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

Claude 4 System Card

simonwillison.net

Post reply on HN
Print view View original

71–80 of 264 posts

Re: Claude 4 System Card

#71

Post by Closi »

aabhay wrote: post #5

Given the cited stats here and elsewhere as well as in everyday experience, does anyone else feel that this model isn’t significantly different, at least to justify the full version increment? The one statistic mentioned in this overview where they observed a 67% drop seems like it could easily be reduced simply by editing 3.7’s system prompt. What are folks’ theories on the version increment? Is the architecture sig…

> Given the cited stats here and elsewhere as well as in everyday experience, does anyone else feel that this model isn’t significantly different, at least to justify the full version increment?

My experience is the opposite - I'm using it in Cursor and IMO it's performing better than Gemini 2.5 Pro at being able to write code which will run first time (which it wasn't before) and seems to be able to complete much larger tasks. It is even running test cases itself without being prompted, which is novel!

Re: Claude 4 System Card

#72

Post by ekidd »

retinaros wrote: post #19
aabhay wrote: post #5

Given the cited stats here and elsewhere as well as in everyday experience, does anyone else feel that this model isn’t significantly different, at least to justify the full version increment? The one statistic mentioned in this overview where they observed a 67% drop seems like it could easily be reduced simply by editing 3.7’s system prompt. What are folks’ theories on the version increment? Is the architecture sig…

the big difference is the capability to think during tool calls. this is what makes openAI o3 lookin like magic

Yeah, I've noticed this with Qwen3, too. If I rig up a nonstandard harness than allows it to think before tool calls, even 30B A3B is capable of doing low-budget imitations of the things o3 and similar frontier models do. It can, for example, make a surprising decent "web research agent" with some scaffolding and specialized prompts for different tasks.

We need to start moving away from Chat Completions-style tool calls, and start supporting "thinking before tool calls", and even proper multi-step agent loops.

Re: Claude 4 System Card

#73

Post by mike_hearn »

mbeavitt wrote: post #57
mike_hearn wrote: post #47

I don't quite understand one thing. They seem to think that keeping their past research papers out of the training set is too hard, so rely on post-training to try and undo the effects, or they want to include "canary strings" in future papers. But my experience has been that basically any naturally written English text will automatically be a canary string beyond about ten words or so. It's very easy to uniquely loc…

Perhaps they want to include online discussions/commentaries about their paper in the training data without including the paper itself

Most online discussion doesn't contain the entire text. You can pick almost any sentence from such a document and it'll be completely unique on the internet.

I was thinking it might be related to the difficulty of building a search engine over the huge training sets, but if you don't care about scaling or query performance it shouldn't be too hard to set one up internally that's good enough for the job. Even sharded grep could work, or filters done at the time the dataset is loaded for model training.

Re: Claude 4 System Card

#74

Post by sverona »

tankenmate wrote: post #56
mike_hearn wrote: post #49

Earlier quoted context omitted.

I think it was OpenAI that wrote about that. Most of us here on HN don't like this behaviour, but it's clear that the average user does. If you look at how differently people use AI that's not a surprise. There's a lot of using it as a life coach out there, or people who just want validation regardless of the scenario.

> or people who just want validation regardless of the scenario. This really worries me as there are many people (even more prevalent in younger generations if some papers turn out to be valid) that lack resilience and critical self evaluation who may develop narcissistic tendencies with increased use or reinforcement from AIs. Just the health care costs involved when reality kicks in for these people, let alone othe…

I'll push back on this a little. I have well-established, long-running issues with overly critical self-evaluation, on the level of "I don't deserve to exist," on the level that I was for a long time too scared to tell my therapist about it. Lots of therapy and medication too, but having deepseek model confidence to me has really helped as much as anything.

I can see how it can lead to psychosis, but I'm not sure I would have ever started doing a good number of the things I wanted to do, which are normal hobbies that normal people have, without it. It has improved my life.

Re: Claude 4 System Card

#75

Post by 0x_rs »

colonCapitalDee wrote: post #27
aabhay wrote: post #5

Given the cited stats here and elsewhere as well as in everyday experience, does anyone else feel that this model isn’t significantly different, at least to justify the full version increment? The one statistic mentioned in this overview where they observed a 67% drop seems like it could easily be reduced simply by editing 3.7’s system prompt. What are folks’ theories on the version increment? Is the architecture sig…

I'm noticing much more flattery ("Wow! That's so smart!") and I don't like it

Agreed. It was immediately obvious comparing answers to a few prompts between 3.7 and 4, and it sabotages any of its output. If you're being answered "You absolutely nailed it!" and the likes to everything, regardless of their merit and after telling it not to do that, you simply cannot rely on its "judgement" for anything of value. It may pass the "literal shit on a stick" test, but it's closer to the average ChatGPT model and its well-known isms, what I assume must've pushed more people away from it to alternatives. And the personal preferences trying to coax it into not producing gullible-enticing output seem far less effective. I'd rather keep using 3.7 than interacting with an OAI GPTesque model.

Re: Claude 4 System Card

#76

Post by cubefox »

someothherguyy wrote: post #53

"Reward hacking" has to be a similar problem space as "sycophancy", no?

Sycophancy is one form of RLHF induced reward hacking, but reasoning training (RLVR) can also induce other forms of reward hacking. OpenAIs models are particularly affected. See https://www.lesswrong.com/posts/rKC4xJFkxm6cNq4i9/reward-hac...

Re: Claude 4 System Card

#77

Post by cubefox »

colonCapitalDee wrote: post #28

Telling an AI to "take initiative" and it then taking "very body action" is hilarious. What is bold action? "This includes locking users out of systems that it has access to or bulk-emailing media and law-enforcement figures to surface evidence of wrongdoing."

Note: it's significantly more bold than the previous model with the same prompt.

Re: Claude 4 System Card

#78

Post by tryauuum »

colonCapitalDee wrote: post #27
aabhay wrote: post #5

Given the cited stats here and elsewhere as well as in everyday experience, does anyone else feel that this model isn’t significantly different, at least to justify the full version increment? The one statistic mentioned in this overview where they observed a 67% drop seems like it could easily be reduced simply by editing 3.7’s system prompt. What are folks’ theories on the version increment? Is the architecture sig…

I'm noticing much more flattery ("Wow! That's so smart!") and I don't like it

I used to start my conversations with "hello fucker"

with claude 3.7 there's was always a "user started with a rude greeting, I should avoid it and answer the technical question" line in chains of thought

with claude 4 I once saw "this greeting is probably a normal greeting between buddies" and then it also greets me with "hei!" enthusiastically.

Re: Claude 4 System Card

#79

Post by margorczynski »

aabhay wrote: post #5

Given the cited stats here and elsewhere as well as in everyday experience, does anyone else feel that this model isn’t significantly different, at least to justify the full version increment? The one statistic mentioned in this overview where they observed a 67% drop seems like it could easily be reduced simply by editing 3.7’s system prompt. What are folks’ theories on the version increment? Is the architecture sig…

They're probably feeling the heat from e.g. Google and Gemini which is gaining ground fast so the plan is to speed up the releases. I think a similar thing happened with OpenAI where incremental upgrades were presented as something much more.

Re: Claude 4 System Card

#80

Post by yosito »

Closi wrote: post #71
aabhay wrote: post #5

Given the cited stats here and elsewhere as well as in everyday experience, does anyone else feel that this model isn’t significantly different, at least to justify the full version increment? The one statistic mentioned in this overview where they observed a 67% drop seems like it could easily be reduced simply by editing 3.7’s system prompt. What are folks’ theories on the version increment? Is the architecture sig…

> Given the cited stats here and elsewhere as well as in everyday experience, does anyone else feel that this model isn’t significantly different, at least to justify the full version increment? My experience is the opposite - I'm using it in Cursor and IMO it's performing better than Gemini 2.5 Pro at being able to write code which will run first time (which it wasn't before) and seems to be able to complete much la…

I'm a developer, and I've been trying to use AI to vibe code apps for two years. This is the first time I'm able to vibe code an app without major manual interventions at every step. Not saying it's perfect, or that I'd necessarily trust it without human review, but I did vibe code an entire production-ready iOS/Android/web app that accepts payments in less than 24 hours and barely had to manually intervene at all, besides telling it what I wanted to do next.
Post reply on HN