# 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-0012&implementations=M-0012:I-0007

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 |
| --- | --- | --- | --- | --- | --- | --- | --- | --- |
| Model identity attestation / Attestable Audits | R2 | Semi-trusted | Analysis | Existing features | 0 / 2 / 1 | unspecified | unspecified | unspecified |

## Claims

No claims chosen.

## Mechanisms

### Model identity attestation

A research prototype that runs AI safety benchmarks inside a trusted execution environment and publishes attestations binding the model, the audit and the results. ([Model identity attestation](https://trustbutveri.fyi/mechanisms/model-identity-attestation/))

- Assessment: selected implementation [Attestable Audits](https://trustbutveri.fyi/implementations/attestable-audits/).
- Readiness: R2 Demonstrated, assessed for showing users that the model answering them is the audited one.
- Claims in this proposal: none of them.
- Threat model: semi-trusted prover. Hardware: existing features. Prover cooperation: required. Attack testing: analysis. Category: On-chip & hardware-enabled.
- 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.


## Properties

**No new hardware needed**

- Model identity attestation


## 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**

- Model identity attestation / Attestable Audits: Analysis


## Limits

**Open significant flaws**

- Relies on the TEE vendor and inherits TEE attacks (theoretical argument, in Attestable Audits; https://trustbutveri.fyi/implementations/attestable-audits/#flaw-1) [1]. Inherited finding. The prototype trusts AWS Nitro, not the Intel TDX or AMD SEV-SNP attestation roots targeted by the cited confidential-VM studies. Those studies are class context, not a demonstrated attack on this Nitro prototype. The design depends on trusting the TEE vendor, AWS in the prototype. The authors cite memory-aliasing, ciphertext side-channel and malicious-interrupt attacks on confidential VMs (BadRAM, CIPHERLEAKS, Heckler). Their answer is to revoke vulnerable base images once such attacks are discovered. Related finding: https://trustbutveri.fyi/mechanisms/tee-remote-attestation/#flaw-5.

  Response: The authors propose revoking vulnerable base images; they do not report a red-team evaluation of the prototype.
- Prompt-based model exfiltration is a residual gap (open question, in Attestable Audits; https://trustbutveri.fyi/implementations/attestable-audits/#flaw-2) [1]. The authors state that "prompt-based model exfiltration during the user interaction step remains a residual gap".

**Family finding context**

- Context for Attestable Audits; applicability depends on the finding's scope. Underlying attestation can be forged or relayed (demonstrated attack, in Model identity attestation; https://trustbutveri.fyi/mechanisms/model-identity-attestation/#flaw-1) [3][4][5][6][7][8]. Inherited finding. Critical for the enclave route against an operator with physical access to affected hardware, or control of an unpatched SEV-SNP hypervisor. It does not apply to the recomputation route. PAL*M excludes physical attacks, and Tinfoil acknowledges this boundary. The enclave route inherits the platform-specific TEE attestation failures. Intel TDX forgery and H100 relay were demonstrated with physical access and host control. Battering RAM defeated AMD SEV-SNP attestation on DDR4 servers; RMPocalypse did so from malicious host software on platforms without AMD's fixes. These demonstrate failures of the trust roots, not of each model-commitment protocol. Related finding: https://trustbutveri.fyi/mechanisms/tee-remote-attestation/#flaw-1.

  Response: The TEE.fail authors report that physical interposer attacks are outside Intel's and AMD's threat models. AMD reports fixes for RMPocalypse.

  Related mechanism: Hardware-enabled guarantees (flexHEG) and guarantee processors (R1, not in the proposal). A tamper-protected enclosure around the chip is the proposed answer when the party that holds the hardware may attack it physically.
- Context for Attestable Audits; applicability depends on the finding's scope. Launch-state attestation does not by itself cover weights loaded later (theoretical argument, in Model identity attestation; https://trustbutveri.fyi/mechanisms/model-identity-attestation/#flaw-2) [9][10]. Attestation measures launch state, and weights are read from disk after boot. A signature checked at load time does not stop a malicious hypervisor from altering the disk afterwards. Tinfoil reports mitigating this with dm-verity checks on every read. Unmeasured runtime configuration remains a general risk.
- Context for Attestable Audits; applicability depends on the finding's scope. For private models, a user can confirm consistency but not content (open question, in Model identity attestation; https://trustbutveri.fyi/mechanisms/model-identity-attestation/#flaw-3) [1][9]. When weights are not published, users can check that the same root hash is served each time, but not what the model is. Pairing the hash with an attested evaluation, as in Attestable Audits, is one proposed remedy.
- Context for Attestable Audits; applicability depends on the finding's scope. Recomputation depends on trusted logging and randomness, and its tolerance leaves a covert channel (demonstrated attack, in Model identity attestation; https://trustbutveri.fyi/mechanisms/model-identity-attestation/#flaw-4) [11][12]. The recomputation variant assumes that every input, output and seed is logged correctly, and that the attacker can neither predict nor manipulate which messages are sampled for verification. Legitimate nondeterminism concentrates at a few token positions, and slow leaks within the tolerated slack remain possible. An independent study showed that an adversary who controls the prompts roughly doubles the bits leaked per token, reducing the exfiltration slowdown from 146–254 times under benign prompts to 60–118 times. The attack targets the exfiltration bound, not the check that outputs match the declared model.

  Related mechanism: Network taps and certifiers (R1, not in the proposal). Taps are proposed to copy and hash traffic on the monitored links, reducing reliance on the prover's own log. This still depends on the monitored boundary and trusted capture.

  Related mechanism: Deterministic and bit-exact inference (R3, not in the proposal). Bit-exact inference would remove the numerical tolerance that leaves this channel.

**Open minor flaws**

- CPU-only enclaves force small, quantized models and high cost (open question, in Attestable Audits; https://trustbutveri.fyi/implementations/attestable-audits/#flaw-3) [1][2]. Memory limits required 4-bit quantization, and the quantized model scored 51.4% on zero-shot MMLU. CPU inference cost 21.7 times as much per token as GPU inference, and the enclave roughly doubled the CPU cost. The authors wrote that H100 confidential computing had no multi-GPU support. NVIDIA's white paper of August 2025 describes a protected-PCIe mode that passes all eight GPUs of a Hopper HGX node to one confidential VM, with NVLink traffic unencrypted.


## 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.

- **TEE remote attestation for AI workloads** (R3 In production, assessed for showing which software ran to a party that distrusts the operator holding the hardware)
  - Model identity attestation waits on it: The prototype needs porting to GPU confidential computing to handle larger models; the authors expect an overhead as small as 5 times there.


## Dependencies

**Missing prerequisites**

- TEE remote attestation for AI workloads (R3 In production, assessed for showing which software ran to a party that distrusts the operator holding the hardware), needed by Model identity attestation

**Blockers**

- Model identity attestation: The prototype needs porting to GPU confidential computing to handle larger models; the authors expect an overhead as small as 5 times there. (performance & compatibility; waits on TEE remote attestation for AI workloads) [1]
- Model identity attestation: As of September 2026 no code has been released for the prototype. (adversarial validation) [1]


## What the verifier sees

- Model weights: shown by none; depends on the design for none; hidden by none; not involved in none; unspecified for Model identity attestation.
- Inputs and outputs: shown by none; depends on the design for none; hidden by none; not involved in none; unspecified for Model identity attestation.
- Training data: shown by none; depends on the design for none; hidden by none; not involved in none; unspecified for Model identity attestation.

## Implementations

- Model identity attestation: [Attestable Audits](https://trustbutveri.fyi/implementations/attestable-audits/) (R2, research prototype); [PAL*M](https://trustbutveri.fyi/implementations/palm/) (R2, research prototype); [Tinfoil model identity (Modelwrap)](https://trustbutveri.fyi/implementations/tinfoil-model-identity/) (R3, product)

## Sources

1. Attestable Audits: Verifiable AI Safety Benchmarks Using Trusted Execution Environments, C. Schnabl et al. (2025). https://arxiv.org/abs/2506.23706
2. NVIDIA Secure AI with Blackwell and Hopper GPUs (White Paper), NVIDIA (2025). https://docs.nvidia.com/nvidia-secure-ai-with-blackwell-and-hopper-gpus-whitepaper.pdf
3. TEE.fail: Breaking Trusted Execution Environments via DDR5 Memory Bus Interposition, J. Chuang et al. (2026). https://tee.fail/
4. Battering RAM: Low-Cost Interposer Attacks on Confidential Computing via Dynamic Memory Aliasing, J. De Meulemeester et al. (2026). https://batteringram.eu/
5. RMPocalypse: How a Catch-22 Breaks AMD SEV-SNP, B. Schlüter & S. Shinde (2025). https://rmpocalypse.github.io/
6. SEV-SNP RMP Initialization Vulnerability (AMD-SB-3020), AMD (2025). https://www.amd.com/en/resources/product-security/bulletin/amd-sb-3020.html
7. A primer on secure enclaves, Tinfoil (2026). https://docs.tinfoil.sh/verification/secure-enclave-primer
8. PAL*M: Property Attestation for Large Generative Models, P. Chantasantitam et al. (2026). https://arxiv.org/abs/2601.16199
9. How Tinfoil Proves Exactly What Model Is Running, Tinfoil Team (2026). https://tinfoil.sh/blog/2026-02-03-proving-model-identity
10. On TEEs for Privacy-Preserving Monitoring in AI Governance, Gloria Z (2026). https://techgov.intelligence.org/blog/on-tees-for-privacy-preserving-monitoring-in-ai-governance
11. Verifying LLM Inference to Detect Model Weight Exfiltration, R. Rinberg et al. (2025). https://arxiv.org/abs/2511.02620
12. Adversarial Entropy Inflation Against Gumbel-Based Inference Verification, N. Kezins (2026). https://arxiv.org/abs/2608.23375