Live data from Hacker News

Changes in the system prompt between Claude Opus 4.6 and 4.7

simonwillison.net

61–70 of 240 posts

Re: Changes in the system prompt between Claude Opus 4.6 and 4.7

#61
post #59

Earlier quoted context omitted.

When you are worth hundreds of billions, people start falling over themselves running to file lawsuits against you. We're already seeing this happen. So spending $50M to fund a team to weed out "food for crazies" becomes a no-brainer.

It's so shameful. We let people buy kitchen knives. But because the kitchen knife companies don't have billions of dollars, we don't go after them. We go after the LLM that might have given someone bad diet advice or made them feel sad. Nevermind the huge marketing budget spent on making people feel inadequate, ugly, old, etc. That does way more harm than tricking an LLM into telling you you can cook with glue.

I don’t feel like that’s a reasonable analogy. Kitchen knives don’t purport to give advice. But if a kitchen knife came with a label that said ‘ideal for murdering people’, I expect people would go after the manufacturer.

Re: Changes in the system prompt between Claude Opus 4.6 and 4.7

#62
post #4

I'm curious as to why 4.7 seems obsessed with avoiding any actions that could help the user create or enhance malware. The system prompts seem similar on the matter, so I wonder if this is an early attempt by Anthropic to use steering vector injection? The malware paranoia is so strong that my company has had to temporarily block use of 4.7 on our IDE of choice, as the model was behaving in a concerningly unaligned w…

No, you underestimate how huge the malware problem right now. People try publish fake download landing pages for shell scripts or even Claude code on https://playcode.io every day. They pay for google ads $$$ to be one the top 1 position. How Google ads allow this? They can’t verify every shell script.

No I am not joking. Every time you install something, there is a risk you clicked a wrong page with the absolute same design.

Re: Changes in the system prompt between Claude Opus 4.6 and 4.7

#63
post #5
post #4

I'm curious as to why 4.7 seems obsessed with avoiding any actions that could help the user create or enhance malware. The system prompts seem similar on the matter, so I wonder if this is an early attempt by Anthropic to use steering vector injection? The malware paranoia is so strong that my company has had to temporarily block use of 4.7 on our IDE of choice, as the model was behaving in a concerningly unaligned w…

I have started to notice this malware paranoia in 4.6, Boris was surprised to hear that in comments, probably a bug

more likely the paranoia behavior was backported. current gen is already being used for bug bounties.

Re: Changes in the system prompt between Claude Opus 4.6 and 4.7

#65
post #49
post #3

The acting_vs_clarifying change is the one I notice most as a heavy user. Older Claude would ask 3 clarifying questions before doing anything. Now it just picks the most reasonable interpretation and goes. Way less friction in practice.

I really hate that change, it's now regularly picking bad interpretation instead of asking.

Yeah, that really feels like a choice that should be user preference.

Re: Changes in the system prompt between Claude Opus 4.6 and 4.7

#66

The eating disorder section is kind of crazy. Are we going to incrementally add sections for every 'bad' human behaviour as time goes on?

Another way to think about it: every single user of Claude is paying an extra tax in every single request

Re: Changes in the system prompt between Claude Opus 4.6 and 4.7

#67
post #4

I'm curious as to why 4.7 seems obsessed with avoiding any actions that could help the user create or enhance malware. The system prompts seem similar on the matter, so I wonder if this is an early attempt by Anthropic to use steering vector injection? The malware paranoia is so strong that my company has had to temporarily block use of 4.7 on our IDE of choice, as the model was behaving in a concerningly unaligned w…

No, you underestimate how huge the malware problem right now. People try publish fake download landing pages for shell scripts or even Claude code on https://playcode.io every day. They pay for google ads $$$ to be one the top 1 position. How Google ads allow this? They can’t verify every shell script. No I am not joking. Every time you install something, there is a risk you clicked a wrong page with the absolute sam…

He's not talking about malware awareness. He's talking about a bug i've seen too which requires Claude justifying for *every* tool call to add extra malware-awareness turns. Like every file read of the repo we've been working on

Re: Changes in the system prompt between Claude Opus 4.6 and 4.7

#68
post #59

Earlier quoted context omitted.

When you are worth hundreds of billions, people start falling over themselves running to file lawsuits against you. We're already seeing this happen. So spending $50M to fund a team to weed out "food for crazies" becomes a no-brainer.

It's so shameful. We let people buy kitchen knives. But because the kitchen knife companies don't have billions of dollars, we don't go after them. We go after the LLM that might have given someone bad diet advice or made them feel sad. Nevermind the huge marketing budget spent on making people feel inadequate, ugly, old, etc. That does way more harm than tricking an LLM into telling you you can cook with glue.

Ad companies prompt injecting consumers. LLM companies countering with guardrails.

Re: Changes in the system prompt between Claude Opus 4.6 and 4.7

#69

The eating disorder section is kind of crazy. Are we going to incrementally add sections for every 'bad' human behaviour as time goes on?

When you are worth hundreds of billions, people start falling over themselves running to file lawsuits against you. We're already seeing this happen. So spending $50M to fund a team to weed out "food for crazies" becomes a no-brainer.

It is a no brainer. If a company of any size is putting out a product that caused cancer we wouldn't think twice about suing them. Why should mental health disorders be any different?

Re: Changes in the system prompt between Claude Opus 4.6 and 4.7

#70

The eating disorder section is kind of crazy. Are we going to incrementally add sections for every 'bad' human behaviour as time goes on?

Yup. Anyone who is surprised by this has not been paying attention to the centralization of power on the internet in the past 10 years.
Post reply on HN