# AI verification proposal

A proposal built with the Proposal Explorer of the AI Verification Tech Map (https://trustbutveri.fyi/), from its records of 2026-10-08. Interactive version: https://trustbutveri.fyi/explorer/?mechanisms=M-0001,M-0024&implementations=M-0001:I-0002

How to read it: a claim is something one party wants to verify about another's AI hardware or software. A mechanism is a general technique for verifying claims; it is "aimed at" a claim when that is its direct purpose, and "supporting" when it contributes without being aimed at it. A claim is addressed when a mechanism in the proposal is aimed at it and is not excluded by the filters; addressed does not mean verified, so check that mechanism's readiness and open flaws. Readiness levels R0 to R4 describe one record's public evidence for its assessed use and are never combined. Definitions: https://trustbutveri.fyi/about/methodology/ (roles, properties and flaws) and https://trustbutveri.fyi/about/readiness/ (readiness levels).

## Filters

Filters apply to mechanisms only and describe the setting the proposal is for.

None set. Every mechanism on the map was available.

## Overview

One row per mechanism, read from its record. Open flaws: critical / significant / minor. The last three columns are the editors' reading of what the verifier sees.

| Mechanism | Readiness | Prover | Attack testing | Hardware | Open flaws | Weights | Inputs and outputs | Training data |
| --- | --- | --- | --- | --- | --- | --- | --- | --- |
| Sampled inference recomputation / DiFR (Divergence From Reference) | R2 | Adversarial | Analysis | None | 0 / 2 / 1 | unspecified | unspecified | unspecified |
| Bounding unexplained information in outputs | R2 | Adversarial | Independent red-team | Retrofit device | 0 / 4 / 0 | depends | depends | not involved |

## Claims

No claims chosen.

## Mechanisms

### Sampled inference recomputation

DiFR checks that an inference provider ran its declared model by comparing output tokens or activations with a trusted re-run using the same random seed. ([Sampled inference recomputation](https://trustbutveri.fyi/mechanisms/sampled-inference-recomputation/))

- Assessment: selected implementation [DiFR (Divergence From Reference)](https://trustbutveri.fyi/implementations/difr/).
- Readiness: R2 Demonstrated, assessed for checking that outputs match the declared model, precision and sampling settings.
- Claims in this proposal: none of them.
- Threat model: adversarial prover. Hardware: none. Prover cooperation: required. Attack testing: analysis. Category: Cryptographic & computational.
- What the verifier sees: model weights unspecified; inputs and outputs unspecified; training data unspecified. This Explorer has no asset-specific exposure assessment for this implementation. Check its source and deployment assumptions.

### Bounding unexplained information in outputs

Limits the hidden information a facility's outputs can carry by measuring how much of those outputs the declared computation fails to predict. ([Bounding unexplained information in outputs](https://trustbutveri.fyi/mechanisms/bounding-unexplained-information/))

- Assessment: mechanism family.
- Readiness: R2 Demonstrated, assessed for bounding how much hidden information can leave in checked inference outputs.
- Claims in this proposal: none of them.
- Threat model: adversarial prover. Hardware: retrofit device. Prover cooperation: required. Attack testing: independent red-team. Category: Isolation & system architectures.
- What the verifier sees: model weights depends; inputs and outputs depends; training data not involved. Depends on where recomputation runs: in a sealed enclosure, or with zero-knowledge proofs, the verifier need not see the weights or the traffic.


## Properties

**Built for an adversarial prover**

- Sampled inference recomputation
- Bounding unexplained information in outputs

**No new hardware needed**

- Sampled inference recomputation


## Attack testing

Published attempts to break a system, including those that found failures. Testing history does not establish that open flaws are resolved.

**Testing history**

- Sampled inference recomputation / DiFR (Divergence From Reference): Analysis
- Bounding unexplained information in outputs: Independent red-team


## Limits

**Open significant flaws**

- Statistical tolerance leaves a covert channel (demonstrated attack, in DiFR (Divergence From Reference); https://trustbutveri.fyi/implementations/difr/#flaw-1) [2][7][9]. Statistical schemes can put an upper bound on an adversary's covert bandwidth, but cannot close it. In the companion exfiltration study, the detector cut exfiltratable information to under 0.5% under benign prompt traffic. It did not cut it to zero. Its authors called the channel's size under adversarial prompts an open empirical question. An independent study attacked that detector, which uses the same Gumbel-margin statistic. An adversary who controls the prompts roughly doubled the bits leaked per token. Across six models, this cut the slowdown from 146–254 times under benign prompts to 60–118 times. The attack widens the exfiltration bound. It does not target DiFR's check that outputs match the declared configuration.
- Speculative decoding and multi-model sampling not evaluated (open question, in DiFR (Divergence From Reference); https://trustbutveri.fyi/implementations/difr/#flaw-3) [1]. The algorithms and experiments cover sampling from a single LLM. Speculative decoding was not evaluated. The authors sketch an extension to one speculative-decoding algorithm, without experiments. They note that other variants would need modified verification and extra metadata.
- Prompt-controlled entropy inflation widens the covert channel (demonstrated attack, in Bounding unexplained information in outputs; https://trustbutveri.fyi/mechanisms/bounding-unexplained-information/#flaw-1) [2][7]. Gumbel-based inference verification tolerates token choices that honest GPU nondeterminism could produce, and the size of that tolerated set grows with the model's output entropy. Kezins, an independent researcher, showed that an adversary who controls the prompt distribution can raise output entropy and roughly double the bits leaked per token. Across six models of 1 to 32 billion parameters, this cut the slowdown from 146–254 times under benign prompts to 60–118 times. Kezins argues that architectures built on the same unexplained-information bound inherit this attack surface, and recommends calibrating tolerances against local token entropy rather than benign traffic.

  Related mechanism: Deterministic and bit-exact inference (R3, not in the proposal). Bit-exact replay would remove the tolerance for numerical noise that sets the size of this channel. The record notes that it needs full hardware and software metadata.
- Information the declared computation explains is not bounded (theoretical argument, in Bounding unexplained information in outputs; https://trustbutveri.fyi/mechanisms/bounding-unexplained-information/#flaw-2) [12][13]. The bound limits unexplained bits only. Outputs that the declared computation fully explains can still carry valuable information: a compression study notes that an adversary with inference access can extract more proprietary information per bit than naive transmission allows.
- Channels other than checked outputs are outside the bound (theoretical argument, in Bounding unexplained information in outputs; https://trustbutveri.fyi/mechanisms/bounding-unexplained-information/#flaw-3) [2][14]. The inference-verification scheme treats side channels as out of scope. A low-trust system design argues that suppressing physical covert bandwidth below kilobits per second is much more achievable than aiming for zero, and that a malicious device can leak one bit of information by deliberately outputting a wrong result.

  Related mechanism: Side-channel suppression for isolated facilities (R1, not in the proposal). Physical side channels need separate suppression, which is this mechanism's purpose.
- The facility-level design is untested (open question, in Bounding unexplained information in outputs; https://trustbutveri.fyi/mechanisms/bounding-unexplained-information/#flaw-4) [12]. The compute-verification architecture is described with protocol details, potential attacks and prototyping plans, but no prototype results have been published.

**Family finding context**

- Context for DiFR (Divergence From Reference); applicability depends on the finding's scope. Tolerance for numerical noise leaves a covert channel (demonstrated attack, in Sampled inference recomputation; https://trustbutveri.fyi/mechanisms/sampled-inference-recomputation/#flaw-1) [2][7][9]. Schemes that accept approximate matches can put an upper bound on an adversary's covert bandwidth, but they cannot close the channel. The weight-exfiltration detector cut exfiltratable information to under 0.5%, not to zero, on a 30-billion-parameter mixture-of-experts model under benign prompt traffic. Its authors called the channel's size under adversarial prompts an open empirical question. An independent study showed that an adversary who controls the prompts roughly doubles the bits leaked per token. Across six models, that cut the slowdown from 146–254 times under benign prompts to 60–118 times. The attack widens the exfiltration bound. It does not target the check that outputs match the declared model.

  Related mechanism: Deterministic and bit-exact inference (R3, not in the proposal). Bit-exact inference would remove the numerical tolerance if exact replay can be deployed with the required weights and configuration.
- Context for DiFR (Divergence From Reference); applicability depends on the finding's scope. Only recorded traffic is checked (theoretical argument, in Sampled inference recomputation; https://trustbutveri.fyi/mechanisms/sampled-inference-recomputation/#flaw-2) [2][10]. Recomputation checks that recorded, declared workloads are correct. It cannot show that the record is complete. The published schemes do not cover hidden workloads run on the same compute, or substituted work. Rinberg et al. say their exfiltration-detection scheme cannot stand alone.

  Related mechanism: Network taps and certifiers (R1, not in the proposal). Taps copy and hash all traffic on the monitored links, which bears on whether the traffic record is complete. They do not show what else ran on the same chips.
- Context for DiFR (Divergence From Reference); applicability depends on the finding's scope. Some inference optimizations are not covered (theoretical argument, in Sampled inference recomputation; https://trustbutveri.fyi/mechanisms/sampled-inference-recomputation/#flaw-3) [1][11]. TOPLOC's authors state that it cannot detect speculative decoding in which a cheaper model does the decoding. They did not test whether it distinguishes types of key-value (KV) cache compression. DiFR was evaluated only on sampling from a single model. Its authors sketch an extension to one speculative-decoding algorithm but do not test it.
- Context for DiFR (Divergence From Reference); applicability depends on the finding's scope. Mixed hardware widens the honest baseline (open question, in Sampled inference recomputation; https://trustbutveri.fyi/mechanisms/sampled-inference-recomputation/#flaw-4) [1]. When honest reference runs span different GPU types, the spread of benign scores grows. In DiFR's tests on Qwen3-30B-A3B, pooling A100 and H200 runs left Token-DiFR unable to separate the two smallest tested changes, a temperature of 1.1 instead of 1.0 and a simulated top-2 sampling bug, at the target false-positive rate, while cross-entropy separated them. Matched provider and verifier environments, or pooling that weights rare large deviations, restored detection.

**Open minor flaws**

- Mixed hardware widens the honest baseline (open question, in DiFR (Divergence From Reference); https://trustbutveri.fyi/implementations/difr/#flaw-2) [1]. For Qwen3-30B-A3B, pooling honest runs across A100 and H200 GPUs and parallelism setups broadened the honest score distribution. Token-DiFR then failed to separate the two smallest tested changes, a temperature of 1.1 instead of 1.0 and a simulated top-2 sampling bug, at the target false-positive rate, while cross-entropy did. The authors report that matched provider and verifier environments, or pooling that weights rare large deviations, restore detection.


## Possible additions

Mechanisms on the map, not in the proposal, that the records connect to an unaddressed or partly addressed claim, an open flaw or a dependency. Pointers, not recommendations: each brings its own readiness level and flaws, and none is claimed to close a flaw.

- **Deterministic and bit-exact inference** (R3 In production, assessed for reproducing open-model inference from receipts in Gensyn's information-market service)
  - Bears on the open significant flaw "Prompt-controlled entropy inflation widens the covert channel" in Bounding unexplained information in outputs. Bit-exact replay would remove the tolerance for numerical noise that sets the size of this channel. The record notes that it needs full hardware and software metadata.
  - Bounding unexplained information in outputs waits on it: Tolerance for numerical nondeterminism sets the size of the residual channel; bit-exact replay would remove it but needs full hardware and software metadata.
- **Side-channel suppression for isolated facilities** (R1 Proposed, assessed for bounding physical covert channels out of a verified enclosure)
  - Bears on the open significant flaw "Channels other than checked outputs are outside the bound" in Bounding unexplained information in outputs. Physical side channels need separate suppression, which is this mechanism's purpose.
  - Bounding unexplained information in outputs waits on it: Physical side channels need separate suppression, and one design treats a low residual bandwidth, rather than zero, as the realistic target.
- **Bandwidth limits and compartmentalization** (R2 Demonstrated, assessed for monitoring inter-node traffic with operator-run software on four GPUs)
  - Bounding unexplained information in outputs waits on it: The prover's compute must be isolated so that all traffic passes through the verifier's interlock; any unmonitored path voids the bound.
- **Network taps and certifiers** (R1 Proposed, assessed for committing a complete record of cluster traffic, so declared inference can be checked)
  - Bounding unexplained information in outputs depends on it.


## Dependencies

**Missing prerequisites**

- Bandwidth limits and compartmentalization (R2 Demonstrated, assessed for monitoring inter-node traffic with operator-run software on four GPUs), needed by Bounding unexplained information in outputs
- Side-channel suppression for isolated facilities (R1 Proposed, assessed for bounding physical covert channels out of a verified enclosure), needed by Bounding unexplained information in outputs
- Network taps and certifiers (R1 Proposed, assessed for committing a complete record of cluster traffic, so declared inference can be checked), needed by Bounding unexplained information in outputs

**Blockers**

- Sampled inference recomputation: The verifier needs the model weights, so outsiders cannot use the method to verify providers of closed-weights models. (privacy & leakage) [1]
- Sampled inference recomputation: The verifier must know and match the provider's sampling procedure, and in one prototype a sampling mismatch in a newer vLLM version produced large spurious logit differences. (performance & compatibility) [1][4]
- Sampled inference recomputation: No independent red-team of DiFR's consistency check has been published, Amodo rates recomputation red-teaming 'not started', and the one independent attack study targets an exfiltration detector built on the same statistic. (adversarial validation) [6][7]
- Bounding unexplained information in outputs: The prover's compute must be isolated so that all traffic passes through the verifier's interlock; any unmonitored path voids the bound. (coverage & hidden compute; waits on Bandwidth limits and compartmentalization) [12]
- Bounding unexplained information in outputs: Physical side channels need separate suppression, and one design treats a low residual bandwidth, rather than zero, as the realistic target. (coverage & hidden compute; waits on Side-channel suppression for isolated facilities) [2][14]
- Bounding unexplained information in outputs: Tolerance for numerical nondeterminism sets the size of the residual channel; bit-exact replay would remove it but needs full hardware and software metadata. (protocol soundness; waits on Deterministic and bit-exact inference) [7][14]
- Bounding unexplained information in outputs: Recomputation over confidential weights and inputs needs a protected setting: prover recomputation in a verifier-controlled enclosure, verifier recomputation in a prover-controlled enclosure, or zero-knowledge proofs. (privacy & leakage) [12]
- Bounding unexplained information in outputs: No prototype of the facility-level architecture exists to red-team. (adversarial validation) [12]


## What the verifier sees

- Model weights: shown by none; depends on the design for Bounding unexplained information in outputs; hidden by none; not involved in none; unspecified for Sampled inference recomputation.
- Inputs and outputs: shown by none; depends on the design for Bounding unexplained information in outputs; hidden by none; not involved in none; unspecified for Sampled inference recomputation.
- Training data: shown by none; depends on the design for none; hidden by none; not involved in Bounding unexplained information in outputs; unspecified for Sampled inference recomputation.

## Implementations

- Sampled inference recomputation: [AI 2040 inference-only verification stack](https://trustbutveri.fyi/implementations/ai-2040-inference-only-verification-plan/) (R1, proposed architecture); [DiFR (Divergence From Reference)](https://trustbutveri.fyi/implementations/difr/) (R2, research prototype); [Low-trust AI compute verification system overview](https://trustbutveri.fyi/implementations/low-trust-compute-verification-system-overview/) (R1, proposed architecture); [SASH confidential network logger](https://trustbutveri.fyi/implementations/sash-confidential-network-logger/) (R1, research prototype); [TOPLOC](https://trustbutveri.fyi/implementations/toploc/) (R3, open-source project)
- Bounding unexplained information in outputs: none on the map

## Sources

1. DiFR: Inference Verification Despite Nondeterminism, A. Karvonen et al. (2025). https://arxiv.org/abs/2511.20621
2. Verifying LLM Inference to Detect Model Weight Exfiltration, R. Rinberg et al. (2025). https://arxiv.org/abs/2511.02620
3. adamkarvonen/difr (GitHub repository), A. Karvonen (2025). https://github.com/adamkarvonen/difr
4. Scaling Recomputation Inference Verification, Amodo Design (2026). https://amododesign.com/notes/2026-09-02-scaling-recomputation-inference-verification/
5. Amodo-Design/Inference-Recomputation-Prototype (GitHub repository), Amodo Design (2026). https://github.com/Amodo-Design/Inference-Recomputation-Prototype
6. AI 2040 Plan A — Verification SITREP, Amodo Design (2026). https://amododesign.com/ai-verification/plan-a-sitrep/
7. Adversarial Entropy Inflation Against Gumbel-Based Inference Verification, N. Kezins (2026). https://arxiv.org/abs/2608.23375
8. An Inference Verification Prototype — Stage 1, Amodo Design (2026). https://amododesign.com/notes/2026-06-29-inference-verification-prototype/
9. Bit-Exact AI Inference Verification Without Performance Tradeoffs, N. Cankaya (2026). https://arxiv.org/abs/2606.00279
10. Example Schemes for Verifying High-Stakes AI Agreements, Amodo Design (2026). https://amododesign.com/notes/2026-06-23-verification-algorithms/
11. TOPLOC: A Locality Sensitive Hashing Scheme for Trustless Verifiable Inference, J. M. Ong et al. (2025). https://proceedings.mlr.press/v267/ong25a.html
12. Verifying AI Compute by Bounding Unexplained Information Exfiltration, J. Petrie & Y. Mühlhäuser (2026). https://openreview.net/forum?id=qtgG5HZSsk
13. Haiku to Opus in Just 10 bits: LLMs Unlock Large Compression Gains, R. Rinberg et al. (2026). https://arxiv.org/abs/2604.02343
14. A System Overview for Near-Term, Low-Trust AI Compute Verification, N. Cankaya (2026). https://intelligence.org/wp-content/uploads/2026/06/A-system-overview-for-near-term-low-trust-AI-compute-verification.pdf