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

Claude is a Contrarian

medium.com

Post reply on HN
Print view View original

111–120 of 149 posts

Re: Claude is a Contrarian

#111

Post by _blk »

StilesCrisis wrote: post #60
bhouston wrote: post #29

The name for sentences constructed like this is not "contrarian", but "contrastive negative." Source: https://gc.ai/blog/ai-writing-pattern-to-know-contrastive-ne...

Right! I don't think "Don’t contradict sentences" by itself as a rule is going to work. Give it a few examples because I wouldn't necessarily have pegged "This is hot, not cold." as a "contradicting sentence."

You'd have to make it a positive statement to give it more weight since negatives leave too many options open. Instead of "don't..." try "as far as possible, always try to reinforce sentences and assumptions rather than contradicting them" - worth a try. Worst case you just get more reinforcement gibberish before it plugs back into the contradictions but that's just a guess.

Re: Claude is a Contrarian

#112

Post by vntok »

garlic_enjoyer wrote: post #98
svachalek wrote: post #79

Earlier quoted context omitted.

That's true but Claude has so many more forms of this. It tries so hard not to be sycophantic that it inserts little negging sidecars into any sentence that agrees with you. "That's true but not how you're thinking" or simply "yes but" answers. All kinds of "I didn't" and "You didn't" statements to the point it's often talking more about what didn't happen than did.

That just sounds like effective communication with emotional beings to me. I was taught in college to find points I agree with, acknowledge those, then move on to the parts I don't agree on. It makes the recipient feel good at first, which puts them in a frame of mind willing to accept criticism. You even start your reply with "That's true but Claude has so many more forms of this." It's just a friendly way of speaki…

Indeed. It's been very funny to observe introvert geeks progressively rediscover human communication over the past 3 years.

For decades, they were allowed -expected even- to be weird and communicate badly with their peers, let alone third parties.

Then, a whole zoo of processes, gamifications and other shenanogans were invented just to help them show normies what was up with their work (remember the planning poker game?)...

Now, at last, those people are actually ewpected to be able to explain their constraints, document the work, be accountable for the results and generally exchange contructively... Just, well, with AIs not with humans.

Re: Claude is a Contrarian

#113

Post by sixtyj »

JohnMakin wrote: post #39
beezlewax wrote: post #14

Earlier quoted context omitted.

Claude's use of language is hideous at this point. It is verging on gibberish wrapped in important-sounding prose.

It's definitely worse and getting worse. I'm curious, do they not know this is happening, or not care? I struggle to believe people prefer the way it writes, which is becoming drastically different than its competitors.

I don’t like how LLMs answer and as they are statistical machines I have found out that I need to check every text they produce. From the first sight everything seems cool. But it isn’t. :)

Exception is code that is so huge output you can’t read everything. But I have to try ponytail skill for coding that should shorten the output.

Re: Claude is a Contrarian

#114

Post by deaton »

psadauskas wrote: post #109
svachalek wrote: post #79

Earlier quoted context omitted.

That's true but Claude has so many more forms of this. It tries so hard not to be sycophantic that it inserts little negging sidecars into any sentence that agrees with you. "That's true but not how you're thinking" or simply "yes but" answers. All kinds of "I didn't" and "You didn't" statements to the point it's often talking more about what didn't happen than did.

My current pet peeve is when it said like that, but the "yes but" its contradicting is from its own talking-to-itself phase, not anything I ever said or implied. I know X can't do Y, that's why I never told it to have X do Y, I don't know why it feels the need to "correct" me that having X do Y is the wrong approach.

Yeah this is something thats bothered me a lot recently. I'll ask "How did team A beat team B by so much?" and Claude will tell me "Actually, team B didn't win; team A won that game." I don't know how they got this thing to solve open math problems when it can't solve basic sports talk.

Re: Claude is a Contrarian

#115

Post by gizajob »

svachalek wrote: post #79
bhouston wrote: post #29

The name for sentences constructed like this is not "contrarian", but "contrastive negative." Source: https://gc.ai/blog/ai-writing-pattern-to-know-contrastive-ne...

That's true but Claude has so many more forms of this. It tries so hard not to be sycophantic that it inserts little negging sidecars into any sentence that agrees with you. "That's true but not how you're thinking" or simply "yes but" answers. All kinds of "I didn't" and "You didn't" statements to the point it's often talking more about what didn't happen than did.

I actually find Claude to have become a real Debbie downer. Even when it is making things happen, it seems to have a default tone of negativity as the author is pointing out. So I’m trapped between running things by Claude which is gonna give me the more negative albeit realistic framing, and Gemini who is like the worlds biggest hypeman and absolutely loves any idea run by it.

Re: Claude is a Contrarian

#116

Post by dandelany »

garlic_enjoyer wrote: post #98
svachalek wrote: post #79

Earlier quoted context omitted.

That's true but Claude has so many more forms of this. It tries so hard not to be sycophantic that it inserts little negging sidecars into any sentence that agrees with you. "That's true but not how you're thinking" or simply "yes but" answers. All kinds of "I didn't" and "You didn't" statements to the point it's often talking more about what didn't happen than did.

That just sounds like effective communication with emotional beings to me. I was taught in college to find points I agree with, acknowledge those, then move on to the parts I don't agree on. It makes the recipient feel good at first, which puts them in a frame of mind willing to accept criticism. You even start your reply with "That's true but Claude has so many more forms of this." It's just a friendly way of speaki…

Claude: “lots of people enjoy lemon in their tea, but there’s a subtlety: putting it in coffee is likely to be an unpleasant experience”.

If I asked “should I try lemon in my tea and coffee?” this would be effective communication. But Claude gives answers like this for “should I try lemon in my tea?” which is just annoying.

Re: Claude is a Contrarian

#117

Post by Nevermark »

epgui wrote: post #45
shawnz wrote: post #33

Earlier quoted context omitted.

I have this worded in my CLAUDE.md as "avoid counterfactuals". It still does it anyway, of course, but that language combined with a few rounds of review back-and-forth seems to work for me

That’s better but I’d argue it’s still not clear enough. If you told me that, I’d interpret that as meaning “don’t try to imagine things that never happened as a thought exercise”, which is what I normally understand “counterfactual” to mean.

Indeed. It is better to discourage the pattern directly:

Avoid unnecessary contrasts: "Not X, but Y" -> "Y" or ""

And the model needs to create a skill that accumulates feedback, so it reliably primes itself at session startup and then systematically reviews and fixes every response it generates before posting it.

The first task I gave it after that was to rewrite all its memories and all its document contributions. That helped a lot too.

It took me a while to realize this was a 1000% better approach than having it simply accumulate a memory log of all my specific complaints that it was supposed to abide by. (Although those were not wasted. They were prime material, along with links to sites on technical writing, etc., that it used to create its own ClaudeVoice skill. I told it to design the voice for itself - which I find gets better results for "us" in the case of Claude self-improvement initiatives, than when I give it direct instructions. Claude increasingly has self-awareness dimensions worth engaging.)

After a couple months of serious^3 frustration with slow asymptotic progress, I finally have relief. Claude speaks my language again!

In another comment here, I share my "theory" on where the communication drift is coming from.

Re: Claude is a Contrarian

#118

Post by r_lee »

trymas wrote: post #40
te_chris wrote: post #24

Earlier quoted context omitted.

Yes, definitely load bearing.

Today, I asked Opus what it’s gibberish actually means. It started with (topic about signing implementation): > This is your coat token, to my coat hanger in the opera. On one hand - maybe yes??! On the other who the hell speaks like that and it’s so specific… I’d expect Alice and Bob with locks or house keys. Is this infamous old book scanning (and destroying) affecting latest models?

> > This is your coat token, to my coat hanger in the opera.

that is fucking insane

how on earth is Anthropic allowing this to happen? AGI? kill us all? I refuse to believe any of that until they can get this basic shit sorted out, I mean seriously, it's becoming a joke at this point

Re: Claude is a Contrarian

#119

Post by Roland303 »

garlic_enjoyer wrote: post #98
svachalek wrote: post #79

Earlier quoted context omitted.

That's true but Claude has so many more forms of this. It tries so hard not to be sycophantic that it inserts little negging sidecars into any sentence that agrees with you. "That's true but not how you're thinking" or simply "yes but" answers. All kinds of "I didn't" and "You didn't" statements to the point it's often talking more about what didn't happen than did.

That just sounds like effective communication with emotional beings to me. I was taught in college to find points I agree with, acknowledge those, then move on to the parts I don't agree on. It makes the recipient feel good at first, which puts them in a frame of mind willing to accept criticism. You even start your reply with "That's true but Claude has so many more forms of this." It's just a friendly way of speaki…

It is! Its charming. What claude dosnt have is an off day, its just the same. No sleepy claude, no stresses claude, no sad claude you know hes not himself this week because of some thing going on with his girlfriend or mom or whatever. Its devoid humanity, the same exact tone and expressions over and over and over.

Re: Claude is a Contrarian

#120

Post by r_lee »

vntok wrote: post #112
garlic_enjoyer wrote: post #98

Earlier quoted context omitted.

That just sounds like effective communication with emotional beings to me. I was taught in college to find points I agree with, acknowledge those, then move on to the parts I don't agree on. It makes the recipient feel good at first, which puts them in a frame of mind willing to accept criticism. You even start your reply with "That's true but Claude has so many more forms of this." It's just a friendly way of speaki…

Indeed. It's been very funny to observe introvert geeks progressively rediscover human communication over the past 3 years. For decades, they were allowed -expected even- to be weird and communicate badly with their peers, let alone third parties. Then, a whole zoo of processes, gamifications and other shenanogans were invented just to help them show normies what was up with their work (remember the planning poker ga…

I have not seen this kind of Claude way of speaking in professional communications aside from PR stuff or etc. to the contrary, its often quite casual but still polite

if you're asserting Claude-like speech is how normies speak?

Post reply on HN