# New security papers show smart bulbs leaking data through light and reinforcement-learning attackers learning to evade edge intrusion detection

A companion study gives a quantitative framework for deciding where machine-learning intrusion detection can run in Internet of Things deployments.

- Published: 2026-10-08T06:25:29.519Z
- Canonical: https://polylog.news/ai/2026-10-08/new-security-papers-show-smart-bulbs-leaking-data-through-li
- Publisher: Polylog (AI desk)
- Section: tech
- Sources: [arXiv cs.CR (optical covert channels)](https://arxiv.org/abs/2610.08868), [arXiv cs.CR (adversarial RL port-scan evasion)](https://arxiv.org/abs/2610.08864), [arXiv cs.CR (IoT IDS feasibility)](https://arxiv.org/abs/2610.08867), [arXiv (Hue covert channel)](https://arxiv.org/pdf/2408.14613)

A new study asks whether commodity smart light-emitting diode (LED) bulbs can carry digital data through brightness changes small enough that a person in the room would struggle to notice them. Smart bulbs are networked devices whose output…

This story is for subscribers. Read it in full at https://polylog.news/ai/2026-10-08/new-security-papers-show-smart-bulbs-leaking-data-through-li (subscription information: https://polylog.news/pricing).