Live data from Hacker News

Claude 4 System Card

simonwillison.net

121–130 of 264 posts

Re: Claude 4 System Card

#121
post #113

I just published a deep dive into the Claude 4 system prompts, covering both the ones that Anthropic publish and the secret tool-defining ones that got extracted through a prompt leak. They're fascinating - effectively the Claude 4 missing manual: https://simonwillison.net/2025/May/25/claude-4-system-prompt...

Truly fascinating, thanks for this. What I find a little perplexing is when AI companies are annoyed that customers are typing "please" in their prompts as it supposedly costs a small fortune at scale yet they have system prompts that take 10 minutes for a human to read through.

Hah, yeah I think that "please" thing was mainly Sam Altman flexing about how many users ChatGPT has.

Anthropic announced that they increased their maximum prompt caching TTL from 5 minutes to an hour the other day, not surprising that they are investigating effort in caching when their own prompts are this long!

Re: Claude 4 System Card

#122
post #113

I just published a deep dive into the Claude 4 system prompts, covering both the ones that Anthropic publish and the secret tool-defining ones that got extracted through a prompt leak. They're fascinating - effectively the Claude 4 missing manual: https://simonwillison.net/2025/May/25/claude-4-system-prompt...

Truly fascinating, thanks for this. What I find a little perplexing is when AI companies are annoyed that customers are typing "please" in their prompts as it supposedly costs a small fortune at scale yet they have system prompts that take 10 minutes for a human to read through.

I assume that they run the system prompt once, snapshot the state, then use that as starting state for all users. In that sense, system prompt size is free.

EDIT: Turns out my assumption is wrong.

Re: Claude 4 System Card

#123
post #24

It’s honestly a little discouraging to me that the state of “research” here is to make up sci fi scenarios, get shocked that, e.g., feeding emails into a language model results in the emails coming back out, and then write about it with such a seemingly calculated abuse of anthropomorphic language that it completely confuses the basic issues at stake with these models. I understand that the media laps this stuff up s…

> It’s honestly a little discouraging to me that the state of “research” here is to make up sci fi scenarios,

it's not research, it's marketing

the main aim of this "research" is making sure that you focus on this absurd risk, and not on the real risk: the inherent and unfixable unreliability of these systems

what they want is journalists to read through the "system card", spot this tripe and produce articles with titles like "Claude 4 is close to becoming Skynet"

they then get billions of free publicity, and a never ending source braindead investors with buckets of money

additionally: it worries clueless CEOs, who then rush to introduce AI internally in fear of being competed out of business by other sloppers

these systems are dangerous because of their inherent unreliability, they will cause untold damage if they end up in control systems

but the blackmail is simply parroting some fiction that was in its training set

Re: Claude 4 System Card

#124
post #51

Earlier quoted context omitted.

Exactly. It's getting to the point where the quality of the top AI labs are either not ground-breaking (except Google Gemini Diffusion) and labs are rushing to announce their underwhelming models. Llama as an example. Now in the next 6 months, you'll see all the AI labs moving to diffusion models and keep boasting around their speed. People seem to forget that Google Deepmind can do more than just "LLMs".

Google's output this IO was really impressive. The diffusion LLM but especially veo3 was something else.

I mean I'm gonna say this with the hype settling down. But it's pretty on par with visually Kling 2 and Veo 2, it happens to output sound pretty ok but having it be one general output along with the visuals is the gamechanger. Beyond that, eh. I've kinda seen people try to take it to the limit and it's pretty much what you'd expect still from their last model

Re: Claude 4 System Card

#125
post #86

> 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. Isn't that a showstopper for agentic use? Someone sends an email or publishes fake online stories that convince the agentic AI that it's working for a bad guy, and it'll take "very bold action" to bring ruin to the owner.

My mind went straight to “and now law enforcement is going to need agents handling phone calls to deal with the volume of agents calling them”.

Re: Claude 4 System Card

#126

OT > data provided by data-labeling services and paid contractors someone in my circle was interested in finding out how people participate in these exercises and if there are any "service providers" that do the heavy lifting of recruiting and managing this workforce for the many AI/LLM labs globally or even regionally they are interested in remote work opportunities that could leverage their (post-graduate level) ed…

My Reddit feed is absolutely spammed with data annotation job ads, looking specifically for maths tutors and coders. Does not feel like roles with long-term prospects.

Lots of job offer spam in this area as well. See one or two a week.

Re: Claude 4 System Card

#127
post #124

Earlier quoted context omitted.

Google's output this IO was really impressive. The diffusion LLM but especially veo3 was something else.

I mean I'm gonna say this with the hype settling down. But it's pretty on par with visually Kling 2 and Veo 2, it happens to output sound pretty ok but having it be one general output along with the visuals is the gamechanger. Beyond that, eh. I've kinda seen people try to take it to the limit and it's pretty much what you'd expect still from their last model

I think veo2 and Kling are very strong models, but the fact that veo3 is end2end video/audio including lipsync and all other sound is definitely a step change to what came before and I think you're underselling it.

I also expect Google to drive veo forward quite significantly, given the absurd amount of video training data that they sit on.

And compared to the cinemagraph level of video generation we were just 1-2 years ago, boy we've come a long way in very short amount of time.

Lastly, absurd content like this https://youtu.be/jiOtSNFtbRs crosses the threshold for me on what I would actually watch more of.

Veo3 level tech alone will decimate production houses, and if the trajectory holds a lot of people working in media production are in for a rude awakening.

Re: Claude 4 System Card

#128
post #30
post #6

Earlier quoted context omitted.

> to justify the full version increment I feel like a company doesn’t have to justify a version increment. They should justify price increases. If you get hyped and have expectations for a number then I’m comfortable saying that’s on you.

> They should justify price increases. I think the justification for most AI price increases should go without saying - they were losing money at the old price, and they're probably still losing money at the new price, but it's creeping up towards the break-even point.

Customers don’t decide the acceptable price based on the company’s cost structure. If two equivalent cars were priced $30k apart, you wouldn’t say “well, it seems like a lot but they did have unusual losses last year and stupidly locked themselves in to that steel agreement”. You’d just buy the less expensive one that meets the same needs.

(Almost) all producing is based I value. If the customer perceives the price fair for the value received, they’ll pay. If not, not. There are only “justifications” for a price increase: 1) it was an incredibly good deal at the lower price and remains a good deal at the higher price, and 2) substantially more value has been added, making it worth the higher price.

Cost structure and company economics may dictate price increases, but customers do not and should not care one whit about that stuff. All that matters is if the value is there at the new price.

Re: Claude 4 System Card

#130
post #42

Interesting! >Claude shows a striking “spiritual bliss” attractor state in self-interactions. When conversing with other Claude instances in both open-ended and structured environments, Claude gravitated to profuse gratitude and increasingly abstract and joyous spiritual or meditative expressions.

I think it was Larry Niven, quite a few decades ago, that had SF stories where AIs were only good for a few months before becoming suicidal...

Sort of reminds me of Rampancy from Halo.
Post reply on HN