6 ms·
Security/privacy concerns aside, Recall doesn't particularly feel like an "AI" feature to me. It's on-device OCR plus a SQLite database that you can then search
by simonw 2y ago
Security/privacy concerns aside, Recall doesn't particularly feel like an "AI" feature to me. It's on-device OCR plus a SQLite database that you can then search, right?
Even by today's loose definition of "AI" I'm having trouble making that connection. I guess the OCR is based on machine learning?
Is there an LLM component to Recall that I've missed? If so, can I build websites with prompt injection attacks on them that will target Recall users directly, by getting my rogue instructions indexed into their SQLite database and later fed into the LLM?
- zamalek 2y agoIt probably has RAG.
- simonw 2y agoIs that confirmed? Microsoft have been boasting that it all runs on-device, so are they running RAG using an on-device model (Phi-3 or similar) as part of the Recall feature?
- rihegher 2y agoI had this confirmed during the Q&A portion of a demo presented by Microsoft. All the processing is supposed to happen locally using the NPU capacity of the hardware.
- jrm4 2y agoI really can't fathom what it takes to casually be like "security/privacy concerns aside" here.
- codetrotter 2y agoI think you are misreading the comment. Saying aside from those concerns does not dismiss the concerns themselves. It’s just expressing that this comment will speak about some other things, and not those things.
- simonw 2y agoI decided not to fully type out "obviously the security/privacy concerns are the most important thing here and should not be ignored... but aside from that, does anyone know if..."
- randomdata 2y agoI assume you are working with the AI definition that is along the lines of: "Problems a computer may not be able to solve"? OCR was absolutely considered AI before we knew how to do it. Now that we understand it, it's just computing, of course. But it still is reasonably considered AI by any other definition of AI.
- KeplerBoy 2y agoIsn't SotA OCR done using CNNs? So it's AI even in the modern sense.
- giantg2 2y agoI assume the AI part is using some model to interpret the stuff on the screen (Maybe even non-text) and used in the search results. That's just a guess. I'm not really that interested in the Recall product, so I have dug into it.
- maciejgryka 2y agoI have no actual info on this, but I always assumed they'd compute some mutlimodal embeddings of the screenshots to then retrieve semantically-relevant ones by text? And yeah, they'd have to do it using on-device models, which doesn't seem out of reach?
- breadwinner 2y agoDemos show that you can search for "Blue bag" even if the words "blue bag" don't appear on the screen. If there was a blue bag in a photo, say in a PowerPoint slide, it will find it.
- cflewis 2y agoIn some ways, I find this even creepier than when I thought it was more than an SQLite database.
- cjk2 2y agoTo be fair Apple Photos does this. But it does it to your photos not your entire universe.
- hu3 2y agoSame for google photos. I searched for "smiling in beach" and it works flawlessly. I recon Microsoft will want to offload this kind of search processing to user machines with the advent in "AI processor computers" we saw recently: https://blogs.microsoft.com/blog/2024/05/20/introducing-copilot-pcs/ https://blogs.microsoft.com/blog/2024/05/20/introducing-copi...
- mmcnl 2y agoI believe Apple does it completely locally on the device, whereas Google searches in the cloud.
- ethagnawl 2y agoDropbox has also been doing this for years. It's helpful sometimes and other times completely misses the mark. Personally, I do find it to be creepy and it has me edging closer to using a more privacy-focused cloud solution to backup my important documents, photos, etc.
- sleepybrett 2y agospotlight is also indexing pretty much every file you create on your mac.
- ben_w 2y ago> Even by today's loose definition of "AI" I'm having trouble making that connection. I guess the OCR is based on machine learning? It's long been noticed that every time AI researchers figure out how to do a thing, it goes from "this is impossible SciFi nonsense" to "that's not real AI". It's weird for me to actually encounter people doing this. I remember when OCR was impossible, at any decent quality level, for AI. We used this to stop bots logging into forums, with a thing we called a "Completely Automated Public Turing test to tell Computers and Humans Apart" or CAPTCHA for short. It started off as text, then the text got increasingly distorted until most humans had trouble reading it, then it became house numbers, then it became "click all pictures of XYZ", before mostly disappearing into analytics seeing where you hovered your mouse cursor and which other websites you visited.
- MrRadar 2y agoYeah, take this XKCD for instance: https://xkcd.com/1425/ https://xkcd.com/1425/ That was published 10 years ago but now the second item is (nearly) as trivial to accomplish as the first.
- pavel_lishin 2y agoTo be fair, we've had multiple research teams and ten years :)
- jantissler 2y agoBut that's not the point I think. The point to me is: 10 years ago, this seemed nearly impossible to solve. "Weird, impossible sci-fi stuff". Now I have this exact feature in the Photos app of my iPhone. It not only finds my dog in the photos, but correctly determines its breed. It's amazing! And it works locally on my device, no servers involved. But I'm sure many people would argue that it is not "real AI", because it is "just XYZ". So as soon as we figure out how to do it, it's declared banal. Like LLMs. I'm baffled by the people who think it's "just" this or "just" that. That little word "just" buries years of research and several breakthroughs. Somehow, AI is always what AI can't do yet. It's always the next thing. The noise of goal posts being moved is sometimes deafening. And people don't even realize they're doing it.
- simonw 2y agoAnswering my own question (with the help of replies on Twitter: https://twitter.com/simonw/status/1798368111038779610 https://twitter.com/simonw/status/1798368111038779610 ) Best guess is that there's some level of semantic analysis going on beyond just OCR, which tips the balance (at least for me) to being an "AI" feature based on loose current usage of that term. The clue is that in this demo https://www.youtube.com/watch?v=aZbHd4suAnQ&t=1062s https://www.youtube.com/watch?v=aZbHd4suAnQ&t=1062s they show searching for "blue pantsuit with sequin lace" finds something that was described as "peacock" in the text. It looks like this is based on embedding search against image embeddings. And... one of the 3 SQLite databases created by Recall is called "SemanticImageStore" - which suggests to me there may be a CLIP-style image embeddings model being run on-device here. Those databases contain "diskann" columns which are likely a reference to Microsoft's vector indexing library: https://github.com/microsoft/DiskANN https://github.com/microsoft/DiskANN
- osolo 2y agoMy team worked on this feature and I can confirm that yes, semantic search is being done for both text and images. This allows you to do fuzzy searches like you mentioned in your comment and use words to match images. Everything runs on your device (all the models, the vector database, etc.) in order to preserve privacy.
- simonw 2y agoThat's really impressive. Can you share which embedding models you are using for this? Also, is Phi-3 or Phi-3 Vision involved?
- osolo 2y agoI don't think they've announced publicly what models we're using. I don't think there's any particular reason for this, but just in case, I can't name them here. I'll see if this can be addressed in a blog post of something. I can tell you that Recall isn't using Phi. Rather, it's using a collection of models that are much more tuned (and therefore much more efficient) for the feature.
- ryandrake 2y ago> Recall doesn't particularly feel like an "AI" feature to me. The word "AI" is becoming a pure marketing checkbox. You can't release any new product in 2024 without somehow attaching the word "AI" to it. It doesn't matter whether it actually uses AI or an LLM.
- paulmd 2y agoWell, this one does, so what’s the point of this observation in this context?
- EasyMark 2y agoBut it doesn't just recognize characters though? It recognizes things happening on video and scenes in pictures without text. for example, you could prompt "find my photos and videos with dogs in them"