# AI verification proposal

A proposal built with the Proposal Explorer of the AI Verification Tech Map (https://trustbutveri.fyi/), from its records of 2026-10-09. Interactive version: https://trustbutveri.fyi/explorer/?mechanisms=M-0004,M-0024&implementations=M-0004:I-0012

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 development status, security evidence and findings. Definitions: https://trustbutveri.fyi/about/methodology/ (roles, properties and findings) and https://trustbutveri.fyi/about/readiness/ (development status).

## 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 failures: critical / significant / minor. The last three columns are the editors' reading of what the verifier sees. Findings are grouped as known failures, scope limitations and open questions. Only known failures count as failures. Counts are an inventory of published findings, not a risk score.

| Mechanism | Development | Security evidence | Prover | Attack testing | Hardware | Open failures | Weights | Inputs and outputs | Training data |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| Zero-knowledge proofs of inference / Low-trust AI compute verification system overview | Proposed | Published security analysis | Adversarial | Analysis | Retrofit device | 0 / 0 / 1 | unspecified | unspecified | unspecified |
| Bounding unexplained information in outputs | Research demonstration | Published attack testing | Adversarial | Independent red-team | Retrofit device | 0 / 1 / 0 | depends | depends | not involved |

## Claims

No claims chosen.

## Mechanisms

### Zero-knowledge proofs of inference

A reference design that existing data centres could add: network taps hash all traffic, and air-gapped, independently sourced checkers later re-run randomly chosen records. ([Zero-knowledge proofs of inference](https://trustbutveri.fyi/mechanisms/zk-proofs-of-inference/))

- Assessment: selected implementation [Low-trust AI compute verification system overview](https://trustbutveri.fyi/implementations/low-trust-compute-verification-system-overview/).
- Development: Proposed (legacy code R1), assessed for screening challenged records to show declared inference compute is not training.
- Security evidence: Published security analysis. Independent evaluation: unassessed. Formal proof: unassessed. Deployment assurance: unassessed.
- Claims in this proposal: none of them.
- Threat model: adversarial prover. Hardware: retrofit device. Prover cooperation: required. Attack testing: analysis. Category: Isolation & system architectures.
- 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.
- Development: Research demonstration (legacy code R2), assessed for bounding how much hidden information can leave in checked inference outputs.
- Security evidence: Published attack testing. Independent evaluation: unassessed. Formal proof: unassessed. Deployment assurance: unassessed.
- 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**

- Zero-knowledge proofs of inference
- Bounding unexplained information in outputs


## Attack testing

Attack testing records published testing for this use. It does not by itself show independent review, a formal proof or that a deployed system is secure.

**Testing history**

- Zero-knowledge proofs of inference / Low-trust AI compute verification system overview: Analysis
- Bounding unexplained information in outputs: Independent red-team


## Limits

**Open significant failures**

- Prompt-controlled entropy inflation widens the covert channel (known failure, demonstrated attack, in Bounding unexplained information in outputs; https://trustbutveri.fyi/mechanisms/bounding-unexplained-information/evidence/flaws/1/) [10][11]. 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.

**Family finding context**

- Context for Low-trust AI compute verification system overview. Findings from the mechanism family appear here as context. They apply to an implementation only when its own record lists them, under the conditions stated there. The proof covers a fixed-point approximation, not the floating-point model (scope limitation, open question, in Zero-knowledge proofs of inference; https://trustbutveri.fyi/mechanisms/zk-proofs-of-inference/evidence/flaws/1/) [1][3][4][5]. Current ZK inference systems prove a quantised version of the network. zkLLM scales values by 2^16 and reports small perplexity changes. Attestable reports quantising matrix multiplications to 8-bit integers while proving other operations in floating point. A verifier therefore learns about the proof-friendly variant, and must separately accept that this variant is the declared model. Trail of Bits built a ResNet-18 backdoor that is dormant in the full-precision model and active after ezkl's quantisation; whether it persists through proving was left for further investigation. A verification system design notes that ZKPs can emulate floating-point operations. Rounding makes floating-point results depend on summation order, so bit-for-bit replay of an accelerator's results needs its original reduction tree. The report calls emulating that tree inside a ZKP an open, intricate problem and asks what it would cost.
- Context for Low-trust AI compute verification system overview. Findings from the mechanism family appear here as context. They apply to an implementation only when its own record lists them, under the conditions stated there. A proof speaks only for the computations that were proven (scope limitation, theoretical argument, in Zero-knowledge proofs of inference; https://trustbutveri.fyi/mechanisms/zk-proofs-of-inference/evidence/flaws/2/) [6]. Attestable writes that "a proof of some computation is not a proof of all computation", and that a proof cannot discover a datacenter that was never declared. Proofs of inference do not by themselves show that no other workload ran on the same or other hardware.

  Related mechanism: Proofs of useful work for capacity accounting (R1, not in the proposal). The record names proof-of-work accounting as the kind of compute accounting needed to show that proven inference was the only work done.
- Context for Low-trust AI compute verification system overview. Findings from the mechanism family appear here as context. They apply to an implementation only when its own record lists them, under the conditions stated there. The model architecture is disclosed (scope limitation, theoretical argument, in Zero-knowledge proofs of inference; https://trustbutveri.fyi/mechanisms/zk-proofs-of-inference/evidence/flaws/3/) [3][7]. ZKML "requires that the model architecture (but not weights) is revealed", and zkLLM assumes a publicly known model structure. Architecture can be commercially sensitive.
- Context for Low-trust AI compute verification system overview. Findings from the mechanism family appear here as context. They apply to an implementation only when its own record lists them, under the conditions stated there. Proofs do not bind computational effort (Hollow-LLM) (scope limitation, demonstrated attack, in Zero-knowledge proofs of inference; https://trustbutveri.fyi/mechanisms/zk-proofs-of-inference/evidence/flaws/4/) [8]. Researchers at the University of Southern California show that a proof of inference certifies that an output is consistent with committed weights under the declared architecture, but not how much computation produced it. In their Hollow-LLM attack, a provider keeps the declared architecture and parameter count but commits to "ghost weights". Some layers pass their inputs through unchanged, and wide layers carry the signal in a small subspace, so a much smaller inner model does the real work. The ghost weights satisfy the verification circuit and yield valid proofs.

  The authors ran the attack with the proof procedure of zkGPT, a separate ZK inference system, on a 6-layer, 512-dimensional transformer declared as up to 12 layers and 1,024 dimensions. Outputs were identical to the inner model's, and serving cost stayed at the inner model's level. An honest model of the declared size cost 2.4 times as much to prefill and 3.1 times as much to decode. Proving cost still grew with the declared architecture.

  The authors note that results may be served before any proof, with the provider building the witness only when a call is selected for audit. They describe their constructions as "compatible with state-of-the-art zkLLM pipelines", and state that the attack does not imply a flaw in the proof system itself. They propose challenge-based audits and ablation tests, which raise the cost of cheating but give no guarantee.

**Open minor failures**

- Deliberate faults leak a bit each (known failure, theoretical argument, in Low-trust AI compute verification system overview; https://trustbutveri.fyi/implementations/low-trust-compute-verification-system-overview/evidence/flaws/2/) [1]. A malicious device can leak one bit by deliberately outputting a wrong result, which blocks a disclosure when the cross-comparison fails. The design therefore needs a pre-agreed budget of tolerated faults.

**Scope limitations**

- Mismatches cannot be attributed to cheating or error (scope limitation, theoretical argument, in Low-trust AI compute verification system overview; https://trustbutveri.fyi/implementations/low-trust-compute-verification-system-overview/evidence/flaws/1/) [1]. A failed hash or replay does not show whether it came from an evasion attempt, a random bit flip or an evaluation error. The author notes that if detected anomalies can plausibly be waved off as malfunctions, deterrence becomes less effective, so the parties need an agreed escalation procedure that ends in attribution.
- Information the declared computation explains is not bounded (scope limitation, theoretical argument, in Bounding unexplained information in outputs; https://trustbutveri.fyi/mechanisms/bounding-unexplained-information/evidence/flaws/2/) [9][12]. 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 (scope limitation, theoretical argument, in Bounding unexplained information in outputs; https://trustbutveri.fyi/mechanisms/bounding-unexplained-information/evidence/flaws/3/) [1][10]. 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.

**Open questions**

- Inspector agents may be manipulable (open question, open question, in Low-trust AI compute verification system overview; https://trustbutveri.fyi/implementations/low-trust-compute-verification-system-overview/evidence/flaws/3/) [1]. Automated compliance screening with LLM-based inspector agents must resist prompt-injection attacks. Adversarially trained systems might hide malicious actions with steganography, which makes backdoor detection an open problem.
- The facility-level design is untested (open question, open question, in Bounding unexplained information in outputs; https://trustbutveri.fyi/mechanisms/bounding-unexplained-information/evidence/flaws/4/) [9]. The compute-verification architecture is described with protocol details, potential attacks and prototyping plans, but no prototype results have been published.

**Not yet demonstrated**

- Zero-knowledge proofs of inference: Proposed (legacy code R1), assessed for screening challenged records to show declared inference compute is not training


## Possible additions

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

- **Deterministic and bit-exact inference** (Operational use (legacy code R3), assessed for reproducing open-model inference from receipts in Gensyn's information-market service)
  - Bears on the open significant failure "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.
  - Zero-knowledge proofs of inference waits on it: Exact replay needs complete hardware and software metadata, and the tolerable slowdown from emulation is an open question.
  - 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.
- **Bandwidth limits and compartmentalization** (Research demonstration (legacy code R2), 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.
- **Tamper evidence for verifier devices** (Research demonstration (legacy code R2), assessed for detecting probing of proposed verifier hardware, using server and electronics prototypes as evidence)
  - Zero-knowledge proofs of inference waits on it: Tamper-evident, rapidly mass-manufacturable and retrofittable enclosures for side-channel defence are an open research question, and physical security against covert communication in every monitored data centre is challenging.
- **Timed challenge-response and memory-occupation challenges** (Research demonstration (legacy code R2), assessed for detecting whether a GPU is doing other work)
  - Zero-knowledge proofs of inference waits on it: Distinguishing one server's DRAM contents from another's by challenge-response timing, and a general challenge-response protocol for diverse data types, are open.
- **Network taps and certifiers** (Proposed (legacy code R1), assessed for committing a complete record of cluster traffic, so declared inference can be checked)
  - Zero-knowledge proofs of inference waits on it: Empirical feasibility of passive optical splitting at 53–112 GBaud under realistic conditions is an open question.
  - Bounding unexplained information in outputs depends on it.
- **Side-channel suppression for isolated facilities** (Proposed (legacy code R1), assessed for bounding physical covert channels out of a verified enclosure)
  - Zero-knowledge proofs of inference waits on it: A mass-manufacturable, good-enough side-channel defence, particularly power-line filtering, has not been constructed or red-teamed.
  - 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.
- **Sampled inference recomputation** (Operational use (legacy code R3), assessed for checking untrusted workers' activations against the declared model, prompt and precision)
  - Bounding unexplained information in outputs depends on it.


## Dependencies

**Missing prerequisites**

- Network taps and certifiers (Proposed (legacy code R1), assessed for committing a complete record of cluster traffic, so declared inference can be checked), needed by Zero-knowledge proofs of inference and Bounding unexplained information in outputs
- Deterministic and bit-exact inference (Operational use (legacy code R3), assessed for reproducing open-model inference from receipts in Gensyn's information-market service), needed by Zero-knowledge proofs of inference
- Sampled inference recomputation (Operational use (legacy code R3), assessed for checking untrusted workers' activations against the declared model, prompt and precision), needed by Bounding unexplained information in outputs
- Bandwidth limits and compartmentalization (Research demonstration (legacy code R2), 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 (Proposed (legacy code R1), assessed for bounding physical covert channels out of a verified enclosure), needed by Bounding unexplained information in outputs

**Shared foundations**

- Network taps and certifiers, relied on by Zero-knowledge proofs of inference and Bounding unexplained information in outputs
- Deterministic and bit-exact inference, relied on by Zero-knowledge proofs of inference and Bounding unexplained information in outputs
- Side-channel suppression for isolated facilities, relied on by Zero-knowledge proofs of inference and Bounding unexplained information in outputs

**Blockers**

- Zero-knowledge proofs of inference: Empirical feasibility of passive optical splitting at 53–112 GBaud under realistic conditions is an open question. (performance & compatibility; waits on Network taps and certifiers) [1]
- Zero-knowledge proofs of inference: Exact replay needs complete hardware and software metadata, and the tolerable slowdown from emulation is an open question. (performance & compatibility; waits on Deterministic and bit-exact inference) [1]
- Zero-knowledge proofs of inference: Tamper-evident, rapidly mass-manufacturable and retrofittable enclosures for side-channel defence are an open research question, and physical security against covert communication in every monitored data centre is challenging. (hardware trust; waits on Tamper evidence for verifier devices) [1]
- Zero-knowledge proofs of inference: A mass-manufacturable, good-enough side-channel defence, particularly power-line filtering, has not been constructed or red-teamed. (coverage & hidden compute; waits on Side-channel suppression for isolated facilities) [1]
- Zero-knowledge proofs of inference: Distinguishing one server's DRAM contents from another's by challenge-response timing, and a general challenge-response protocol for diverse data types, are open. (coverage & hidden compute; waits on Timed challenge-response and memory-occupation challenges) [1]
- Zero-knowledge proofs of inference: The threat model is under-developed and needs input from cybersecurity and AI threat-modelling experts. (adversarial validation) [1]
- 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) [9]
- 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) [1][10]
- 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) [1][11]
- 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) [9]
- Bounding unexplained information in outputs: No prototype of the facility-level architecture exists to red-team. (adversarial validation) [9]


## 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 Zero-knowledge proofs of inference.
- 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 Zero-knowledge proofs of inference.
- Training data: shown by none; depends on the design for none; hidden by none; not involved in Bounding unexplained information in outputs; unspecified for Zero-knowledge proofs of inference.

## Implementations

- Zero-knowledge proofs of inference: [Attestable zero-knowledge inference prover](https://trustbutveri.fyi/implementations/attestable-zk-inference/) (R1, product); [EZKL](https://trustbutveri.fyi/implementations/ezkl/) (R2, product); [Low-trust AI compute verification system overview](https://trustbutveri.fyi/implementations/low-trust-compute-verification-system-overview/) (R1, proposed architecture); [zkLLM](https://trustbutveri.fyi/implementations/zkllm/) (R2, research prototype)
- Bounding unexplained information in outputs: none on the map

## Sources

1. 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
2. Fingerprinting All AI Cluster I/O Without Mutually Trusted Processors, N. Cankaya et al. (2026). https://arxiv.org/abs/2606.10724
3. zkLLM: Zero Knowledge Proofs for Large Language Models, H. Sun et al. (2024). https://doi.org/10.1145/3658644.3670334
4. Proving LLMs at Scale, Attestable (2026). https://attestable.com/blog/proving-llms-scale
5. Zkonduit EZKL Security Assessment, F. Casal et al. (2025). https://github.com/trailofbits/publications/blob/master/reviews/2025-03-zkonduit-ezkl-securityreview.pdf
6. Pacing AI Requires Proof, Attestable (2026). https://attestable.com/blog/pacing-ai-requires-proof
7. ZKML: An Optimizing System for ML Inference in Zero-Knowledge Proofs, B.-J. Chen et al. (2024). https://doi.org/10.1145/3627703.3650088
8. Hollow-LLM Attack: Computationally Trivial Weights in Zero-Knowledge Verification of LLM Inference, C. Gong et al. (2026). https://arxiv.org/abs/2607.28884
9. Verifying AI Compute by Bounding Unexplained Information Exfiltration, J. Petrie & Y. Mühlhäuser (2026). https://openreview.net/forum?id=qtgG5HZSsk
10. Verifying LLM Inference to Detect Model Weight Exfiltration, R. Rinberg et al. (2025). https://arxiv.org/abs/2511.02620
11. Adversarial Entropy Inflation Against Gumbel-Based Inference Verification, N. Kezins (2026). https://arxiv.org/abs/2608.23375
12. Haiku to Opus in Just 10 bits: LLMs Unlock Large Compression Gains, R. Rinberg et al. (2026). https://arxiv.org/abs/2604.02343