P020 Side-channel defences at scale
Can side-channel defences (shielding, filters, jamming, wardens) be mass-manufactured and retrofitted to cap a tray's leakage below a stated level, and at what unit cost?
Engineering side-channel defences (TEMPEST, filters, jamming, wardens) at scale.
Terms: TEMPEST: the standard body of techniques for shielding equipment so it does not leak information through stray electromagnetic emissions. wardens: monitors that watch for such leakage.
Status
Analyzed A paper design from a research sprint the author calls 'a first-pass viability study': enclosures, sensors, look-through jamming, inspections and physical security with an estimated bill of materials below 1% of cluster cost (this survey's cost goal of 1% of facility cost, met on paper only); power-filter volume cost 'has proven difficult' to pin down; no attenuation or leakage-rate cap measured; the 39 dB look-through suppression is cited from commodity radios in work not held; wardens: no work ('unexplored').
I present an example of a defense architecture towards the end of this post. It leverages a synergy of cheap enclosures, sensors, jamming, unannounced hardware inspections and physical security … at an estimated (!) Bill-of-Materials (BOM) below 1% of the cost of the monitored cluster
Posed by
- AI 2040 (2026), Get Involved in Verification:
No plan yet for quickly scaling side-channel defences on a frontier cluster; only early theoretical pieces exist.
(Completeness, L129) (source states the problem but not the quantity the question asks for)
Also raised in: Cankaya (2026), A System Overview for Near-Term, Low-Trust AI Compute Verification (v0.2); Cankaya (2026), Research Note: The Fundamentals and Feasibility of Secure Network Taps for Verifying AI Datacenter Use; Cankaya (2026), Suppressing Side Channels in an Untrusted Data Center via Retrofitted Defenses (MIRI); Amodo (2026), AI 2040 Plan A — How You Can Help; Amodo (2026), AI 2040 Plan A — Verification SITREP.
Sub-questions
| ID | Sub-question | What the sources report |
|---|---|---|
| P020.1 | Can tamper-evident, shielded enclosures or TEMPEST-proof trays be mass-manufactured and retrofitted onto a frontier cluster quickly, and what attenuation caps tray EM leakage below a stated leakage rate? | open |
| P020.2 | What residual covert capacity per power cable remains after a TEMPEST filter plus continuous look-through (bps)? | partly answered |
| P020.3 | What is the volume unit cost of a TEMPEST-grade filter for a rack power whip ($)? | open |
| P020.4 | What self-interference suppression (dB) does continuous look-through jamming achieve on power-line and ultrasonic channels (39 dB shown only in commodity radios)? | partly answered |
| P020.5 | What does an OEO isolation boundary per pretraining pod cost, as % of pod cost? | partly answered |
| P020.6 | What does physical guarding and anti-bypass security cost relative to tap/enclosure BOM, and can anti-tamper assurance dispense with guards? | implied by a source, not stated as open |
| P020.7 | Can an RF/power 'warden' detect a covert channel (at a stated rate) out of a GPU tray at a stated detection rate and false-alarm rate (targets to be agreed)? | open |
Possibly relevant work found by automated search
Found by AI agents searching for this problem's sub-questions, mostly by keyword and citation in the OpenAlex scholarly database and on the web; a second AI pass reviewed each one, and we spot-checked some by hand. They seemed reasonably related, as far as one can tell from outside each paper's field. Some come from neighbouring fields, such as arms control and financial audit, and bear on the problem by analogy. Not reviewed against the status above; a paper listed here does not change that status. 928 of 1052 quotes checked word for word against the paper's text.
For P020.1: Can tamper-evident, shielded enclosures or TEMPEST-proof trays be mass-manufactured and retrofitted onto a frontier cluster quickly, and what attenuation caps tray EM leakage below a stated leakage rate?
For P020.2: What residual covert capacity per power cable remains after a TEMPEST filter plus continuous look-through (bps)?
For P020.4: What self-interference suppression (dB) does continuous look-through jamming achieve on power-line and ultrasonic channels (39 dB shown only in commodity radios)?
For P020.5: What does an OEO isolation boundary per pretraining pod cost, as % of pod cost?
For P020.6: What does physical guarding and anti-bypass security cost relative to tap/enclosure BOM, and can anti-tamper assurance dispense with guards?
For P020.7: Can an RF/power 'warden' detect a covert channel (at a stated rate) out of a GPU tray at a stated detection rate and false-alarm rate (targets to be agreed)?
* Full text not reviewed; more likely to contain errors.
Data
- problems.csv — one row per problem
- problem_status.csv — status, scope and evidence per problem
- problem_sources.csv — the posed-by quote and other documents per problem
- leaves.csv — all sub-questions, including those without a checkable form
- leaf_sources.csv — every source row behind each sub-question