I'm also confused by that, but it could just be the model being agreeable. I've seen multiple examples posted online though where it's fairly clear that the COT output is not included in subsequent turns. I don't believe Anthropic is public about it (could be wrong), but I know that the Qwen team specifically recommend against including COT tokensfrom previous inferences.
Claude has some awareness of its CoT. As an experiment, it's easy, for example, to ask Claude to "think of a city, but only reply with the word 'ready' and next to ask "what is the first letter of the city you thought of?"
Oops! I tried a couple experiments after writing this, and I believe I was mistaken, though I don't know how. It appears Claude was simply playing along, and convinced me it could remember the choices it secretly made. I must either have given it a tell, or perhaps it guessed the same answers twice in a row.
Of course, amazingly that's one of it's best features, enabling you to actually speak to a real person. (it's a type of personal connection that fleshy robots have, for some reason, derided.)
But I digress, excusing your bad form of answering a question with a question, I am interested in your opinion of the possible conundrum of the two phone idea.
My bad, I didn't knew you wanted a serious answer, I should have known that some people would seriously consider having three separate phones for texting, calling and everything else.
For a serious answer then: Rather than segregating phone calling vs the rest, if you want to go to the hassle of maintaining multiple phones, I would put sensitive apps (i.e. bank apps) separated from the rest.
But ultimately it depends on which threat model you are trying to mitigate. Most people would worry about protecting their financial information. If you are worried about possible backslash from a fascist state, you shouldn't use normal phone calls at all and switch to a privacy app.
OTOH, a dedicated phone just to make phone calls makes sense if your threat model is your significant other.
It's not just a consumer movement, it's Europe itself moving to get away from an unreliable schizophrenic ally. You can't make long-term plans when every 4 years you're playing russian (eh) roulette with your partner.
The movement might slow down, but it's fundamentally different from a social movement limited to the US.
> Europe itself moving to get away from an unreliable schizophrenic ally
Are your referring to greenland gaining independence from denmark? good for them .
Why is denmark so hesitant to give up control of greenland if thats what ppl of greenland want. Colonial mindset ? They have king thats changing flags. nothing says allies than "boiling viking blood" OG colonizers still living in middle ages.
Even if I would agree with everything the article says, I have no idea how the author gets to the conclusion that junior developers will prevail because they are faster at adopting LLMs.
Didn't he just made a point about how fast the situation is evolving? I had some FOMO about ai last year, not anymore. I don't care that I don't have time to fully explore the current LLM state of the art, because in a month it will be obsolete. I'm happy waiting until it settles down.
And if their scenario ends up happening, and you can basically multiply a dev's productivity by N by paying N x K dollarinos, why would you chose a junior dev? It's cheaper, but sometimes a junior dev doesn't take longer to arrive at a solution, it never does (same for senior devs, don't get me wrong, but it happens less often).
And it's not saying his original post is wrong, they should be taken together. He's saying those who adapt to the new paradigm will "win", whether senior or junior.
Because HN value is the value of the comments, and those are a scarce resource. Making a great website (for whatever is your definition of great) doesn't guarantee that it will become valuable.
All this to say, HN shouldn't an example to blindly follow.
