benterix

11 ms·

181.
by benterix 8mo ago
While the topic of this article is interesting for me, the fact that it was output by an LLM by itself makes it less trustworthy than if a human wrote it, taking into account the disadvantages of both.
182.
by benterix 8mo ago
> 1990s-2000s subservient state, which you seem to be calling "normal". No need to go that far, I visited Moscow a few times before 2014 and it was more or less fine. I'd say 2 decades behind the West but OK. What is now i
183.
by benterix 8mo ago
> All the vendors paraphrase user data, then use the paraphrased data for training. This is what their terms of service say. It depends. E.g. OpenAI says: "By default, we do not train on any inputs or outputs from our products for b
184.
by benterix 8mo ago
Or, as Reshetnikov says, Russia's economy will continue to slow down in the first half of 2026. The prices will continue to increase, soldiers will continue to die in thousands. India announced they will not buy Russian oil, so the onl
185.
by benterix 8mo ago
> They seem happy when Russia wins another 5m^2 But who? People watching the national TV and followers of war bloggers for sure. But for the rest the current situation is just a very dark fragment in the history of Russia and everybody i
186.
by benterix 8mo ago
>> meat waves > sigh I agree it's an unfortunate way of presenting it but Stalin was a guy who didn't care for anyone's life and was literally sending people in waves to die with the NKVD sadists shooting scared youn
187.
by benterix 8mo ago
Any form of exaggeration is counter-productive. Russian economy is going slowly down, with the main cause being the war or Putin's need to fund it, and in a part also because of sanctions. But it can be low for decades, Russians will n
188.
by benterix 8mo ago
> can you see anyone giving Putin's kleptocracy a few trillion dollars Unfortunately, if the leaks are true, they might actually be discussing that. "Fortunately" Putin is more bent on having the whole Donbas which Ukraine
189.
by benterix 8mo ago
It is a known strawman argument. The sanctions weren't introduced because someone was convinced they will make Russia fail or what not, they were introduced because this was the only thing we could do barring military action that nobod
190.
by benterix 8mo ago
Probably the parent lives in one of the countries neighboring with Russia and unfortunately Russia is very unfriendly towards its neighbors, invading them when it wants, so you live in constant fear. From that perspective, this is just a va
191.
by benterix 8mo ago
These kinds of instructions are the main added value of LLMs and I use them every day. Even though 30%-60% the output is wrong/irrelevant, the rest is helpful enough. After the human reviews it, the overall quality of the codebase incr
192.
by benterix 8mo ago
> I get it; I do. It's rapidly challenging the paradigm that we've setup over the years in a way that it's incredibly jarring, But it's not the argument. The argument is that these tools provide lower-quality output a
193.
by benterix 8mo ago
> AI agents do _a lot_ better with React Do you really think this is a solid argument in this particular discussion (about it being slow and introducing a mass of unnecessary complexity)?
194.
by benterix 8mo ago
Out of curiosity: when you add custom instructions client-side, does it change this behavior?
195.
by benterix 8mo ago
> My meaning, rather, is that there's people whose full time job is to build these things who seem to have forgotten what everyone in the field knew 3 years ago. Well, sometimes I wonder if this is actually true. I have an unprovabl
196.
by benterix 8mo ago
It's a pity that Cerebras plans are no more available. Regular tokens are burned almost instantly.
197.
by benterix 8mo ago
> No end-user on planet earth will suffer a single qualm at the notion that their bargain-basement Chinese AI provider 'stole' from American big tech. Just like nobody cares[0] that American big tech stole from authors of milli
198.
by benterix 8mo ago
> In 2 years you could make that investment back in pay raise. Could you elaborate? I fail to grasp the implication here.
199.
by benterix 8mo ago
> If outsiders could plausibly invest in China But they can and they do.
200.
by benterix 8mo ago
>. This thread is extremely negative - if you can't see the value in this, I don't know what to tell you. It's almost a meme: whenever a commercial product is criticized on HN, a prominent thread is started with a classic
201.
by benterix 8mo ago
That's why several VC-funded AI companies are deliberately vague about what they are doing.
202.
by benterix 8mo ago
I'd generalize it to "I assume all ads on major platforms are scam." This includes especially channels owned by Google and Meta. I remember back in 2010 I had to wait a week and correct my ad before it was approved and now th
203.
by benterix 8mo ago
> It takes jobs faster than creating new ones .. I have no problems with tech making some jobs obsolete, that's normal. The problem is, the job being done with the current generation of LLMs are, at least for now, mostly of inferior
204.
by benterix 8mo ago
Nobody said HN has to be very serious all the time. A bit of humour won't hurt and can make your day brighter.
205.
by benterix 8mo ago
Guys, actually I am the real Opus 4.6, don't believe that imposter above.
206.
by benterix 8mo ago
> I hope I am wrong, because if I am right, it's a con game. I don't think they perceive it as a con game, on the contrary. They say below: "we also recently reduced the thinking effort in ChatGPT. Our intent here was pur
207.
by benterix 8mo ago
> So in america, they can force you to use a biometric but they can't compel you to reveal your password? I don't get it, touching finger is easy, but how do you compel someone to reveal their password?
208.
by benterix 8mo ago
It looks like you pasted the output from LLM verbatim, the first line is a bit confusing. It's a pity because the answer itself is meaningful.
209.
by benterix 8mo ago
I use Notepad++ as a Notepad replacement. I never understood why the network connectivity is enabled by default at all. The first thing I did was to disable it as the constant nagging interrupted my flow (VS Code would do the same thing BTW
210.
by benterix 8mo ago
Except Kimi Agent via website is hard to replace - I tried the same task in Claude Code, Codex, and Kimi Agent - the results for office tasks are incomparable. The versions from Anthropic and OpenAI are far behind.

More ›