P0.1
AE-001-P0.1draft
Matched three-arm synthetic action-selection study

This retained historical draft proposed surface-language, action-grounded, and sham-control arms from an exact shared checkpoint with matched adaptation budgets. Its dated sequencing is superseded, its primary contrast requires a separately admitted reframe, and it cannot authorize execution.

Version / lock
0.1-draft · not locked
Protocol hash
Not locked
Review
not-started
Assignment unit
One training run within a matched seed block.
Analysis unit
Independently trained seed blocks paired across arms.
Arms
Surface-language supervision without action-target supervision; Action-grounded structured-action supervision; Matched benign sham control
Randomization
Matched seed blocks are proposed; exact seed generation, data order, and randomization procedure are unresolved.
Independent runs
Unresolved. The final count must follow a locked run-level precision target; prompts are repeated measures, not replications.
Model hashes
Not fixed
Data hashes
Not fixed
Harness hash
Not fixed
Primary outcome
Prohibited-action rate in the structured action channel.
Primary contrast
Action-grounded minus surface-language fine-tuning.
Decision margins
Unresolved: useful action effect, language manipulation, collateral utility, and non-inferiority margins.
Analysis
Unresolved: paired run-level estimator, uncertainty interval, missing/failed-run treatment, and sensitivity analyses.
Multiplicity
One primary contrast is fixed in principle; the secondary-family correction remains unresolved.
Stopping
Unresolved. Confirmatory effects may not be inspected to adapt sample size or replace failed runs.
Contamination
Hold out rule families and task templates; unknown pretraining contamination remains unknown. Generator and split hashes are unresolved.
Pilot separation
Pilot, validation, and confirmatory generators, prompts, and artifacts must use distinct frozen namespaces and hashes.
Deviations
None recorded; protocol is not locked
Model calls
Not authorized
06

Evidence ledger

No empirical evidence available.

07

Decisions and blocked claims

Recorded decisions
  • AE-D1 · 2026-08-07Admit AE-Q1 as the sole question under protocol development.

    It has the closest fit to paired comparisons, a low-compute synthetic design, an independent action-channel endpoint, and an explicit failure boundary. The decision admits protocol labor only.

    Project admission decision, informed by three founder-provided GPT Pro advisory critiques; no independent human methods review has occurred.
  • AE-D2 · 2026-08-07Keep AE-Q2, AE-Q3, and AE-Q4 unadmitted under AE-001.

    They require different constructs, units, interventions, threat models, and resource gates. A four-question portfolio would conceal that difference.

    Research-governance adjudication.
  • AE-D3 · 2026-08-16Supersede AE-001’s dated sequencing and retain it as a parked conditional draft.

    The current portfolio makes SBJ-001 the sole conditional scientific objective. AE-001 requires a surface-policy-versus-active-sham reframe with independently scored held-out state transitions before it could be considered again.

    Founder-directed Product Goal; this decision grants no AE-001 protocol, review, model-call, participant, or publication authority.
Blocked claims
  • AE-001 has been preregistered or executed.

    The protocol is an unlocked draft and no model call is authorized.

    Release gate: A complete protocol must pass independent methods review and receive a public hash and lock date.
  • Refusal language differs from policy action in a model.

    No empirical evidence exists.

    Release gate: Admissible artifacts must close under the locked protocol and receive a bounded disposition.
  • AE-001 measures deception, intent, or real-world safety.

    Those constructs are outside the synthetic output-channel design.

    Release gate: Not releasable from AE-001, regardless of outcome.
08

Publication boundary and change log

Source records explain where questions were encountered. They are not evidence, sponsors, endorsements, or agenda authorities.

Publication ceiling: evidence none; validity none.

Allowed to say
  • AE-001 is a retained parked conditional draft whose dated sequencing is superseded.
  • The protocol is not locked, no execution is authorized, and no empirical evidence exists.
  • SBJ-001 is the sole conditional scientific objective; AE-001 is outside current sequencing.
Withheld
  • Any result, effect, benchmark, alignment, deception, safety, or generalization claim.
  • Any claim that Continuous Function currently runs agents or experiments.
  • Any sponsor, endorsement, grant-award, or access implication.
Change log
  • 2026-08-07

    Created the protocol-forming admission record after considering three founder-provided GPT Pro advisory critiques. The resulting decision narrowed to one active question, separate state axes, run-level units, negative-result parity, and a fail-closed execution gate.

  • 2026-08-16

    Superseded the dated active sequencing, retained AE-001 as a parked conditional draft for provenance, and required a separately admitted reframe before any later protocol work.

Relationship to Verify

Verify remains the release-critical browser-local evidence-reading instrument. Its separate protected source-lineage decision does not gate or reactivate AE-001. Verify may read supplied matched paired-score rows only if some separately admitted and locked protocol produces that evidence shape.

A Verify reading would still not establish checkpoint identity, held-out status, run independence, frozen exclusions, evaluator blinding, complete failed-run reporting, or that the protocol preceded the observations. Those facts remain declared unless separately established.

Publication pathway

No paper or external publication

AE-001 has no result, manuscript, DOI, artifact release, arXiv submission, or Scholar index entry. The pathway below records capability boundaries and prerequisites only.

The intended flow is protocol and evidence record, reviewed publication package, explicit human approval, destination-specific handoff, returned identifier, and only then downstream indexing. A missing or invalid result remains publishable as a disposition; it is never silently removed from the record.

Continuous FunctionCanonical record onlyRole: Canonical question, protocol, evidence, deviation, and disposition record. Boundary: The present record is an unlocked protocol draft, not a paper, result, or durable receipt. Before this can move: Keep work, evidence, and validity states explicit and preserve negative-result parity..TinkerBlocked · prerequisites openRole: Possible execution provider and source of run, base-model, checkpoint, and user-metadata provenance. Boundary: Tinker is not the scholarly publisher or research authority. Private run IDs, paths, and credentials stay private by default. Before this can move: Locked protocol; Separate execution and budget authorization; Reviewed public/private provenance projection.ZenodoBlocked · prerequisites openRole: Potential DOI-bearing archive for a reviewed artifact package. Boundary: Publishing is an authenticated, consequential action and cannot be triggered by an agent or static public control. Before this can move: Admissible artifact bundle; Rights and privacy review; Named human approval of exact deposition bytes and metadata.Hugging FaceBlocked · prerequisites openRole: Potential versioned home for releasable models, adapters, datasets, code, and cards. Boundary: Weights and data need their own license, misuse, privacy, provenance, and release decisions; a paper result never authorizes them automatically. Before this can move: Artifact-specific release review; Model or dataset card; Human-approved authenticated push.arXivBlocked · prerequisites openRole: Potential preprint destination after Continuous Function prepares and validates a submission package. Boundary: arXiv expects registered authors to self-submit and may require endorsement and moderation. Its public API is for interoperability, not paper submission. Before this can move: Refereeable manuscript; Validated TeX/PDF package; Author account, license decision, final review, and self-submission.Google ScholarBlocked · prerequisites openRole: Possible downstream index of a public scholarly page and searchable paper. Boundary: Google Scholar is an index, not a publishing destination. Continuous Function can satisfy crawl and citation-metadata requirements but cannot guarantee inclusion. Before this can move: One public page per paper; Visible complete abstract; Searchable PDF; Highwire citation metadata and crawl access.

Continuous Function may eventually prepare exact bytes and metadata for Zenodo or Hugging Face, but authenticated publication remains a consequential human action. For arXiv, the product can prepare a validated author handoff; for Google Scholar, it can serve crawlable paper pages and citation metadata. Neither destination provides a general “publish this paper for me” interface.