Proposal Explorer

Add claims to see which ones the mechanisms address.

Operational use

• Operational use Hardware-attested weight binding for hardware-attested weight binding showing users that a service runs its committed weights

Built for an adversarial prover

Bandwidth limits and compartmentalization

No new hardware needed

Hardware-attested weight binding

Failures since mitigated

• Launch-state attestation does not by itself cover weights loaded later in Hardware-attested weight binding ^{8 18}

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.

• Bandwidth limits and compartmentalization: Analysis
• Hardware-attested weight binding: Independent red-team

Open critical failures

• Underlying attestation can be forged or relayed in Hardware-attested weight binding

  Critical for weight binding against an operator with physical access to affected hardware, or with control of the hypervisor on an AMD SEV-SNP platform without AMD's fixes. 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. ^{9 10 14 15 16 17}

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

  Known failure · Demonstrated attack · Critical · Open · Inherited finding. On the record · Related finding in TEE remote attestation for AI workloads

  Related mechanism Proposed Hardware-enabled guarantees (flexHEG) and guarantee processors: A tamper-protected enclosure around the chip is the proposed answer when the party that holds the hardware may attack it physically. A pointer, not evidence that this failure is mitigated. Add

Open significant failures

2 failures in 1 mechanism

• Operator control of pod routing collapses the bound in Bandwidth limits and compartmentalization

  Lucid's analysis finds that if the operator can freely assign pods to routers, it could dedicate a whole cell of 100 or more pods to one pipeline stage. The bound then falls to about 90–220x uncompressed and as low as about 25x with compression. The proposed mitigation, auditor-controlled random assignment that is periodically re-randomized, has not been implemented. ²

  Known failure · Theoretical argument · Significant · Open. On the record

• Parallel scale-up switches are hard enforcement points in Bandwidth limits and compartmentalization

  In GB200 topologies, GPUs reach GPUs in other nodes through NVSwitches without a NIC on the path. Amodo notes that limits are hard to enforce there because many switches work in parallel, so compromising one or two would bypass the limit. ⁵

  Known failure · Theoretical argument · Significant · Open. On the record

Scope limitations

• Undeclared local storage raises per-pod capacity in Bandwidth limits and compartmentalization

  More memory or storage per pod helps an adversary. Lucid requires per-pod storage to be declared, capped and physically inspected. ²

  Scope limitation · Theoretical argument. On the record

• Training within one pod is not covered in Bandwidth limits and compartmentalization

  Lucid's bounds concern pre-training models larger than the pods are sized for. Training models that fit in one pod, fine-tuning and reinforcement-learning post-training within one pod are outside the modelled threat. ²

  Scope limitation · Open question. On the record

• For private models, a user can confirm consistency but not content in Hardware-attested weight binding

  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. ^{8 19}

  Scope limitation · Open question. On the record

Open questions

• Low-communication training reduces the bandwidth training needs in Bandwidth limits and compartmentalization

  DiLoCo matched fully synchronous training on 8 workers while communicating 500 times less. Rahman writes that this family of methods theoretically allows large-scale training with less than 100 Mbps. Lucid includes these methods in its bounds, but notes that extreme activation compression, architectures with unusually small inter-layer widths, or modular paradigms could erode the margin. ^{2 6 7}

  Open question · Theoretical argument. On the record

Mechanisms on the map that are not in the proposal. Pointers, not recommendations. ?

• Operational use TEE remote attestation for AI workloads Add
  • Hardware-attested weight binding waits on it. Attestation that resists physical attackers, for the enclave variant.
• Research demo Tamper evidence for verifier devices Add
  • Bandwidth limits and compartmentalization waits on it. Shaping devices and routing assignments must be trusted by both parties; Amodo has not yet fully analysed resilience to a compromised DPU.
• Proposed Network taps and certifiers Add
  • Bandwidth limits and compartmentalization waits on it. The verifier must know that all traffic leaving a pod crosses the capped, monitored links.
• Proposed Hardware-enabled guarantees (flexHEG) and guarantee processors Add

  Excluded by filters: needs new chips

  • Bears on the open critical failure “Underlying attestation can be forged or relayed” in Hardware-attested weight binding. A tamper-protected enclosure around the chip is the proposed answer when the party that holds the hardware may attack it physically.

Missing prerequisites

• Research demo Tamper evidence for verifier devices needed by Bandwidth limits and compartmentalization Add
• Operational use TEE remote attestation for AI workloads needed by Hardware-attested weight binding Add

Blockers

5 blockers recorded

• Bandwidth limits and compartmentalization
  • No cap that a verifier can check has been implemented or red-teamed. Adversarial validation ²
  • The verifier must know that all traffic leaving a pod crosses the capped, monitored links. Coverage & hidden compute. Waits on Network taps and certifiers ³
  • Shaping devices and routing assignments must be trusted by both parties; Amodo has not yet fully analysed resilience to a compromised DPU. Hardware trust. Waits on Tamper evidence for verifier devices ^{2 5}
  • Advances in low-communication training could shrink the margin that the cap enforces. Capacity bounds ^{2 6 7}
• Hardware-attested weight binding
  • Attestation that resists physical attackers, for the enclave variant. Hardware trust. Waits on TEE remote attestation for AI workloads ¹⁴

From the family or selected implementation's record. ?

Model weights

Hidden by Hardware-attested weight binding.

Not involved: Bandwidth limits and compartmentalization.

Inputs and outputs

Not involved: Bandwidth limits and compartmentalization.

Unspecified for Hardware-attested weight binding. Check the implementation record.

Training data

Not involved: Bandwidth limits and compartmentalization and Hardware-attested weight binding.

Bandwidth limits and compartmentalization

• Proposed AI 2040 inference-only verification stack Proposed architecture, AI Futures Project
• Proposed RAND secure inference data center (SIDC) design Proposed architecture, RAND

Hardware-attested weight binding

• Research demo Attestable Audits Research prototype, University of Cambridge
• Research demo PAL*M Research prototype, University of Waterloo
• Operational use Tinfoil model identity (Modelwrap) Product, Tinfoil

1. Verification Plan, R. Dean (2026). Original
2. Traffic Shaping for Workload Classification, Lucid Computing (2026). Original
3. A System Overview for Near-Term, Low-Trust AI Compute Verification, N. Cankaya (2026). Original
4. De-risking Interconnect Limits for AI Verification, A. Scher et al. (2026). Original
5. The Tray as a Bandwidth Boundary, Amodo Design (2026). Original
6. DiLoCo: Distributed Low-Communication Training of Language Models, A. Douillard et al. (2024). Original
7. Does Distributed Training Undermine Compute Governance?, R. Rahman (2026). Original
8. How Tinfoil Proves Exactly What Model Is Running, Tinfoil Team (2026). Original
9. PAL*M: Property Attestation for Large Generative Models, P. Chantasantitam et al. (2026). Original
10. A primer on secure enclaves, Tinfoil (2026). Original
11. Backend infrastructure, Tinfoil (2026). Original
12. How verification works in Tinfoil, Tinfoil (2026). Original
13. modelwrap: Reproducible dm-verity read-only image of Huggingface models, Tinfoil (2026). Original
14. TEE.fail: Breaking Trusted Execution Environments via DDR5 Memory Bus Interposition, J. Chuang et al. (2026). Original
15. Battering RAM: Low-Cost Interposer Attacks on Confidential Computing via Dynamic Memory Aliasing, J. De Meulemeester et al. (2026). Original
16. RMPocalypse: How a Catch-22 Breaks AMD SEV-SNP, B. Schlüter & S. Shinde (2025). Original
17. SEV-SNP RMP Initialization Vulnerability (AMD-SB-3020), AMD (2025). Original
18. On TEEs for Privacy-Preserving Monitoring in AI Governance, Gloria Z (2026). Original
19. Attestable Audits: Verifiable AI Safety Benchmarks Using Trusted Execution Environments, C. Schnabl et al. (2025). Original