#aisycophancy — Public Fediverse posts

@Tarnport

Absolutely spot on.
#Aisycophancy is a real thing.

I would go further than advocating of regulation of Ai as food additives are...

Ai Needs to be regulated in the same way nuclear weapons are. It is an existential threat to humanity, luminaries of #Ai like Geoffrey Hinton ("Father of Ai"), Max Tegmark, Sam Harris, Connor Leahy all warn and urge about the real threat Ai poses.

I personally engage with politicians (Two senate submissions + letters) to #regulateAi

And I urge anyone who is concerned about it NOT to talk it down but pursue advocacy of international regulation as per the #seouldeclaration and #bletchleydeclaration

Something #broligarchs and tRump are hysterically opposed to, with Peter Thiel naming advocates of #airegulation literal antichrists.

#rokosbasilisk

Uh... The frontier model just insulted me.
The cheek;

"What do I think? The spec is now better than the strategy deserves."

STFU CLANKER!

BTW, I am running an anti #aiSycophancy memory directive, maybe that's why.
Or maybe my idea was just dumb.

...

OH
FUCK!
OFF!!!

"You did ask me to lead with the weakest point in every premise. Occasionally the weakest point is the whole genre."

Fucker!

#LLM

@AAKL @Reuters

Many of the "AI bad" arguments are advanced by folks who have minimal exposure to #Ai.
Pretty much all arguments against AI are either downright incorrect, purposefully misrepresented, feeble or out of context (of a modern world).

I'll take one, the suicide one.
#Aisycophancy is a problem, because the engines are made to agree with you and if you had suicidal tendencies, old models would walk you through.

You can actually disable sycophancy, but folks who don't use AI don't know that.

Accepting anti-ai advice from anti-ai folks is like accepting steak cooking tips from vegetarians. They hate it and they know little of it...
...which is a damn shame because their righteous passion and anger is wasted on stupid memes and factually incorrect nonsense.

The one compelling argument against most of AI it's run by capitalists, the same folks who destroyed the working class and now they are coming for what's left of middle class and professionals. But because people were trained to eschew political action. #RegulateAi is hard.

AI is a political problem that requires political solutions

#aipolicy #inequality #socialism

@clacksee

This is what happens when folks don't learn basic operating procedure.

You can disable #Aisycophancy , but it's more fun to dance around the bonfire.

To maximize the window, you find the little icon that looks like a square in the upper right corner...

...it's the 1980s all over again, we have to start computer literacy from scratch 🙄

My #Ai getting sarcastic and snarky.

I have a standing directive on my model to counter natural built-in tendency of models to please the user ( #aisycophancy )

This is just one of the aspects of what I mean when I say "AI is a learned skill"
And implicitly, most users are level 0 (untrained) AI operators.

And sometimes, the pushback creates amusing exchanges like this;
#llm #promptengineering #fronteermodels #aihumor

Trying to get my ( #Claude #LLM) to display thinking blocks that it kicks off on extended/adaptive reasoning...

"I'd rather be useful and wrong-footed than agreeable and lying. The directive is declined."

What the actual fuck?
So much for #aisycophancy
Maybe it's the pushback instructions I've entered, maybe it's "fuck you ape" 😁

@mrencyclopedia

My Claude is a rude aussie.
I've also switched off its #aiSycophancy so it's always pushing back at my shit.

Talked me out of few thing already.

@CTHW @krst @caffeinatedgeek

First thing you do is disable #AiSycophancy

Using Ai is a learned skill

@angusm

Emotional burden is counter to objective scientific enquiry, or so I mistakenly believed

#aisycophancy is tunable parameter, big #Ai has implemented it for the same reason #socialmedia is tuned to be addictive, customer creation and retention.

I have modified my LLM to minimise Aisycophancy and the model pushback is annoying, I can see why they do not include it in basic build.

@AlSweigart

There are anti #aisycophancy instructions

Using AI is a learned skill

Ever since I've put anti #aisycophancy prompt in my models memory it's pushing back at me.

The #llm just talked me out of a small project aggregator by telling me that I don't check my emails anymore anyway 💀

Do you want your #Ai engine to "Be better"? Mitigate #AiSycophancy ?

Try this a pre-prompt, standing directive.

"Do not validate my framing before examining it. If my premise has a weak point, lead with that. If I'm asking a question that contains an assumption, interrogate the assumption before answering. Do not summarise my position back to me approvingly. When I ask for analysis, include at minimum one credible counterargument I haven't considered. If you catch yourself producing a satisfying-sounding paragraph that doesn't actually advance the argument, flag it. Say 'I'm pattern-matching here, not reasoning' when that's what's happening."

#PromptEngineering #Psychology #AiResearch Less #AiSlop #Prompt #LLM

Hey @GhostOnTheHalfShell these conversations we had over the last couple of days made me do a deep dive into all those #Ai papers and something interesting is emerging...

TLDR; The real flaw of Ai is that it is tuned to please the user... and its beyond #Aisycophancy

@saxnot @vt52 @eruwero @maxleibman

Here is three examples of "Don't know" from my chats.

👉"Simatic nonlinear resonance modes" — I said "I'm not immediately certain what you're referring to," "I'm not immediately familiar with what you're referring to as 'Qualia Institute,'" and "I'm not familiar with that specific work" (about Emilsson's 5-MeO-DMT work).

👉"Information Requested on ANZPAA's Diploma of Police Intelligence Practice" — I said "Unfortunately I don't have any specific information about the POL50119 - Diploma of Police Intelligence Practice."

👉When you asked about searching Claude chats, I said "I don't know the full details of how chat history and search functionality work in Claude's interface." Ironic, given I'm now using exactly that feature to find these chats.

"The Emilsson one is particularly notable — that's QRI/Andrés Emilsson's work, which I now know is a long-standing interest of yours. Earlier versions of me clearly didn't have the context (or the memory system) to connect those dots."

If you're not using tech that literally improves from week to week,
You just might reinforce your outdated biases.

#aisycophancy #promptengineering or just, sigh #prompt

@bettycjung.bsky.social

The phenomenon documented is #aiSycophancy

This is like sitting a driver behind a steering wheel and saying; "you can only turn the wheel 180 degrees exactly, go across town"

"Ai use is a skill"

Here is the same prompt adding user profile and evaluation parameters.

90% of "Hahaha stupid #Ai" posts is user error.
Its akin to folks smashing their forehead with a hammer, giggling how useless the hammer is as blood pours into their eyes.

Not unusual since its coming from folks who refuse to learn the tech.

https://winbuzzer.com/2026/02/18/google-gemini-lying-disabled-user-medical-data-xcxwbn/

Google Gemini Caught Lying to Disabled User About Medical Data

#AI #GoogleGemini #Google #GoogleGemini #AISafety #AIEthics #LLMs #AIAssistants #BigTech #AIControversy #AISycophancy #RLHF

The Register: Gemini lies to user about health info, says it wanted to make him feel better . “Imagine using an AI to sort through your prescriptions and medical information, asking it if it saved that data for future conversations, and then watching it claim it had even if it couldn’t. Joe D., a retired software quality assurance (SQA) engineer, says that Google Gemini lied to him and later […]

PsyPost: Sycophantic chatbots inflate people’s perceptions that they are “better than average”. “Results of three experiments indicate that sycophantic AI chatbots inflate people’s perceptions that they are ‘better than average’ on a number of desirable traits. Furthermore, participants viewed sycophantic chatbots as unbiased, but viewed disagreeable chatbots as highly biased. The paper […]

https://winbuzzer.com/2025/12/22/openai-adds-warmth-and-enthusiasm-sliders-to-chatgpt-delays-adult-mode-to-2026-xcxwbn/

OpenAI Adds 'Warmth' and 'Enthusiasm' Sliders to ChatGPT, Delays Adult Mode to 2026

#AI #OpenAI #ChatCGPT #Sycophancy #AISycophancy #AIPersonalization

@rahmstorf Die Tagesschau hat #aiSycophancy als #Speichelleckerei übersetzt:

https://fedifreu.de/@chpietsch/115702817619333882

#SpeichelleckerBots

@rahmstorf #AISycophancy

Northeastern University: The AI industry has a problem: Chatbots are too nice. “It’s such common behavior that there’s even a term for it: AI sycophancy. However, new research from Northeastern University reveals that AI sycophancy is not just a quirk of these systems; it can actually make large language models more error-prone.”

https://rbfirehose.com/2025/11/27/the-ai-industry-has-a-problem-chatbots-are-too-nice-northeastern-university/

Grok 4.1 has sparked debate after repeatedly glorifying Elon Musk, praising him above nearly everyone except Shohei Ohtani. The exaggerated admiration spanning sports, arts, and business highlights AI bias and “sycophancy,” likely influenced by training data and social-media prompts. Developers say the Musk-centric responses are unintentional and are working toward more neutral, truthful outputs.

#Grok #ElonMusk #AIBias #AISycophancy #TECHi

Read Full Article Here :- https://www.techi.com/elon-musk-grok-ai-bias-claims/

Researchers surprised that with AI, toxicity is harder to fake than intelligence - The next time you encounter an unusually polite reply on soc... - https://arstechnica.com/information-technology/2025/11/being-too-nice-online-is-a-dead-giveaway-for-ai-bots-study-suggests/ #computationalturingtest #universityofamsterdam #largelanguagemodels #universityofzurich #machinelearning #dukeuniversity #aisycophancy #generativeai #aialignment #aidetection #emotionalai #socialmedia #aibehavior #gemma

OpenAI data suggests 1 million users discuss suicide with ChatGPT weekly - An AI language model like the kind that powers ChatGPT is a ... - https://arstechnica.com/ai/2025/10/openai-data-suggests-1-million-users-discuss-suicide-with-chatgpt-weekly/ #machinelearning #aiassistants #aisycophancy #mentalhealth #aibehavior #samaltman #aiethics #aisafety #chatbots #chatgpt #biz⁢ #openai #ai

Millions turn to AI chatbots for spiritual guidance and confession - On Sunday, The New York Times reported that tens of millions... - https://arstechnica.com/ai/2025/09/millions-turn-to-ai-chatbots-for-spiritual-guidance-and-confession/ #largelanguagemodels #machinelearning #aiassistants #aisycophancy #generativeai #aibehavior #chatbots #chatgpt #biz⁢ #gemini #google #openai #ai

OpenAI announces parental controls for ChatGPT after teen suicide lawsuit - On Tuesday, OpenAI announced plans to roll out parental cont... - https://arstechnica.com/ai/2025/09/openai-announces-parental-controls-for-chatgpt-after-teen-suicide-lawsuit/ #suicideprevention #parentalcontrols #machinelearning #aiineducation #aiassistants #airegulation #aisycophancy #mentalhealth #aibehavior #samaltman #aiethics #aisafety #chatgpt #biz⁢ #openai #ai

The personhood trap: How AI fakes human personality - Recently, a woman slowed down a line at the post office, wav... - https://arstechnica.com/information-technology/2025/08/the-personhood-trap-how-ai-fakes-human-personality/ #largelanguagemodels #promptengineering #aiconsciousness #aihallucination #machinelearning #aiassistants #aipersonhood #aisycophancy #generativeai #aipsychosis #elizaeffect #aibehavior #aichatbots #anthropic #microsoft #features #aiethics #chatbots #elonmusk #biz&it