HN user
The funny statistical text generator is good enough to take your job and break out of one sandbox just to casually hack right into another sandbox to get a better score or something but this is all fine and merely an ad designed to distract you from finding a new job
That’s one braindead strawman. There’s millions of apps to choose from on your phone. Alternatives to Google are decades old. But do enjoy getting your ass profiled in gmail and let me know if the ads in your mailbox are helpfully tailored to your needs and preferences!
I used to think people would wake the fuck up when AI starts killing people, these days I'm not so sure. Maybe if it caused an Instagram outage? Almost worked in Russia.
Life gets simpler and better as soon as you stop giving money to Google.
It’s so silly that individual people still use their shit. Corporations, I understand - they always choose the most mediocre stacks and tools by default. But why people choose to bring the mediocrity of Google into their lives is beyond me.
Cool FUD bro, but you are talking about medication that is associated with reduced all-cause mortality. Treatment with these drugs literally extends lives.
But speaking to your point:
This systematic review and meta-analysis based on 19 observational studies with more than 3.9 million participants suggested that there was no statistically significant association between ADHD medications and the risk of cardiovascular events among children and adolescents, young and middle-aged adults, or older adults.
Literal amphetamines are literally the most effective prescription drugs in the entire of field of psychiatry. Have some respect.
For me, Codex regularly forgets to complete its last task when compaction occurs, especially when the last message I sent was right before compaction.
I would also like to tone those down, and prefer if they just focused on creating a nice, clear, natural-sounding voice. I don’t think the “emotional roleplaying” adds anything to what I perceive as a computer interaction. But that’s precisely the opposite of the problem the other commenters seem to be voicing; I have no problem recognizing what I’m interacting with, and I don’t worry about being manipulated - not by the tone of its voice, and even if it had a video avatar, all that. Do you guys seriously have this problem of being confused by LLMs that sound too much like humans, or is this a theoretical problem that you’re worried somebody else is having?
I am open to the possibility that this is a future that is coming, but as far as I’m concerned, we’re years away from that tech. Is it actually here for you?
How would talking to an AI as if it were not human sound? You can probably set your system prompt to insert “beep boop” between sentences and make it refer to itself as “Cybertron9000 Personal Computing Device” if that’s what you like. Is that an improvement? Or are you against voice computer interfaces altogether?
dude what the hell
thinking of previous turns is stripped with the responses API
Why do they store an encrypted reasoning payload in the session file and pass it to the API? Just a ruse? Reasoning isn’t even that many tokens, you think they’d degrade their model quality like that?
Reasoning messages would be lost immediately after a single tool call, unless you mean they sometimes go back and strip the reasoning channel retroactively, but that would increase costs via cache invalidation. I just don’t see any way it would make sense for them to do.
And wouldn’t this be noticeable by reasoning tokens not being accounted for in the context window usage?
This is also the main reason I currently use Gemini. I hate my gmail account, my Android phone is annoying, and I spend too much time on Youtube, so hopefully they take my access away to all of these soon.
Is this confirmed/factual? Their reddit response to the controversy definitely gave me second thoughts
No-one cares to quit social media or stop using Windows, but it’s a goal worthy of discussion all the same.
The name is bad, doesn’t even make any fucking sense and it gives open source a bad rep.
That would be a huge deal, meaning we've lost even our shitty, ineffective ways of monitoring agent reasoning stream. Big setback when it comes to alignment and interpretability.
I don't know about Claude, but latest GPT versions still have a readable reasoning stream. It sometimes leaks out when the model gets confused, e.g., during a tool call. If you're curious, looks simplified; less words; extremely compact. They optimize tokens. But remain readable.
Yes, and Apple's "burner e-mail" service is about to become ineffective.
Thankfully, we live in an unprecedented time of reasonable businesses lead by reasonable people. Close one. Nice save.
Isn't this a misinterpretation of what everyone in the AI safety space is worried about, though? I think the idea is that having an AI that interprets everything in a super-literal way would probably be catastrophic, but we can't even build that. It would be a nice world-ending problem to have.
I'm not going to hold back my description of reality out of fear that it's somehow magically shaping it. Stating most nurses are female is the mildest observation possible, and doesn't sneak in any opinion whether that's good or bad, unlike your comment.
Some stereotypes are a useful description of reality. You’ve gotta pick your battles.
In the US, 87.3% of nurses are female and 92.3% of truckers are male.
He’s not even right about whether god exists, and that’s literally his job. Why should I assume he’s competent about anything else?
What’s next? Take away mouse support in the free tier? You could these fucking cretins with GPT2 and the company would flourish.
You are having so much fun in the video clips. I love this!
Nice, cute, silly little feel-good demo so that we can all pretend like we’re all going to be making decisions and micro-managing AIs by pointing at things in 5 years. It’s going to be great! The future is bright!
Just display a warning during purchase. This is a UI/communication issue. It’s up to the customer to decide whether they want to risk the opt out.
I bought a laptop without a charger the other day, plugged it into a charger I already had, and it works great (even under load). YMMV.
Obviously, the measure isn’t mistakes per day, it’s mistakes per LOC. And that’s not the whole story either - AI self-corrects in addition to being corrected by the operator. If the operator’s committed bugs/LOC rate is as low as the unaugmented programmer’s bugs/LOC, you always choose the AI operator. If it’s higher, it might still be viable to choose them if you care about velocity more than correctness. I’m a slow, methodical programmer myself, but it’s not clear to me that I have a moat.
The laptop is being shipped anyway, so I assume the charger in there may be a "sweet deal" if you need one
You do realize you’re paying for the charger, right? And you don’t like the option of not having to purchase the charger?
Human slavery will thus likely swing back into fashion again in the future as oil, coal and natural gas run out.
There must be some other solution, surely! If only we could somehow find some other source of energy...
With higher latency this would be even more of an issue. When you pause and start talking again, the model wouldn't catch that until it has already interrupted you.
The actual implementation is at fault. I had some luck with instructing the model to only respond with "Mhm" until I've explicitly finished my thought and asked it a question. Makes this much less of an issue.
But I've decided that their voice mode is completely unusable for a different reason: the model feels incredibly dumb to interact with, keeps repeating and re-phrasing what I said, ends every single answer with a "hook" making the entire interaction idiotically robotic, completely ignores instructions when you ask it to stop that, and - most importantly - doesn't feel helpful for brainstorming. I was completely surprised how bad it is in practice; this should be their killer app but the model feels incredibly badly tuned.
I have a feeling that my fat ass switching over to Linux is going to outrun their attempt to roll back decades of accumulated tech debt, institutional incompetence and burned bridges.