Do the Waymo Lidar Lasers harm human eyes significantly?
37
1kṀ5141
resolved Jan 6
Resolved
NO

Lidar over 1400 nm is definitely capable of cornea damage. The question is what about 900~1000 nm. That is more common on avs. They claim they are safe… I don't know.

Market context
Get
Ṁ1,000
to start trading!

🏅 Top traders

#NameTotal profit
1Ṁ190
2Ṁ82
3Ṁ48
4Ṁ32
5Ṁ31
Sort by:

This resolves based on "common sense" and my common sense says we'd have seen reports and such by now if it was an issue at the level of "significantly".

@EvanDaniel Gemini supports.

Based on current scientific evidence, regulatory standards, and documented safety records through early 2026, this market should likely resolve to No. There is no evidence that Waymo’s LiDAR lasers cause significant (or any) harm to human eyes.

Suggested Resolution: NO

Reasoning:

  • Class 1 Certification: Waymo’s LiDAR systems are certified as Class 1 laser products by the FDA and the International Electrotechnical Commission (IEC 60825-1). By definition, Class 1 lasers are "eye-safe under all conditions of normal use," including long-term staring or viewing with optical aids like magnifying glasses.

  • Biological Protection: Waymo utilizes 1550nm wavelengths for its long-range sensors. This wavelength is "retina-safe" because it is absorbed by the cornea and the fluid (vitreous humor) in the human eye before it can reach the sensitive retina. While extremely high power could theoretically cause corneal burns, Waymo's power output is strictly regulated to stay orders of magnitude below that threshold.

  • No Documented Injuries: Despite Waymo vehicles driving tens of millions of miles in densely populated areas like San Francisco, Phoenix, and Los Angeles, there have been zero documented medical cases or lawsuits involving eye damage from their sensors. Most reported "injuries" involving Waymo are related to vehicle collisions, not sensor exposure.

  • The "Camera Confusion": There are documented cases of LiDAR damaging camera sensors (e.g., at tech shows like CES). This occurs because camera lenses lack the protective fluid found in eyes and focus light directly onto a delicate silicon sensor. This technical vulnerability in cameras is often misinterpreted by the public as a risk to human vision.

Supporting Sources:

  1. Official Safety Reports: Waymo’s Safety Report and technical blog posts confirm their adherence to Class 1 eye-safety standards (IEC 60825-1).

  2. Regulatory Standards: The FDA Laser Products guidelines define Class 1 as the safest possible category, equivalent to the lasers used in CD players and barcode scanners.

  3. Scientific Research: Studies published in journals like Nature and by the SPIE Digital Library explain that 1550nm LiDAR is inherently safer for the eye than visible or 905nm light because the ocular fluid is opaque to that frequency, acting as a natural shield for the retina.

  4. Accident Databases: NHTSA (National Highway Traffic Safety Administration) crash data and SGO (Standing General Order) reports through 2025 list collision-related injuries but show no entries for radiation or laser-induced harm.

@Kire_ Can you provide any supporting info to go with the resolution you're suggesting?

@EvanDaniel I don't have any supporting info besides the lack of reports. Feel free to resolve N/A

@Kire_ The question seems obviously applicable to me; n/a seems bad for the usual reasons: https://manifold.markets/dreev/how-bad-is-it-to-resolve-a-market-a

I just mostly assumed that if you were tagging the mods, you had a specific thing in mind you'd hope we would do, and a reason for it.

@EvanDaniel Yeah that's my bad - I was thinking about this more and I realize that my behavior of just blindly tagging mods is not super helpful.

@Kire_ No worries, I'm just trying to encourage habits I'd like to see! I do appreciate people bringing things to the attention of the mods when needed. And "I don't know what to do here but I think something should be done" is also a fine reason to ask the mods for help, but it is nice to make clear if that's what you're doing. We have a lot of reports right now, so I'm trying to encourage people to be clear about what they're asking for.

What is considered resolution here? You can’t just say “I don’t know” and not have a market resolution condition…

@ZackFrank This will resolve on common sense, wager at your own risk!

TL;DR: Don't focus on wavelength, modern LIDARs are often in unsafe wavelengths. Look at the other parts of the safety story.

Three basic LIDAR safety stories:

  • Short-range / low-throughput fixed sensors (generic industrial parts): These are usually spec'd to an intensity that doesn't cause damage.

  • Long-range spinning sensors ("Chicken bucket" LIDARs, e.g. Velodyne): Rely on a spinning mirror to control dwell time; use centrifugal force as the optical path interlock. So the beam doesn't have to be eye-safe because it's only pointed at you for 0.1% of the time. But the point cloud isn't instantaneous -- each point is from a different time -- which can be pathological in high-speed driving.

  • Long-range software-controlled lidars (e.g. every single VC-funded product since the Luminar G): usually eye-unsafe wavelengths and intensities but rely on software safeties to avoid having the laser dwell in any particular solid angle for long enough to cause damage. These are awesome for the software people because they don't have the "rolling shutter" effect of spinners, but some have had poor safety practices in the past.

AFAIK (not a waymo expert) Waymo uses all three -- they have short-range sensors for low-speed obstacle work like parking, four small spinners on the roof, and an undisclosed "instantaneous" sensor that's presumably the third sort.

I think you'd want to look at the third sort and the quality of control software, and look at whether the vendor's FCC certs are up-to-date and not obviously photocopies and so forth. That information must be somewhere...

© Manifold Markets, Inc.TermsPrivacy