I genuinely don't understand why (at least power users) users don't seem to understand this, they never "own" their account, thousand have lost everything due to a mistake, login once via anonymizing solution and so-on and bam, account ban, I've lost so much money in "crypto" (on some exchanges) because of 2-FA perma loss due to Gmail ban back in 2015, storing your business on Google and mixing-up with your personal life is just reckless at this point, it's not like people don't "know" that they can lose everything from 1 day to another, I hope this movement of moving out from Google will be much more generalized.
Yes, token use can be tracked the same way, just have to MITM everything. The ToS is a non-issue as it's not a legal issue, unless you plan to do business with Anthropic, not really an issue as you can always go to API later-on, in which case, Anthropic can't supposedly "ban you" as they are saying they don't record prompts.
I by myself use now more than 15 accounts combined of all providers + API as well for external providers, more than 50K$ equivalent a month in API tokens, my team is doing the same thing, it's not really that much once you figured out the real automation loops and workflows, solving 300 issues a day with guarantees is common.
I feel that a lot of users are still stuck on Claude code or tools like this and don't really have a real argument about why they are even following the thread at all, everything has to be async for serious automation, you shouldn't even be seeing what Claude or any other model is replying (everything has to be digested with another model to increase relevancy and accuracy of the message so you can read faster (like a bot)), it's irrelevant, only human in the loop when a decision must be made, the rest has to be loops with all model, typical e2e, regression, computer use test, video into frames into all model loop and so-on.
Yeah sure, going from an open-source social media algorithm to people throwing Nazi signs (not even the person that published it) is a bit completely out of place but alright, let's stay stuck on that for the next 50 years.
I'll be honest, cool yes, but forget about adoption, the first impression is that it's just a quick thing that is meant to die due to its copycat design, which downgrade the seriousness of the project itself.
Well I doubt anyone would be surprised with a backdoor in MS product, there have been many of them already, I frankly doubt anyone with "disk encryption" on Windows would think that it's NSA-proof (or script-kiddy clever, as shown in this article :))
To be fair it seems worse on the other side here on HN at least, I rarely see positive comments about Elon, so both sides seems to be doing the echo-chamber and defender mode, it's getting seriously absurd to not be able to have talk on technology itself without it getting into politic, especially for non-US residents :/
"could"? They for sure bought already all data they could put their hands-on, it's logical on a business perspective, they probably have a giant DB with a ton of world citizen, because why not?
Driving skill (and road manners) is also a serious issue, not only Internet quality (it's mostly solved nowadays with dual 5G/dual residential, Starlink is also available, np), getting a driver license is basically just paying a fixer for $200 (equivalent in PHP) and even if you attend the school genuinely and all, it's still super easy versus the west.
The user should be, user gets info from an LLM, a machine or a web post, it doesn't change anything, the one submitting the documents is the one responsible imo and it should remain that way.
It's not really applicable nowadays as you can just ask any agent to solve most linux issues for you, it's literally 1 prompt away and you can snapshot your FS to be safe prior and lock it down to avoid agent mistake.
I'm almost sure they do, for the sole reason that when you get a few M$ to set you for life like this, you'd rather start erasing all kind of proof possible (even if your opsec is really excellent) to slowly start building-up your new life, maintaining evidences anywhere is stupid and those guys are far from stupid for being able to pull stuff like this.
The alternative would be to ruin a part of the life of shitload of students, you find it better? It's not a dumb move at all, most companies pay ransom because the alternative is worse.
Right, however it's not really true in practice and we have stats for it. Generally it's enough money to set them for life and there is some sort of "moral code" around it as well, the same thing that allow darkmarkets to run.