Discussion
Simon Willison’s Weblog
varispeed: Before Opus 4.7, the 4.6 became very much unusable as it has been flagging normal data analysis scripts it wrote itself as cyber security risk. Got several sessions blocked and was unable to finish research with it and had to switch to GPT-5.4 which has its own problems, but at least is not eager to interfere in legitimate work.edit: to be fair Anthropic should be giving money back for sessions terminated this way.
embedding-shape: > The new <acting_vs_clarifying> section includes: When a request leaves minor details unspecified, the person typically wants Claude to make a reasonable attempt now, not to be interviewed first.Uff, I've tried stuff like these in my prompts, and the results are never good, I much prefer the agent to prompt me upfront to resolve that before it "attempts" whatever it wants, kind of surprised to see that they added that
sigmoid10: I knew these system prompts were getting big, but holy fuck. More than 60,000 words. With the 3/4 words per token rule of thumb, that's ~80k tokens. Even with 1M context window, that is approaching 10% and you haven't even had any user input yet. And it gets churned by every single request they receive. No wonder their infra costs keep ballooning. And most of it seems to be stable between claude version iterations too. Why wouldn't they try to bake this into the weights during training? Sure it's cheaper from a dev standpoint, but it is neither more secure nor more efficient from a deployment perspective.
mysterydip: I assume the reason it’s not baked in is so they can “hotfix” it after release. but surely that many things don’t need updates afterwards. there’s novels that are shorter.
sigmoid10: Yeah that was the original idea of system prompts. Change global behaviour without retraining and with higher authority than users. But this has slowly turned into a complete mess, at least for Anthropic. I'd love to see OpenAI's and Google's system prompts for comparison though. Would be interesting to know if they are just more compute rich or more efficient.
formerly_proven: Surely the system prompt is cached across accounts?
winwang: That's usually not how these things work. Only parts of the prompt are actually loaded at any given moment. For example, "system prompt" warnings about intellectual property are effectively alerts that the model gets. ...Though I have to ask in case I'm assuming something dumb: what are you referring to when you said "more than 60,000 words"?
sigmoid10: What you're describing is not how these things usually work. And all I did was a wc on the .md file.
