Detecting AI Training in Datacentre Power

arXiv:2610.30182v1 Announce Type: cross Abstract: 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.

Method detects AI training in datacentre power, making hidden training detectable and expensive.

The method improves detection of unauthorized AI training, enhancing governance and accountability, which is positive. However, the need for detection implies potential for covert use, reflecting ongoing mistrust. Its cost imposition on hidden training adds deterrence but may encourage resource inefficiency or adversarial behaviors. Overall positive but with some mixed societal implications.

This is a narrow technical contribution enabling verification of AI training-run compliance, which modestly supports governance infrastructure. It is simulation-based and speculative in real-world application, but the implication, better tools for trustless oversight of large AI training, is mildly positive for societal control over AI development.

The method enables effective detection of AI training, promoting transparency and compliance. It makes hidden training both detectable and costly, enhancing AI governance.

SOURCE     arXiv cs.CY (Computers and Society) (tier 1)
   ↓
DOCUMENT   4fae7cf3-6d62-4807-a92b-f3c8f6d68c15
           https://arxiv.org/abs/2610.30182
   ↓
EVIDENCE   2 extracted excerpts
   ↓
MODEL RUN  4 runs, methodology 1.0.1
   ↓
SCORE      +38  (Favourable)
   ↓
CONFIDENCE 78%