←back to thread

Niantic announces “Large Geospatial Model” trained on Pokémon Go player data

(nianticlabs.com)
383 points bookstore-romeo | 4 comments | 19 Nov 24 20:02 UTC | HN request time: 0.678s | source
Show context
dankwizard ◴[21 Nov 24 03:51 UTC] No.42200923[source]▶
We do this at MyFitnessPal.

When users scan their barcode, the preview window is zoomed in so users think its mostly barcode. We actually get quite a bit more background noise typically of a fridge, supermarket aisle, pantry etc. but it is sent across to us, stored, and trained on.

Within the next year we will have a pretty good idea of the average pantry, fridge, supermarket aisle. Who knows what is next

replies(13): >>42200951 #>>42200954 #>>42200955 #>>42200966 #>>42200972 #>>42200976 #>>42200980 #>>42201021 #>>42201023 #>>42201114 #>>42201284 #>>42201321 #>>42202322 #
ryanschaefer ◴[21 Nov 24 03:59 UTC] No.42200951[source]▶
I’d be interested in how your privacy policy allows this. I can’t find where it mentions photos are stored or used for training purposes…
replies(2): >>42200979 #>>42201060 #
1. ipaddr ◴[21 Nov 24 04:02 UTC] No.42200979[source]▶
I would be more interested on why you believe something like this isn't baked into most privacy policies.

I'm not shocked but I'm shocked you are shocked.

replies(2): >>42201014 #>>42201054 #
2. ryanschaefer ◴[21 Nov 24 04:09 UTC] No.42201014[source]▶
I’m not exactly shocked that it could exist. But this usage (beyond the scope of processing barcodes) seems like it couldn’t be construed to fit into the normal avenues of data collection under a privacy policy. Also with regard to training specifically, this policy was created in late 2020 so I don’t know how it would cover generative models.
replies(1): >>42201041 #
3. ◴[21 Nov 24 04:14 UTC] No.42201041[source]▶
4. moreofthis ◴[21 Nov 24 04:17 UTC] No.42201054[source]▶
Giving their policy an (admittedly quick) skim there doesn't seem to be any section that mentions AI, LLMs, training any kind of model, using image data from barcode pictures, etc. I'd be very curious to see the explanation of how this is baked into the policy.