• 0 Posts
  • 22 Comments
Joined 2 years ago
cake
Cake day: October 30th, 2023

help-circle






  • On a “respond to an individual query” level, yeah it’s not that much. But prior to response the data center had to be constructed, the entire web had to be scraped, the models trained, the servers continually ran regardless of load. There’s also way too many “hidden” queries across the web in general from companies trying to summarize every email or product.

    All of that adds to the energy costs. This equivocation is meant to make people feel less bad about the energy impact of using AI, when so much of the cost is in building AI.

    Furthermore, that’s the median value–the one that falls right in the middle of the quantity of queries. There’s a limit to how much less energy a query to the left of the median can use; there’s a significantly higher runway to the right of the median for excess energy use. This also only accounted for text queries; images and video generation efforts are gonna use a lot more.



  • Relying on a small child to stay on the ground in order to not accidentally kill themselves is a great way to end up with a dead kid.

    Furniture should be anchored to a wall, guns locked in safes with the safety enabled and ammo removed, drugs in child resistant packaging locked in a cabinet, drawers and cabinets secured, etc.

    Kids climb stuff, get into things, find things they shouldn’t, AND they emulate what they see their parents do. Putting something out of reach is nowhere near secure enough.




  • Probably not.

    This kind of thing relies on the fact that the emitter and environments are static, impacting the propagation of the signals in a predictable way and that each person, having a unique physique, consistently interferes with that propagation in the same way. It’s a tool that reports “the interference in this room looks like the same interference observed in these past cases.”

    Search and rescue is a very dynamic environment, with no opportunity to establish a local baseline, and with a high likelihood that the physiological signal you are looking for has been altered (such as by broken or severed limbs).

    There are some other WiFi sniffing technologies that might be more useful for S&R such as movement detection, but I’m not sure if that will work as well when the broadcaster is outside the environment (as the more rubble between the emitter and the target the weaker your signal from reflections against the rubble).

    Don’t think of this as being able to see through walls like with a futuristic camera, think of this as AI assisted anomaly detection in signal processing (which is exactly what the researchers are doing).