OpenAI buys smartphone camera maker Glass Imaging for $300M(techcrunch.com) |
OpenAI buys smartphone camera maker Glass Imaging for $300M(techcrunch.com) |
they had a really cool optical approach that I kind of felt immediately was doomed in the consumer market but would likely find a cool, valuable niche. The unique distortion patterns and front element I felt probably ruled out cinema but beyond that I had no idea. looks like they pivoted at some point but their expertise in lens accommodation does seem like it could be helpful.
Putting a lot of effort into pushing complexity into software to reduce manufacturing complexity is precisely the sort of thing you worry about when you're getting serious about building consumer products.
Just yesterday I saw a post I cant recall if on here or reddit, someone coded an entire OS with Claude Code, I would not be surprised if they invest AI compute at building an OS themselves, though I'd never buy a phone from these companies.
The person got it running on bare hardware.
Found it:
https://www.reddit.com/r/ClaudeAI/comments/1wfpydl/i_asked_c...
This becomes even tougher if they go the Apple route and don't allow other phone manufacturers to use their OS. Alternatively, if they go down the Windows Phone route, we saw how that played out considering that Google _pays_ manufacturers to use Android
Making OS is very hard, but getting necessary adoption is likely even harder
I, as someone who would rather leave the industry than become a manager of either people or autonomous agents, or worse just a meat-based relay, am truly obsolete.
Do you want fries with that?
The whole vibe coding an OS might become a thing in the next few years, but you still have to QA and test it, and there simply hasn't actually been enough time for that.
Moreover the interaction between firmware and the OS is a massive headache to overcome, especially if you are dealing with battery powered devices.
I totally get why they wanted a custom userland, built it myself for embedded devices. But IME Linux kernel is not terribly bad these days.
AI is rapidly getting to such a place where the you don't really need a developer ecosystem to get your device off the ground. As "plugins" into your agent at some later point -- sure, why not.
There is a solution sitting there, which is running Android and having the AI run it behind the scenes I guess, but is suspect that would end up looking a lot like we have now.
So they are almost certianly making an alexa type thing with cameras. So they can build a device capable of recording the entire context of a household.
This could also be where China gets another win because of their de-Googled Android-not-Android ecosystem which would enable more competition in their local market than can be achieved elsewhere.
But if the LLM isn't local, I'm not buying.
For the mass market, the AI industry has a trust problem.
> For the mass market, the AI industry has a trust problem.
Oh how far we've fallen! Everyone should invite an OpenAI into their homes to watch and hear their most intimate moments. Just think of the value! Next time someone fights with their spouse, the machine can email a summary with advice the next day! Mass bliss will surely follow!
ChatGPT is _enormous_. It doesn't matter if you don't trust it, half the planet does. They don't need to get the HN bubble on board.
Edit to add: the absolute state of the replies to this. Hilarious.
I think it's more likely it would be a radical departure from existing mobile OS.
More like an entirely generative experience. No discrete apps. The phone screen just morphs into whatever purpose you need at the time.
Lol, almost forgot about that cringe $6.5b video announcement - thanks for the reminder!
No vision, let alone strategy.
Probably they are buying the staff, not the technology.
Surely OpenAI can replicate that with their own tools in an afternoon and a few million tokens.
Hardware is the next thing which is far more valuable IP and with its staff.
AI model vendors will likely be torn between developing their own hardware to avoid hardware dependency or moving forward together.
Quietly, Apple and Google will take advantage of this trend.
Will be interesting to see how the legal battle plays out as well.
I'm sick of the Android/iOS duopoly. Let them cook. The market is so ripe for disruption.
- doesnt make ANY cameras
- its tech (AI slop over taken picture) is used by one lonely Chinese vendor
A lot of software people live under the illusion that hardware bugs don't happen, but they're ubiquitous and it's the job of firmware and OS people to reduce their impact.
Edit: I should add anyone that's worked with me knows I strongly advocate offline fuzz testing of all software components, especially to deal with unexpected states coming from the hardware side. This has proven unreasonably effective, and absolutely is something agents can help with, but they won't solve the whole problem.
OpenAI is as bad as the rest. Anthropic likely is too but they seem a hair better. Still bad though.
Will you operate the bread baking machine, own a bread factory ($++) or make eco-handmade artisan bread ($--).
Or do something completely different? Trade railway shares?
Bread baking factories could clearly destroy civilization as we know it, create incredible economic inequality, make a permanent underclass, etc etc.
Basically you have about 130mw power budget. You need to do position, display, sound output, transcription, etc.
https://www.secretbatcave.co.uk/software/musing-on-openais-d... << an explanation here
Modern smartphones won’t be supplanted by tech that could just run on a phone. That doesn’t make any sense. For any new tech to take off and replace phones, it has to provide something that isn’t simply additive to the phone. Otherwise, it just makes more sense to build it into the existing phone.
“A phone, but it has AI and no apps” isn’t even additive. It’s purely a loss of functionality.
I don't necessarily agree that “A phone, but it has AI and no apps” is a good product, but it could be considered additive.
When the iPhone came out, it was "a pocket computer, but without a physical keyboard and mouse" and some might have argued that it was subtractive because of that. Maybe we shouldn't be thinking about "apps" as something inherently required by any future device ever created.
1. A screen that covers the whole device.
2. A device that doesn’t suck for touch.
The loss of the physical keyboard was indeed actually a loss, but was offset by meaningful improvements.
My problem with most of these proposed devices is that they are purely subtractive. I can run ChatGPT or Claude or whatever on my phone right now. Where is the part where a new type of device delivers some new experience or capability?
I don't see that happening unless the new hardware and the way people interact with it is truly "magical" in some sense. Smartphones as a mature device category have so many things going for them that anything new will likely come as a complement for them (which apple can capitalise on) vs a complete replacement.
Actually, no. It's the irrelevant changes, such as Liquid Glass, which are the problem, where we all are made to learn new stuff to keep some designers busy.
> I don't see that happening unless the new hardware and the way people interact with it is truly "magical" in some sense.
But a truly AI enabled interface _would be_ magical. Being able to have a phone where your interface to it is completely built around your preferences, and not change at the whim of some external designer, for example. Where apps are merely things it uses on your behalf in the background . . . Neither Apple or Google could do this because it would disrupt their own cash cows too much.
There is also a lot of scope for using neural nets to drive peripherals (including moving parts) in non obvious ways which could add to it all.
Liquid glass didn't require any relearning at all, the iOS interface changed aesthetics, not functionality. The OS has worked largely the same for the past 10 years since the iPhone X.
But I fail to see how an AI generated interface would be any better. What if after several back and forth with the AI, it still can't get the design right for you just so? It's not deterministic, you might get a different result every time you open your phone. Expecting third party services to leave their APIs open just so OpenAI phones can consume them is awfully optimistic as the moment such a device becomes an existential threat to third parties, APIs will start getting locked down.
By all means, OpenAI should try, not saying they shouldn't, but I'm not optimistic it'll be a success. The smartphone is already basically peak form factor. No one is going to want to carry around a second device, which is why every other AI device/pin is DOA. It has to come out of the gate ready to fully replace someone's smartphone and the surrounding ecosystem of it or it also is DOA. Good camera? Support spotify/apple music? messaging, video calls, airdrop/nearbyshare, works with smartwatches and fitness trackers, etc. If it doesn't, it will fail.
You want me to buy another Android phone AND proprietary peripherals too? In this economy?
Only the HN bubble is interested in a new publicly subsidized AI phone option, over the AI phone already in their pocket at least.
Edit: what is this, Reddit?
> Edit: what is this, Reddit?
Pick a lane brother
for assistant glasses to be useful, you need to know who, what, when, where for each operation that the user cares about. In the experiements that I ran, that took about 35 watts continuous on the headset (using a jetson) and at least 2 big GPU machines in the backend. This was in 2024, so perhaps its not quite as heavy now. but its nowhere near 130mw
I don't think I've seen a phone take over enterprise the way that the Blackberry once dominated, now its either iPhone or Android. You could argue that your executives are safer on a more locked down system. You'd 100% want to get everyone who makes an MFA app or service on your device on the other hand.
RIM essentially invented MDM which is what allowed it to take off in the enterprise, via BlackBerry Enterprise Server. Android and iOS didn't fully unseat them in the enterprise until third party MDMs were up to par and lock down the consumer devices just as effectively.
So whatever new phone/OS that wants to take a stab at the enterprise, out of the gate, must support MDM profiles from all the major ones at least (InTune, ManageEngine, MaaS360, etc.). Then, as you said, all the MFA apps, at least should have Slack/Teams, office apps, etc.
Not that it can't be done, but the cost of entry is much higher today if anyone wants to overcome the switching costs.
User: Hello Uber support? Yes, the app shows my driver is in the ocean off the west coast of Africa? And his ETA is -2147483648 minutes?
Uber: Your app, your problem.
User: Hello phonebrain, can you fix my Uber app?
Phonebrain: You're absolutely right, I must have misunderstood the API docs. Squiggling... ... Shuffling ...
Phonebrain: Error, you've hit your usage limit. Please wait 12 hours, or switch to pay-as-you-go app development mode. I cannot estimate how much it will cost to fix your Uber app.
If an "AI Phone" looks to be an existential threat to third party app devs and services, except everyone to do everything they can kill said AI Phone to protect their own products.
To be clear I don’t think we are fully ready for this to work. But directionally the trend looks significant.
No App Store, only web apps, not even 3G, only Yahoo push mail, and tied to AT&T - who were loathed, because in many places they just sucked. And it succeeded anyway.
It was when Exchange push support came in the first major OS upgrade that corporate adoption exploded.