First screen, captured 1 day ago. Hover to scroll the page, or open the video.

Leaking Secrets from Air-Gap Computers by Spelling Radio Signals from RAM (2024)

What HN thinks

Air gaps leak, Faraday cages get recommended, and somehow the conversation ends with a Roomba plotting its escape.

Clever but impractical attack
It needs malware already on the machine plus a receiver within a few metres for about 100 bps, so many see it as academic, with people and supply chains being the real risk.
Just use a Faraday cage
Shielding is cheap relative to the secrets involved and standard for SCIF/TEMPEST setups, though some note very low frequencies are harder to block.
AI containment worries
Some see covert channels as proof AI can't be fully contained, while others point out labs already gave agents internet access anyway.

Consensus: Most agree that the attack needs a prior compromise and close physical proximity, and that electromagnetic shielding largely defeats it. Commenters also note that Mordechai Guri is the prolific leading researcher on air-gap exfiltration.

Recommended in the comments

Powerline networking (product): Mentioned as working well when needed, though one ham radio operator called it RF-noisy and antisocial.
RAMEAR ↗ (tool): Shared by its author as an earlier working demo of the same attack.

Summarised by Claude from the comment thread 1 day ago.

Points: 202
Comments: 96
Best rank: #6
Posted: 2026-11-01 by hapax_legomena
Category: Security

air-gap side-channel-attacks covert-channels electromagnetic-emanations software-defined-radio data-exfiltration arxiv

All captures:

Full page

Captured 1 day ago. Click for the whole screenshot.

Looks similar

Screenshot of arxiv.org: Proving and improving the Platonic Representation Hypothesis with JEPAs

Proving and improving the Platonic Representation Hypothesis with JEPAs (arxiv.org)

156 points by Anon84 1 day ago | 78 comments

arXiv paper by Balestriero and LeCun proving a Platonic Representation Theorem for JEPAs and showing good views align modalities more efficiently than next-token prediction.

jepa platonic-representation-hypothesis representation-learning multimodal-alignment

Screenshot of arxiv.org: The alien space of science: Coherent but cognitively unavailable directions

The alien space of science: Coherent but cognitively unavailable directions (arxiv.org)

195 points by martincsweiss 11 days ago | 80 comments

arXiv abstract page for a NeurIPS 2026 paper proposing a method to sample scientifically coherent research directions that existing researcher communities are unlikely to propose.

research-ideation llm scientific-discovery novelty

Screenshot of arxiv.org: Monoidal computer I: Basic computability by string diagrams (2012)

Monoidal computer I: Basic computability by string diagrams (2012)(arxiv.org)

3 points by measurablefunc 10 days ago | 0 comments

arXiv abstract page for Dusko Pavlovic's 2012 paper proposing a model of computation based on monoidal categories and string diagrams.

category-theory string-diagrams monoidal-categories computability

Screenshot of arxiv.org: Executable Archaeology: Reanimating the Heuristic Compiler (1961) and KBEmacs (1984)

Executable Archaeology: Reanimating the Heuristic Compiler (1961) and KBEmacs (1984)(arxiv.org)

27 points by abrax3141 5 days ago | 0 comments

arXiv paper on reanimating two historical AI programming systems, Simon's Heuristic Compiler (1961) and MIT's KBEmacs (1984), by running their original code with help from Claude Code.

software-archaeology heuristic-compiler kbemacs herbert-simon

Screenshot of arxiv.org: Qubes OS Security in the Public Record

Qubes OS Security in the Public Record (arxiv.org)

75 points by sciences44 3 months ago | 12 comments

Screenshot of arxiv.org: Why is AI generated language so hard to understand?

Why is AI generated language so hard to understand?(arxiv.org)

194 points by canjobear 8 days ago | 167 comments

arXiv paper analyzing psycholinguistic features that make AI-generated text harder to understand, attributing it to optimization for information density.

llm psycholinguistics ai-generated-text readability