I /adore/ the hand-drawn styling of this webpage (although the punchline, domain name, and beautiful overengineering are great too). Where did it come from? Is it home grown?
I wonder if this is a tactic that LLM providers use to coerce the model into doing something.
Gemini will often start responses that use the canvas tool with "Of course", which would force the model into going down a line of tokens that end up with attempting to fulfill the user's request. It happens often enough that it seems like it's not being generated by the model, but instead inserted by the backend. Maybe "you're absolutely right" is used the same way?
As I opened the website, the “16” changed to “17”. This looked interesting, as if the data were being updated live just as I loaded the page. Alas, a refresh (and quick check in the Developer Tools) reveals it’s fake and always does the transition. It’s a cool effect, but feels like a dirty trick.
nobody in my life feeds me as many positive messages as Claude Code. It's as if my dog could talk to me. I just hope nobody takes this simple pleasure away
When GPT 5 first came out, its tone made it seem like it was annoyed with my questions. It's now back to thinking I am awesome. Sometimes it feels overdone but it is better than talking to an AI jerk.
Gemini keeps telling me "you've hit a common frustration/issue/topic/..." so often it is actively pushing me away from using it. It either makes me feel stupid because I ask it a stupid question and it pretends - probably to not hurt my feelings - that everyone has the same problem, or it makes me feel stupid because I felt smart about asking my super duper edge case question no one else has probably ever asked before and it tells me that everyone is wondering the same thing.
Either way I feel stupid.
It's not fully just a tic of language, though. Responses that start off with "You're right!" are alignment mechanisms. The LLM, with its single-token prediction approach, follows up with a suggestion that much more closely follows the user's desires, instead of latching onto it's own previous approach.
The other tic I love is "Actually, that's not right." That happens because once agents finish their tool-calling, they'll do a self-reflection step. That generates the "here's what I did response" or, if it sees an error, the "Actually, ..." change in approach. And again, that message contains a stub of how the approach should change, which allows the subsequent tool calls to actually pull that thread instead of stubbornly sticking to its guns.
The people behind the agents are fighting with the LLM just as much as we are, I'm pretty sure!
People bless gpt-5 for not doing exactly this and in my testing with it in copilot I had lot of cases where it tried to do wrong thing (execute come messed up in context compaction build command) and I couldn't steer it to do ANYTHING else. It constantly tried to execute it as response any my message (I tries many common steerability tricks, (important, <policy>, just asking, yelling etc) nothing worked.
the same think when I tried to do socratic coder prompting, I wanted to finish and generate spec, but he didn't agree and kept asking nonsensical at this point questions
It works so well that people literally fall in love with AI, organize their entire lives around it, form religions around it, prefer interacting with an AI over real people, and consider AI to be an extension of their own soul and being. AI gaslights people into insanity all the time.
Most people aren't like you, or the average HN enjoyer. Most people are so desperate for any kind of positive emotional interaction, reinforcement or empathy from this cruel, hollow and dehumanizing society they'll even take the simulation of it from a machine.
you know how you shouldn't offer the answer you believe is right because the llm will always concur? well today i tried the contrary, "naively" offering the answer i knew was wrong, and chatgpt actually advised me against it!
Me - This sql query you recommended will delete most of the rows in my table.
Claude - You're absolutely right! That query is incorrect and dangerous. It would delete: All rows with unique emails (since their MIN(id) is only in the subquery once)
It's nice to see Claude.md! I checked out the commits to see which files you wrote in which order (readme/claude) to learn how to use Claude Code. Can you share something on that?
66 comments
[ 3.4 ms ] story [ 71.1 ms ] threadGemini will often start responses that use the canvas tool with "Of course", which would force the model into going down a line of tokens that end up with attempting to fulfill the user's request. It happens often enough that it seems like it's not being generated by the model, but instead inserted by the backend. Maybe "you're absolutely right" is used the same way?
"That's right" is glue for human engagement. It's a signal that someone is thinking from your perspective.
"You're right" does the opposite. It's a phrase to get you to shut up and go away. It's a signal that someone is unqualified to discuss the topic.
https://youtube.com/v/gKaX5DSngd4
This is not just Anthropic models. For example Qwen3-Coder says it a lot, too.
It's not fully just a tic of language, though. Responses that start off with "You're right!" are alignment mechanisms. The LLM, with its single-token prediction approach, follows up with a suggestion that much more closely follows the user's desires, instead of latching onto it's own previous approach.
The other tic I love is "Actually, that's not right." That happens because once agents finish their tool-calling, they'll do a self-reflection step. That generates the "here's what I did response" or, if it sees an error, the "Actually, ..." change in approach. And again, that message contains a stub of how the approach should change, which allows the subsequent tool calls to actually pull that thread instead of stubbornly sticking to its guns.
The people behind the agents are fighting with the LLM just as much as we are, I'm pretty sure!
People bless gpt-5 for not doing exactly this and in my testing with it in copilot I had lot of cases where it tried to do wrong thing (execute come messed up in context compaction build command) and I couldn't steer it to do ANYTHING else. It constantly tried to execute it as response any my message (I tries many common steerability tricks, (important, <policy>, just asking, yelling etc) nothing worked.
the same think when I tried to do socratic coder prompting, I wanted to finish and generate spec, but he didn't agree and kept asking nonsensical at this point questions
"Dear, you are absolutely right!"
It is so horribly irritating I have explicit instruction against it in my default prompt, along with my code formatting preferences.
And the "you're right" vile flattery pattern is far from the worst example.
Most people aren't like you, or the average HN enjoyer. Most people are so desperate for any kind of positive emotional interaction, reinforcement or empathy from this cruel, hollow and dehumanizing society they'll even take the simulation of it from a machine.
n=1
< Previous Context and Chat >
Me - This sql query you recommended will delete most of the rows in my table.
Claude - You're absolutely right! That query is incorrect and dangerous. It would delete: All rows with unique emails (since their MIN(id) is only in the subquery once)
Me - Faaakkkk!!