They didn't, they asked an LLM to create a webpage, and it did, and they didn't notice that particular bit of sloppery, any more than people who post AI-generated memes on Facebook don't notice that their niece has six fingers now.
They can absolutely keep them. Or they can just "lose" them "accidentally". Who exactly would force them to give them back? Or put another way, who exactly will punish them if they break the law?
They can’t keep them permanently, but unfortunately they have in the past kept them from some people for years, until the hardware was past its useful life. It’s a good idea to only travel with electronics you don’t mind losing (and not just because of this).
You have a worldview that’s incompatible with the reality of the current US legal system where things work the way you believe they should, rather than the way they actually do.
Morality and direct commonsense interpretations of law do not apply when there are literally unlimited resources stacked against you. But, assuming you can wait the potential ~10y to receive your device back that it will take to get your device returned to you, good on you. If you think that the current SCOTUS will rule in your favor, good on you.
The reality is, we live in a time where the most horrendous interpretation of the law is the one that will happen. And it won’t be in your favor.
I don’t see how what you said contradicts anything I said, and I suspect you didn’t even read it. Yes, people have before had their devices seized for years, and courts wouldn’t return it sooner. That’s literally what I said.
Honestly, I think this is still fine for most people. The probability of a border agent asking me to unlock my phone is very, very low. The inconvenience of using a burner phone is high.
If they do take my phone (completely shut down, unlikely they'll be able to break in) and it's gone forever, that sucks, but then I get a new phone, restore from a backup, and move on with my life. Given that the probability of getting to this point is very low, I'm comfortable with the risk.
But sure, if I was at high risk of being detained at the border due to my profession, country of origin, ethnicity, etc., I'd probably look at this differently.
Sometimes you can use a semicolon where you use an em dash, but that's only coincidence: They serve different functions and usually you can't.
Rae bought all the candy at the store - Milky Way, Reese's, M&Ms, everything - and set out for Kay's party.
That can't be rewritten as,
Rae bought all the candy at the store; Milky Way, Reese's, M&Ms, everything; and set out for Kay's party.
Semicolons set off independent clauses - what would otherwise be complete sentences. Em dashes create breaks in the flow of the sentence; commas are more similar.
This is a fun challenge, I think: make the AI generate text with no obvious AI shibboleths. I haven't found a good way to do it at all yet, the more you ask it not to, the more you lower the output quality, which paradoxically makes it more sloppy, not less. Funnily enough I think "lower-power" LLMs are much better at this, like Haiku is much better at keeping voice on track with those kinds of instructions.
I have done it for a long time now. It's actually pretty straight forward to do it, but you have to run it in a pipeline, and you have to add deterministic checks as well.
There was an odd key to unlocking it sounding like a human. I spent like a month on this topic last year, and I randomly stumbled upon something that ended up massively improving generated tests.
It was a shot in the dark after failing so often. And then it suddenly sounded good.
I took off work to drive from Houston to Little Rock to see the eclipse at 100% and I'm so glad I did. I can't imagine prioritizing a temporary job over a once in a lifetime event.
I had a colleague who really wanted to see the eclipse but felt bad about taking off from work, so he didn't go. He ended up being laid off within a year. I feel terrible for him. I also learned how important it is to keep perspective on what matters.
They don't need to reason that evidence was destroyed. They just need to reason they there was an attempt to destroy potential evidence, whether or not there was actually any evidence. Any kind of interference with an investigation in any way is likely subject to prosecution.
The ability to start a Claude Code session in a temporary "config free" mode, similar to how incognito works for chrome, would be a nice feature to make this easy to try out, especially since model releases are so frequent.
Yep! Only (potential) downside to --bare is it runs in print mode; can't kick off an interactive session with it. The flags I listed above are a close approximation.
Anthropic had something about API-only in the description a couple months back, but they've been waking all those back for months since the fable access rollback fallout
reply