arXiv subjects

Mauricio Baker

Publications and source records attributed to Mauricio Baker.

7 recordsLinked to original sources

arXiv2026

Detecting AI Training in Datacentre Power

AI governance agreements require methods to detect large model training runs on a monitored computing cluster, without trusting the operator. We develop a method to detect synchronous model training in facility-level power by Viterbi tracking of the wandering iteration cadence. On simulated datacentre power, when the operator makes no attempt to hide, the Viterbi tracker detects at least $96\%$ of training runs against a declared inference workload, at a $1\%$ false-alarm rate and within $0.04$ of the Neyman--Pearson optimum, given a meter that samples at $2$ Hz or faster. Of eight adversarial strategies an operator might use to hide, only extreme variation of the work per iteration evades the tracker. Its strongest form hides training in up to $73\%$ of five-minute observations, but over a day at least $87\%$ of such runs are still caught, at a $0.1\%$ false-alarm rate per run. This adversarial strategy costs no throughput, but in small-scale training the operator needs $93$--$267\%$ more compute to reach the same loss. A modest passive meter is sufficient to make hidden training both detectable and expensive, which makes power metering a promising part of compute verification that does not depend on trusting the operator.

cs.CR↗

arXiv2026

Can Power Draw Constrain Covert Compute? Limits of Analogue Verification for AI Governance

Frontier AI treaties or agreements on limiting computation require external verification; an external auditor must be able to confirm how much computation actually ran and that parties are adhering to the agreement. Analogue, off-chip measurements such as power draw provide an information channel for verification. It is unknown how well these analogue channels can constrain computation against an adversary who actively tries to subvert the audit. We derive a closed form for $β$, the largest hidden computation a power trace cannot exclude, as a fraction of the declared machine capacity. Measurements on NVIDIA A100 GPUs constrain $β= 1.16$ in the worst case, while adversarial matched-energy strategies are shown to hide at least $β= 0.41$ of compute. Analogue power measurements alone therefore constrain compute weakly. Additional restrictions granted by the threat model, such as the ability of the verifier to re-execute the declared work at an observed operating point, let the verifier push $β$ down to $0.059$ in the maximally restricted case. This gives a quantitative estimate of what analogue measurements can contribute to compute verification.

cs.CY↗

arXiv2026

How Sovereign Is Sovereign Compute? A Review of 775 Non-U.S. Data Centers

Previous literature has proposed that the companies operating data centers enforce government regulations on AI companies. Using a new dataset of 775 non-U.S. data center projects, this paper estimates how often data centers could be subject to foreign legal authorities due to the nationality of the data center operators. We find that U.S. companies operate 48% of all non-U.S. data center projects in our dataset when weighted by investment value - a proxy for compute capacity. This is an approximation based on public data and should be interpreted as an initial estimate. For the United States, our findings suggest that data center operators offer a lever for internationally governing AI that complements traditional export controls, since operators can be used to regulate computing resources already deployed in non-U.S. data centers. For other countries, our results show that building data centers locally does not guarantee digital sovereignty if those facilities are run by foreign entities. To support future research, we release our dataset, which documents over 20 variables relating to each data center, including the year it was announced, the investment value, and its operator's national affiliation. The dataset also includes over 1,000 quotes describing these data centers' strategic motivations, operational challenges, and engagement with U.S. and Chinese entities.

cs.CY↗

arXiv2026

Insights for an AI Whistleblower Office from 30 Case Studies

Whistleblower programmes are a promising tool for uncovering noncompliance with AI regulations. This paper aims to help policymakers design an AI whistleblower programme by giving them an understanding of whistleblowers' motivations, and of the overall whistleblowing process. We take an empirical approach, assembling a dataset of 30 case studies of whistleblowers. This dataset includes dozens of features of each case, which range from 1978 to 2020 and span 15 industries. Our findings suggest that whistleblower programmes will be more effective if they financially reward whistleblowers, provide protections for whistleblowers, enable whistleblowers to report anonymously, are adequately staffed and funded, and provide advice to potential whistleblowers. We provide ten concrete policy recommendations for an AI whistleblower programme at the end of this paper.

cs.CY↗

arXiv2025

Verifying International Agreements on AI: Six Layers of Verification for Rules on Large-Scale AI Development and Deployment

The risks of frontier AI may require international cooperation, which in turn may require verification: checking that all parties follow agreed-on rules. For instance, states might need to verify that powerful AI models are widely deployed only after their risks to international security have been evaluated and deemed manageable. However, research on AI verification could benefit from greater clarity and detail. To address this, this report provides an in-depth overview of AI verification, intended for both policy professionals and technical researchers. We present novel conceptual frameworks, detailed implementation options, and key R&D challenges. These draw on existing literature, expert interviews, and original analysis, all within the scope of confidentially overseeing AI development and deployment that uses thousands of high-end AI chips. We find that states could eventually verify compliance by using six largely independent verification approaches with substantial redundancy: (1) built-in security features in AI chips; (2-3) separate monitoring devices attached to AI chips; and (4-6) personnel-based mechanisms, such as whistleblower programs. While promising, these approaches require guardrails to protect against abuse and power concentration, and many of these technologies have yet to be built or stress-tested. To enable states to confidently verify compliance with rules on large-scale AI development and deployment, the R&D challenges we list need significant progress.

cs.CY↗

arXiv2025

Open Problems in Technical AI Governance

AI progress is creating a growing range of risks and opportunities, but it is often unclear how they should be navigated. In many cases, the barriers and uncertainties faced are at least partly technical. Technical AI governance, referring to technical analysis and tools for supporting the effective governance of AI, seeks to address such challenges. It can help to (a) identify areas where intervention is needed, (b) identify and assess the efficacy of potential governance actions, and (c) enhance governance options by designing mechanisms for enforcement, incentivization, or compliance. In this paper, we explain what technical AI governance is, why it is important, and present a taxonomy and incomplete catalog of its open problems. This paper is intended as a resource for technical researchers or research funders looking to contribute to AI governance.

cs.CY↗

arXiv2023

Nuclear Arms Control Verification and Lessons for AI Treaties

Security risks from AI have motivated calls for international agreements that guardrail the technology. However, even if states could agree on what rules to set on AI, the problem of verifying compliance might make these agreements infeasible. To help clarify the difficulty of verifying agreements on AI$\unicode{x2013}$and identify actions that might reduce this difficulty$\unicode{x2013}$this report examines the case study of verification in nuclear arms control. We review the implementation, track records, and politics of verification across three types of nuclear arms control agreements. Then, we consider implications for the case of AI, especially AI development that relies on thousands of highly specialized chips. In this context, the case study suggests that, with certain preparations, the foreseeable challenges of verification would be reduced to levels that were successfully managed in nuclear arms control. To avoid even worse challenges, substantial preparations are needed: (1) developing privacy-preserving, secure, and acceptably priced methods for verifying the compliance of hardware, given inspection access; and (2) building an initial, incomplete verification system, with authorities and precedents that allow its gaps to be quickly closed if and when the political will arises.

cs.CY↗