---
title: "Optical receiver updates AI model memory directly from light"
url: https://www.parallelquant.com/posts/optical-receiver-updates-ai-model-memory-directly-from-light-145e8c
source_name: "IEEE Spectrum"
source_url: https://spectrum.ieee.org/ai-in-robotics
published: 2026-07-26T13:00:01.000Z
topics: ["chips", "research"]
publisher: "Parallel Quant"
---

# Optical receiver updates AI model memory directly from light

*2026-07-26 · Source: [IEEE Spectrum](https://spectrum.ieee.org/ai-in-robotics)*

Cornell Tech researchers built an optical receiver that updates its own memory directly from photocurrents produced by a beamed light array, rather than just reading encoded data like a QR code. The design, presented at the IEEE/JSAP Symposium on VLSI Technology & Circuits, aims to let AI model parameters be loaded optically to cut memory-update energy costs.

**Why it matters:** Energy and power constraints are already a live bottleneck for AI infrastructure, as seen in the PJM grid strain and Google's first negative-cash-flow quarter tied to AI spending. A hardware approach that cuts the energy cost of updating model parameters in data centers, self-driving cars, or edge robots targets that bottleneck directly, though it's early-stage research rather than a deployable product.

**Topics:** chips, research

---
Read the original: https://spectrum.ieee.org/ai-in-robotics
Canonical: https://www.parallelquant.com/posts/optical-receiver-updates-ai-model-memory-directly-from-light-145e8c
Published by Parallel Quant — https://www.parallelquant.com