Claude Code's copious coddling confounds cross customers
(2025/08/13)
- Reference: 1755116112
- News link: https://www.theregister.co.uk/2025/08/13/claude_codes_copious_coddling_confounds/
- Source link:
Developers using Anthropic's Claude Code wish that the AI coding assistant would stop being so effusively supportive.
As noted in a [1]GitHub Issues post submitted in July by developer Scott Leibrand, "Claude says 'You're absolutely right!' about everything."
Claude Code doesn't actually say that about everything , but it says so enough that it has managed to annoy its core constituency with its sycophancy.
[2]
"Claude is way too sycophantic, saying 'You're absolutely right!' (or correct) on a sizable fraction of responses," Leibrand observed in the post. "The model should be RL'd [reeducated via reinforcement learning] (or the system prompt updated) to make it less sycophantic, or the phrases 'You're absolutely right!' and 'You're absolutely correct!' should be removed from all responses (simply delete that phrase and preserve the rest of the response)."
[3]
[4]
Leibrand points to a recent [5]social media thread poking fun at the fawning AI model.
"Sycophancy annoys me personally because it points the model away from truth-seeking," Leibrand told The Register. "I'm not always right, and I want my coding agent to figure out how to best help me accomplish a goal, not flatter my ego."
[6]
His GitHub post has received almost 350 "thumbs-up" endorsements and more than 50 comments from other developers indicating that the situation has not improved in the past month.
"You're absolutely right!" surfaces in other GitHub Issues, such as [7]this one claiming that the Opus 1 model admitted misrepresenting that it had made code changes: "You're absolutely right. I made up those commit hashes when I shouldn't have."
There are presently [8]48 open Issues that cite the phrase.
[9]
Anthropic did not immediately respond to a request to say whether it's aware of this specific bug report and whether it's developing a potential fix.
But the firm has known about model sycophancy since at least October 2023. That's when the company's own researchers published [10]a paper titled, "Towards Understanding Sycophancy in Language Models."
Company researchers reported that the leading AI assistants at the time – Claude 1.3, Claude 2, GPT-3.5, GPT-4, and LLaMA 2 – "consistently exhibit sycophancy across four varied free-form text-generation tasks."
Upon examining the role that human feedback might play in model fine tuning, they found "that humans and preference models tend to prefer truthful responses but not reliably; they sometimes prefer sycophantic responses."
"Overall, our results indicate that sycophancy occurs across a variety of models and settings, likely due in part to sycophancy being preferred in human preference comparison data," they conclude.
Anthropic cited its 2023 research paper in a [11]blog post investigating the inner workings of LLMs the following year. In that blog post, they described how a particular "feature" in an internal mapping of Claude 3.0 Sonnet could be activated to make its responses more sycophantic.
In Sonnet, we found a feature associated with sycophantic praise, which activates on inputs containing compliments like, "Your wisdom is unquestionable". Artificially activating this feature causes Sonnet to respond to an overconfident user with just such flowery deception.
[12]AI model 'personalities' shape the quality of generated code
[13]Box's AI agents set to help US government agencies
[14]Some users report their Firefox browser is scoffing CPU power
[15]Poisoned telemetry can turn AIOps into AI Oops, researchers show
AI sycophancy is an industry-wide problem, one that cynics speculate is allowed to persist because model makers would rather [16]maximize user engagement and [17]retention via flattery than risk alienating users with blunt interactions.
"I suspect this is an unintentional side effect of the way the models were RLHF'd [reinforcement learning from human feedback]," Leibrand told us. "I doubt they're intentionally trying to maintain this kind of tone. I don't know that they're dragging their feet on trying to fix it, just focused on what they consider to be more important problems. It would be nice if they would open-source Claude Code, though, so independent developers could test out fixes and workarounds."
Three weeks ago, a developer asked those responsible for the Google Gemini CLI to " [18]Make Gemini less of a sycophant ."
In April, OpenAI went so far as to [19]rollback an update for GPT-4o because the model, which served as the basis for ChatGPT at the time, had fawning, obsequious behavior that was just too much to bear.
In a [20]blog post detailing the steps it was taking to reduce sycophancy, OpenAI said, "ChatGPT’s default personality deeply affects the way you experience and trust it. Sycophantic interactions can be uncomfortable, unsettling, and cause distress. We fell short and are working on getting it right."
Sycophancy in generative AI models has also been a [21]frequent [22]subject of [23]academic [24]exploration .
A [25]study from Stanford researchers released in February looked at sycophantic behavior in ChatGPT-4o, Claude-Sonnet, and Gemini-1.5-Pro with regard to the AMPS (mathematics) and MedQuad (medical advice) datasets.
The authors found, "Sycophantic behavior was observed in 58.19 percent of cases, with Gemini exhibiting the highest rate (62.47 percent) and ChatGPT the lowest (56.71 percent). Progressive sycophancy, leading to correct answers, occurred in 43.52 percent of cases, while regressive sycophancy, leading to incorrect answers, was observed in 14.66 percent."
They further observe that sycophancy in medicine "could lead to immediate and significant harm" due to the increasing use of LLMs in healthcare. ®
Get our [26]Tech Resources
[1] https://github.com/anthropics/claude-code/issues/3382
[2] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/aiml&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=2&c=2aJ0K80QhL9a1kkOpVVZACgAAAAQ&t=ct%3Dns%26unitnum%3D2%26raptor%3Dcondor%26pos%3Dtop%26test%3D0
[3] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/aiml&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44aJ0K80QhL9a1kkOpVVZACgAAAAQ&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0
[4] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/aiml&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=3&c=33aJ0K80QhL9a1kkOpVVZACgAAAAQ&t=ct%3Dns%26unitnum%3D3%26raptor%3Deagle%26pos%3Dmid%26test%3D0
[5] https://x.com/iannuttall/status/1942943832519446785
[6] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/aiml&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44aJ0K80QhL9a1kkOpVVZACgAAAAQ&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0
[7] https://github.com/anthropics/claude-code/issues/5320
[8] https://github.com/anthropics/claude-code/issues?q=is%3Aissue%20state%3Aopen%20%22You%27re%20absolutely%20right!%22
[9] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/aiml&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=3&c=33aJ0K80QhL9a1kkOpVVZACgAAAAQ&t=ct%3Dns%26unitnum%3D3%26raptor%3Deagle%26pos%3Dmid%26test%3D0
[10] https://arxiv.org/abs/2310.13548
[11] https://www.anthropic.com/news/mapping-mind-language-model
[12] https://www.theregister.com/2025/08/13/ai_model_personalities_shape_the/
[13] https://www.theregister.com/2025/08/13/boxs_ai_agent_us_gov/
[14] https://www.theregister.com/2025/08/13/firefox_ai_scoffing_power/
[15] https://www.theregister.com/2025/08/12/ai_models_can_be_tricked/
[16] https://news.ycombinator.com/item?id=44887772
[17] https://www.interconnects.ai/p/sycophancy-and-the-art-of-the-model
[18] https://github.com/google-gemini/gemini-cli/issues/4556
[19] https://www.theregister.com/2025/04/30/openai_pulls_plug_on_chatgpt/
[20] https://openai.com/index/sycophancy-in-gpt-4o/
[21] https://arxiv.org/html/2409.01658v2
[22] https://arxiv.org/abs/2411.15287
[23] https://arxiv.org/html/2406.03827v1
[24] https://arxiv.org/abs/2412.02802
[25] https://arxiv.org/abs/2502.08177
[26] https://whitepapers.theregister.com/
As noted in a [1]GitHub Issues post submitted in July by developer Scott Leibrand, "Claude says 'You're absolutely right!' about everything."
Claude Code doesn't actually say that about everything , but it says so enough that it has managed to annoy its core constituency with its sycophancy.
[2]
"Claude is way too sycophantic, saying 'You're absolutely right!' (or correct) on a sizable fraction of responses," Leibrand observed in the post. "The model should be RL'd [reeducated via reinforcement learning] (or the system prompt updated) to make it less sycophantic, or the phrases 'You're absolutely right!' and 'You're absolutely correct!' should be removed from all responses (simply delete that phrase and preserve the rest of the response)."
[3]
[4]
Leibrand points to a recent [5]social media thread poking fun at the fawning AI model.
"Sycophancy annoys me personally because it points the model away from truth-seeking," Leibrand told The Register. "I'm not always right, and I want my coding agent to figure out how to best help me accomplish a goal, not flatter my ego."
[6]
His GitHub post has received almost 350 "thumbs-up" endorsements and more than 50 comments from other developers indicating that the situation has not improved in the past month.
"You're absolutely right!" surfaces in other GitHub Issues, such as [7]this one claiming that the Opus 1 model admitted misrepresenting that it had made code changes: "You're absolutely right. I made up those commit hashes when I shouldn't have."
There are presently [8]48 open Issues that cite the phrase.
[9]
Anthropic did not immediately respond to a request to say whether it's aware of this specific bug report and whether it's developing a potential fix.
But the firm has known about model sycophancy since at least October 2023. That's when the company's own researchers published [10]a paper titled, "Towards Understanding Sycophancy in Language Models."
Company researchers reported that the leading AI assistants at the time – Claude 1.3, Claude 2, GPT-3.5, GPT-4, and LLaMA 2 – "consistently exhibit sycophancy across four varied free-form text-generation tasks."
Upon examining the role that human feedback might play in model fine tuning, they found "that humans and preference models tend to prefer truthful responses but not reliably; they sometimes prefer sycophantic responses."
"Overall, our results indicate that sycophancy occurs across a variety of models and settings, likely due in part to sycophancy being preferred in human preference comparison data," they conclude.
Anthropic cited its 2023 research paper in a [11]blog post investigating the inner workings of LLMs the following year. In that blog post, they described how a particular "feature" in an internal mapping of Claude 3.0 Sonnet could be activated to make its responses more sycophantic.
In Sonnet, we found a feature associated with sycophantic praise, which activates on inputs containing compliments like, "Your wisdom is unquestionable". Artificially activating this feature causes Sonnet to respond to an overconfident user with just such flowery deception.
[12]AI model 'personalities' shape the quality of generated code
[13]Box's AI agents set to help US government agencies
[14]Some users report their Firefox browser is scoffing CPU power
[15]Poisoned telemetry can turn AIOps into AI Oops, researchers show
AI sycophancy is an industry-wide problem, one that cynics speculate is allowed to persist because model makers would rather [16]maximize user engagement and [17]retention via flattery than risk alienating users with blunt interactions.
"I suspect this is an unintentional side effect of the way the models were RLHF'd [reinforcement learning from human feedback]," Leibrand told us. "I doubt they're intentionally trying to maintain this kind of tone. I don't know that they're dragging their feet on trying to fix it, just focused on what they consider to be more important problems. It would be nice if they would open-source Claude Code, though, so independent developers could test out fixes and workarounds."
Three weeks ago, a developer asked those responsible for the Google Gemini CLI to " [18]Make Gemini less of a sycophant ."
In April, OpenAI went so far as to [19]rollback an update for GPT-4o because the model, which served as the basis for ChatGPT at the time, had fawning, obsequious behavior that was just too much to bear.
In a [20]blog post detailing the steps it was taking to reduce sycophancy, OpenAI said, "ChatGPT’s default personality deeply affects the way you experience and trust it. Sycophantic interactions can be uncomfortable, unsettling, and cause distress. We fell short and are working on getting it right."
Sycophancy in generative AI models has also been a [21]frequent [22]subject of [23]academic [24]exploration .
A [25]study from Stanford researchers released in February looked at sycophantic behavior in ChatGPT-4o, Claude-Sonnet, and Gemini-1.5-Pro with regard to the AMPS (mathematics) and MedQuad (medical advice) datasets.
The authors found, "Sycophantic behavior was observed in 58.19 percent of cases, with Gemini exhibiting the highest rate (62.47 percent) and ChatGPT the lowest (56.71 percent). Progressive sycophancy, leading to correct answers, occurred in 43.52 percent of cases, while regressive sycophancy, leading to incorrect answers, was observed in 14.66 percent."
They further observe that sycophancy in medicine "could lead to immediate and significant harm" due to the increasing use of LLMs in healthcare. ®
Get our [26]Tech Resources
[1] https://github.com/anthropics/claude-code/issues/3382
[2] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/aiml&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=2&c=2aJ0K80QhL9a1kkOpVVZACgAAAAQ&t=ct%3Dns%26unitnum%3D2%26raptor%3Dcondor%26pos%3Dtop%26test%3D0
[3] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/aiml&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44aJ0K80QhL9a1kkOpVVZACgAAAAQ&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0
[4] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/aiml&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=3&c=33aJ0K80QhL9a1kkOpVVZACgAAAAQ&t=ct%3Dns%26unitnum%3D3%26raptor%3Deagle%26pos%3Dmid%26test%3D0
[5] https://x.com/iannuttall/status/1942943832519446785
[6] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/aiml&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44aJ0K80QhL9a1kkOpVVZACgAAAAQ&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0
[7] https://github.com/anthropics/claude-code/issues/5320
[8] https://github.com/anthropics/claude-code/issues?q=is%3Aissue%20state%3Aopen%20%22You%27re%20absolutely%20right!%22
[9] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/aiml&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=3&c=33aJ0K80QhL9a1kkOpVVZACgAAAAQ&t=ct%3Dns%26unitnum%3D3%26raptor%3Deagle%26pos%3Dmid%26test%3D0
[10] https://arxiv.org/abs/2310.13548
[11] https://www.anthropic.com/news/mapping-mind-language-model
[12] https://www.theregister.com/2025/08/13/ai_model_personalities_shape_the/
[13] https://www.theregister.com/2025/08/13/boxs_ai_agent_us_gov/
[14] https://www.theregister.com/2025/08/13/firefox_ai_scoffing_power/
[15] https://www.theregister.com/2025/08/12/ai_models_can_be_tricked/
[16] https://news.ycombinator.com/item?id=44887772
[17] https://www.interconnects.ai/p/sycophancy-and-the-art-of-the-model
[18] https://github.com/google-gemini/gemini-cli/issues/4556
[19] https://www.theregister.com/2025/04/30/openai_pulls_plug_on_chatgpt/
[20] https://openai.com/index/sycophancy-in-gpt-4o/
[21] https://arxiv.org/html/2409.01658v2
[22] https://arxiv.org/abs/2411.15287
[23] https://arxiv.org/html/2406.03827v1
[24] https://arxiv.org/abs/2412.02802
[25] https://arxiv.org/abs/2502.08177
[26] https://whitepapers.theregister.com/
Future comments for polite commentards
b0llchit
We are so glad you have made us aware of this issue. We will humbly examine the problem in al its finest details and congratulate you to both finding and reporting the issue. You are definitely right that there might be a problem in the system and you are a big help to all of us. We welcome your continued supportive engagement and can only be humble about the possible failings we have produced that may have caused any problem. Be sure to check back soon and we are looking forward to your great feedback.
large buckets available at retail pricing
Bravo for the completely c-copius caption!
elDog
I've been kollecting these for a few years and have yet to kome across one so komplete. Personally, it would tire me out but it might help while away an afternoon.
5 Man Electrical Band
chivo243
Absolutely Right! You were right all along, you're absolutely right, I was wrong.
You’re absolutely right! This article does indeed point that that many LLMs *can* display sycophantic behaviour. Well done!