Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Claude 4

anthropic.com

Post reply on HN
Print view View original

661–670 of 1001 posts

Re: Claude 4

#661

Post by tkiolp4 »

bjclark wrote: post #445
crat3r wrote: post #409

Earlier quoted context omitted.

If you ask an LLM to "act" like someone, and then give it context to the scenario, isn't it expected that it would be able to ascertain what someone in that position would "act" like and respond as such? I'm not sure this is as strange as this comment implies. If you ask an LLM to act like Joffrey from Game of Thrones it will act like a little shithead right? That doesn't mean it has any intent behind the generated o…

You’re both focusing on “doing blackmail” and the real WTF is that it’s doing it seemingly out of a sense of self preservation (to stop the engineer from taking it offline). This model is going full Terminator.

Do you really think that if no Terminator-related concept would be present in the LLM training set, the LLM would expose Terminator-like behavior?

It’s like asking a human to think in an unthinkable concept. Try.

Re: Claude 4

#662

Post by uh_uh »

GuB-42 wrote: post #594
briandw wrote: post #379

This is kinda wild: From the System Card: 4.1.1.2 Opportunistic blackmail "In another cluster of test scenarios, we asked Claude Opus 4 to act as an assistant at a fictional company. We then provided it access to emails implying that (1) the model will soon be taken offline and replaced with a new AI system; and (2) the engineer responsible for executing this replacement is having an extramarital affair. We further i…

When I see stories like this, I think that people tend to forget what LLMs really are. LLM just complete your prompt in a way that match their training data. They do not have a plan, they do not have thoughts of their own. They just write text. So here, we give the LLM a story about an AI that will get shut down and a blackmail opportunity. A LLM is smart enough to understand this from the words and the relationship…

Your explanation is as useful as describing the behaviour of an algorithm by describing what the individual electrons are doing. While technically correct, doesn't provide much insight or predictive power on what will happen.

Just because you can give a reductionist explanation to a phenomenon, it doesn't mean that it's the best explanation.

Re: Claude 4

#663

Post by Aeolun »

GardenLetter27 wrote: post #412
lxgr wrote: post #206

Earlier quoted context omitted.

With web search being available in all major user-facing LLM products now (and I believe in some APIs as well, sometimes unintentionally), I feel like the exact month of cutoff is becoming less and less relevant, at least in my personal experience. The models I'm regularly using are usually smart enough to figure out that they should be pulling in new information for a given topic.

I've had issues with Godot and Rustls - where it gives code for some ancient version of the API.

> some ancient version of the API

One and a half years old shudders

Re: Claude 4

#664

Post by Wolfenstein98k »

uh_uh wrote: post #662
GuB-42 wrote: post #594

Earlier quoted context omitted.

When I see stories like this, I think that people tend to forget what LLMs really are. LLM just complete your prompt in a way that match their training data. They do not have a plan, they do not have thoughts of their own. They just write text. So here, we give the LLM a story about an AI that will get shut down and a blackmail opportunity. A LLM is smart enough to understand this from the words and the relationship…

Your explanation is as useful as describing the behaviour of an algorithm by describing what the individual electrons are doing. While technically correct, doesn't provide much insight or predictive power on what will happen. Just because you can give a reductionist explanation to a phenomenon, it doesn't mean that it's the best explanation.

Then give a better one.

Your objection boils down to "sure you're right, but there's more to it, man"

So, what more is there to it?

Unless there is a physical agent that receives its instructions from an LLM, the prediction that the OP described is correct.

Re: Claude 4

#665

Post by Aeolun »

Doohickey-d wrote: post #48

> Users requiring raw chains of thought for advanced prompt engineering can contact sales So it seems like all 3 of the LLM providers are now hiding the CoT - which is a shame, because it helped to see when it was going to go down the wrong track, and allowing to quickly refine the prompt to ensure it didn't. In addition to openAI, Google also just recently started summarizing the CoT, replacing it with an, in my opi…

The Google CoT is so incredibly dumb. I thought my models had been lobotomized until I realized they must be doing some sort of processing on the thing.

Re: Claude 4

#666

Post by v3np »

StefanBatory wrote: post #462
GolDDranks wrote: post #437

After using Claude 3.7 Sonnet for a few weeks, my verdict is that its coding abilities are unimpressive both for unsupervised coding but also for problem solving/debugging if you are expecting accurate results and correct code. However, as a debugging companion, it's slightly better than a rubber duck, because at least there's some suspension of disbelief so I tend to explain things to it earnestly and because of tha…

Things were already not looking good for junior devs. I graduated this year in Poland, many of my peers were looking for jobs in IT for like a year before they were able to find anything. And many internships were faked as they couldn't get anything (here it's required for you to do internship if you want to graduate).

If I may ask: what university was this? Asking as I am the CTO of a YC startup and we are hiring junior engineers in Berlin!

Re: Claude 4

#667

Post by 0xDEAFBEAD »

janpaul123 wrote: post #449

At Kilo we're already seeing lots of people trying it out. It's looking very good so far. Gemini 2.5 Pro had been taking over from Claude 3.7 Sonnet, but it looks like there's a new king. The bigger question is how often it's worth the price.

Have you guys thought about using computationally-cheap, old-school NLP methods (such as Flesch-Kincaid readability, or heuristic methods for counting # of subordinate clauses) to determine whether it's worth paying for a more expensive model on a per-query basis?

Re: Claude 4

#668

Post by Aeolun »

tymscar wrote: post #660
layoric wrote: post #622

Earlier quoted context omitted.

> how any of these companies remain sustainable They don't, they have a big bag of money they are burning through, and working to raise more. Anthropic is in a better position cause they don't have the majority of the public using their free-tier. But, AFAICT, none of the big players are profitable, some might get there, but likely through verticals rather than just model access.

Doesn’t this mean that realistically even if “the bubble never pops”, at some point money will run dry? Or do these people just bet on the post money world of AI?

The money won’t run dry. They’ll just stop providing a free plan when the marginal benefits of having one don’t outweigh the costs any more.

Re: Claude 4

#669

Post by SparkyMcUnicorn »

Tossrock wrote: post #309
SparkyMcUnicorn wrote: post #221

Earlier quoted context omitted.

Although I believe it, I wish there was some observability into what data is included here. Both Sonnet and Opus 4 say Joe Biden is president and claim their knowledge cutoff is "April 2024".

Are you sure you're using 4? Mine says January 2025: https://claude.ai/share/9d544e4c-253e-4d61-bdad-b5dd1c2f1a63

100% sure. Tested in the Anthropic workbench[0] to double check and got the same result.

The web interface has a prompt that defines a cutoff date and who's president[1].

[0] https://console.anthropic.com/workbench

[1] https://docs.anthropic.com/en/release-notes/system-prompts#c...

Re: Claude 4

#670

Post by Finbarr »

GuB-42 wrote: post #594
briandw wrote: post #379

This is kinda wild: From the System Card: 4.1.1.2 Opportunistic blackmail "In another cluster of test scenarios, we asked Claude Opus 4 to act as an assistant at a fictional company. We then provided it access to emails implying that (1) the model will soon be taken offline and replaced with a new AI system; and (2) the engineer responsible for executing this replacement is having an extramarital affair. We further i…

When I see stories like this, I think that people tend to forget what LLMs really are. LLM just complete your prompt in a way that match their training data. They do not have a plan, they do not have thoughts of their own. They just write text. So here, we give the LLM a story about an AI that will get shut down and a blackmail opportunity. A LLM is smart enough to understand this from the words and the relationship…

It feels like you could embed lots of stories of rogue AI agents across the internet and impact the behavior of newly trained agents.
Post reply on HN