I had Claude write me a script to probe all the GCP Vertex models. I can say with confidence that I know how to call Argon, because Claude told me how, I did it, saw it was modestly better than Flash for my use case and I decided it wasn't worth the money right now, for this use case.
I'm fairly confident I can't call Carbon, and I am okay with that.
>I had Claude write me a script to probe all the GCP Vertex models. I can say with confidence that I know how to call Argon, because Claude told me how,
Wow that's embarrassing ! Is there no "product owners" or people "fighting for the users" anymore at Google ?
The problem with that is then you would have GPT 6.1 Max with max thinking which is different from GPT 6.1 Small with max thinking.
If you look back the naming problem started around the point that OpenAI started doing reasoning models (o1, 4o) where they had to describe this fork in model behavior that let a smaller model think longer and be smarter.
You can come up with naming schemes that don't clash, but there's something to be said for this approach where the thinking level is like you describe and the model name requires you to learn that Astra is the largest model.
That seems like revisionist history. Amazon started out with Prime Video bundled with Prime as a whole, and didn't raise the price. After 5-6 years they made it an add-on, then they added an ad-supported tier.
But if anything, the racket is bundling free shipping with ad-free video and everything else. Raising prices periodically to keep up with inflation is not a racket. And whether you would prefer to get downgraded to the advertising plan or automatically pay more is a personal choice, and automatically downgrading you rather than renewing you at a higher price is in some ways a more sensible default. (Though I'd prefer to be upgraded to the higher price generally and be notified so I have the option to downgrade or cancel, though downgrade really isn't an option I would consider.)
The primary issue is that many people prepaid for a year of Prime, which included ad-free streaming, and that ad-free service was changed mid-contract. I’d have no problem if the included streaming tier or overall price was changed at renewal.
That's fair, changing mid-year is definitely a problem and arguably illegal under current law, though really if we had functional antitrust I'm not sure that bundling would be legal to begin with.
They rolled out the new ToS to all customers, the original ToS they agreed to when signing up gave Disney permission to do that. There is nothing illegal going on here.
There's obviously a limit to what you can change via ToS. Yes, their ToS says they can alter the deal at their discretion. Adding ads to a premium ad-free tier seems like fundamentally a violation of the presumed contract. (Specifically talking about one-year plans, saying that they can make these kinds of fundamental alterations to the deal mid-year sounds like it ought to be illegal if they're selling one-year plans.)
They are assholes that lost my business forever for doing shady shit. I had paid for an annual plan with no-ads. It doesn't have to be illegal to be anti-consumer. I vote with my wallet.
The margins on GPUs are estimated at 80% right now. So the price would most likely drop by 70% just by building the additional manufacturing capacity. Assume some modest reductions in cost and the price should drop by 80-90%.
The only reason computers are so expensive is that demand is so out of control.
> I think it's worth noting that this compromise is much worse than the private keys simply being stolen by a researcher. They are inherently weak single-byte keys that anyone can factor, and the issue cannot be contained.
I would bet that there are some proofs whose simplest form are simply incomprehensible to humans. Explaining them in terms a human could possibly understand is a fool's errand. However the theorem itself may be much simpler, and the theorem is the thing that's most worth translating.
Attempting to translate the proof is valuable, but we have to be realistic about what is and isn't possible.
I would guess that minimum length proofs will end up like code golfing and demoscene hacks, where part of the code is also used as texture somewhere, etc. So many overlapping uses that don't make much sense but by happenstance allow making the proof shorter but less clear. The shortest proof is certainly not the most elegant. The shortest proof is also impossible to find in general due to halting problem etc. The AI isn't finding shortest proofs, and better AI will be better at finding clear and elegant proofs.
I've had pretty good luck having Gemini or Claude use SAM as a tool. Gemini is really good at identifying things in images, but it doesn't do a great job doing segmentation per se, but SAM on its own can't really decide what you need to segment.
They're looking for like, plumbers' email and quickbooks which tells you what kind of plumbing jobs they do and how much they get paid, so they know what kind of things to make robots for.
Well, it's an 'exploit' in the sense that normally we assume that the constant factors are modest enough so that O(n) can be reliant upon to be faster than O(2^n) or even than O(n^2) in practice.
I can't use a passkey on a device unless I can install my password manager on that device and I am comfortable giving that device access to my passkeys. Passwords are simply unparalleled in their flexibility. You're never going to be locked out because isn't working today.
I think the bit about TOTP was mostly a joke, the problem with passkeys is that they're billed as a replacement for passwords but they mix lots of MFA concerns in and make interoperability essentially impossible a lot of the time.
> I can't use a passkey on a device unless I can install my password manager on that device and I am comfortable giving that device access to my passkeys.
That's simply not true. I'm able to log onto GitHub on my corporate device using a passkey that's kept in 1Password on my iPhone.
That's fine if you've got a couple of supported devices. The hardware support is however very bad. If my hardware is unsupported (and most hardware in the world is unsupported) that's not very helpful.
This "use a passcode from another device such as my phone" tech mentioned above is just Bluetooth LTE. Every phone and most laptops and desktops in the last decade or so has Bluetooth LTE support. Hardware support is very good.
Passkeys are built on Webauthn (and ultimately sort of synonymous with it) and that has always been the web standard for all physical key support such as Yubikey and is basically the same as it has been for decades when dealing with Passkeys only with updated UX on top. Hardware support here is pretty good, too.
Bluetooth is very hard to use. This just isn't about hardware support, it's about "how long does it actually take the average person to do a Bluetooth-mediated workflow, how consistent is it between devices, if someone is trying it with a new device, what is the learning curve?" If someone can't figure out how it's supposed to work between their iPhone and their Windows laptop, there might theoretically be support. Practically speaking it's broken most of the time.
The deeper problem is that the default behavior is storing a passkey directly on a specific device and hardware-locking it to that device. That's great if you're an enterprise or whatever and you've got SSO set up well. For the average person, "I lose my iPhone and it effectively resets all my passwords" is a hopelessly unusable workflow.
I store passkeys in a 3rd party password manager which has an app on my phone, so I'm only relying on Apple to handle the bluetooth mediation and they do that very well.
I maybe lucky, but outside of headphones bluetooth has never failed to work for me.
Now with a 3rd party password manager, I can install it on any device and all my passwords work there too. However, with Apple, it stores it on keychain so if you have an internet connection, your passwords are backed up.
If you don't have an internet connection, you're not the average person.
That is a hopelessly unusable workflow which is why the default behavior on iOS is storing Passkeys in iCloud Passwords and having cloud sync between devices and cloud backup/restore.
On Windows you can use Apple's iCloud Passwords app for Passkeys now directly, in supported browsers because for whatever reason Apple's engineers didn't want to build a Windows plugin for it (yet?) or you can use the Bluetooth option. Windows 11 makes it very easy to use the Bluetooth option and for better/worse some of the requirements that made Windows 11 stop supporting older hardware were also those making sure that Bluetooth LTE is reliable especially for Passkeys.
In the Bluetooth flow the "only" user facing mediation step is the QR code. Some of Microsoft, Apple, and Google's finest all worked together to make sure that UX is about as simple as it can be with just about no learning curve. If someone is used to pulling up a QR code in the camera app and clicking the default app it wants to pop up, "it just works".
If you hear Bluetooth and think of complex pairing dances, that is nothing like how Passkey signings over Bluetooth work. The devices don't need to be "joined" or "paired", they just need clear enough radios for a quick couple of messages, which a QR code kicks off.
I'm not sure I see the difference between a complex pairing dance and needing to install the iCloud Passwords app on Windows (I've used iTunes for Windows.) Writing down a password is much more reliable than expecting any of that to work.
I'm just trying to point out that you seem to be fighting strawmen and yelling about last decade's apps and technology as if that was the current state.
Writing down a password is simpler, but it is also stopped being reliable for the vast majority of users because it stopped being adequate security for the average user. Passwords are too easily stolen. Passwords are too easily reused because human memory is quite fallible. Passwords are too easily caught by a man-in-the-middle attack or a social engineering attack. Passwords are too easily leaked in a database breach. Even when properly hashed and salted, password hashes still are weak to rainbow tables and other such attacks.
Passkeys trade off a bit of supposed simplicity (if you ignore modern advice like use different passwords for everything and always use a password manager because you will never remember all your passwords, it's not humanly possible if they are also all secure enough passwords). But the tech is already at a very usable level of simplicity for average users. It's just like learning a password manager. For some average users it is even easier than learning a password manager the standards and standardized user flows have simplified a lot of the hard edges of the password manager, including the above mentioned "use a QR code to provide temporary access on some other device to your Passkey without syncing your full Password Manager".
Most people don't use password managers. I am not even sure that password managers really qualify as a reasonable amount of complexity for the average person. There's a cost/benefit and I think passwords + SMS 2-step-auth is a reasonable balance between usability and security for most people.
Obviously, I use a password manager and I use MFA where possible, but even that I find to be a little questionable from an availability standpoint because I have a hard time ensuring that I have multiple MFA devices located in different buildings, so a fire or a flood or something could totally lock me out of all my accounts, some of them possibly permanently, which is the thing that scares me more than getting hacked.
Passkeys (combined with full-disk-encryption by default) scare me because they make the assumption that most people would prefer to lose access to their photo library than risk someone breaking in and getting them, and I actually think most people would prefer the converse. If a burglar breaks into my house they can steal my TV, I can live without it, but please don't steal my photo albums (problem is they're the same thing.)
That's not really true, a QR code pops up, and you can use your other device to read it.
Even if your other device is a mainframe computer thousands of miles away, so long as you can photograph the QR code, you can decode it and send the data string for the passkey device needs to it.
The issue that many have with passkeys is that someone will tie a passkey to an ecosystem that they don't personally control like Apple's and then they can't get into their services without an apple device that's onboarded into Apple's ecosystem.
The other device has to respond "in nearby radio space" via Bluetooth LTE when using the QR code bootstrap, but that's about the only requirement and it can be just about any device.
But you can be locked out if the provider decides you need to change your password after a breach, or you forget the password, or you enter the wrong password in too many times.
They’re both equally disposable. You can reset a password just like you can reset a passkey.
I'm fairly confident I can't call Carbon, and I am okay with that.
reply