wgrover

I think my "cautious optimism" mostly came from the fact that Gemini found subtle problems with the counterfeit packages that I (someone who looks at fakes as part of my job) didn't notice - like mismatched info between the tube and box, and a malformed Irish postal code. If I didn't notice those, then I seriously doubt that a consumer glancing at a package would notice them. In that case, Gemini's insights would be valuable info for a consumer.

Gemini wrongly called an authentic product a fake, but that was mainly because there really were typographical errors in the authentic product's ingredient list. That's more of an indictment of the manufacturer than it is Gemini.

Finally, a lot of the things that Gemini got wrong seemed to me like they could reasonably be attributed to things like optical artifacts in the photos (glare, shadows, stuff like that). Better/more photos might improve that.

All that being said, this is obviously a tiny study of a single AI tool with a single brand of cosmetic product, so it's probably premature for me to optimistic, even cautiously. I've edited the last section of the writeup accordingly.

That's fascinating! If any of your work using AI to identify reproductions is publicly sharable, I'd love to learn more about it (amateur antique collector here). But if not I understand.

Regarding scaling - I think you're right if the AI mostly detects typos, like it did in this work.

I'm hopeful that future AI models (perhaps trained for this purpose, not a general-purpose model like I used here) might be able to identify more subtle variations in e.g. injection molding patterns, the surface finish of a pill, things that would be hopefully a lot harder for counterfeiters to fix than a typo.

Thanks! I'm putting "doing cool stuff with really basic ideas" in my bio... :)

Your comment makes me think of LabVIEW - if you're not familiar with it, it uses a visual programming language ("G") in which data flows down wires. The color of the wire indicates the wire's data type (blue for ints, orange for floats, green for bools, pink for strings) and the width of the wire indicates the dimensionality of the data (thin line = scalar, thick line = 1D array, double-thick = 2D array). The color is more than decorative or even assistive - it's essential to understanding the program.

Fans of the band Wilco will recognize one of the Conet Project's recordings as the source of the woman repeating "Yankee Hotel Foxtrot" in the song "Poor Places" from the eponymous album. Wilco failed to license the sample and the resulting lawsuit gave the Conet Project a portion of Wilco's royalties on that track.

And I remember going to Radio Shack at the mall as a kid and wanting that Armatron robot so badly! Being able to get these things now for pretty cheap and fixing them and playing with them with my kids has been such a treat.

A parenting tip for those with geeky kids:

You can get some of these (like the smaller TOMY robots) for pretty cheap on eBay. They're usually broken, but the innards of these robots are so interesting, just taking them apart is a learning experience. TOMY was brilliant at making seemingly sophisticated toys that were actually run by a single DC motor; all the movements, sounds, sensing etc. were implemented using gears and cam shafts and other mechanisms. A great way to learn about simple machines, and a bonus if you (or your kid) can repair them and bring a 40-year-old robot back to life.

In the site's defense, it is emulating like ten different operating systems as you scroll.

I'm guessing the title was meant to be a nod toward the meaning of the English word iota, "an extremely small amount".