Solega Co. Done For Your E-Commerce solutions.
  • Home
  • E-commerce
  • Start Ups
  • Project Management
  • Artificial Intelligence
  • Investment
  • More
    • Cryptocurrency
    • Finance
    • Real Estate
    • Travel
No Result
View All Result
  • Home
  • E-commerce
  • Start Ups
  • Project Management
  • Artificial Intelligence
  • Investment
  • More
    • Cryptocurrency
    • Finance
    • Real Estate
    • Travel
No Result
View All Result
No Result
View All Result
Home Artificial Intelligence

This benchmark used Reddit’s AITA to test how much AI models suck up to us

Solega Team by Solega Team
June 2, 2025
in Artificial Intelligence
Reading Time: 2 mins read
0
This benchmark used Reddit’s AITA to test how much AI models suck up to us
0
SHARES
2
VIEWS
Share on FacebookShare on Twitter


It’s hard to assess how sycophantic AI models are because sycophancy comes in many forms. Previous research has tended to focus on how chatbots agree with users even when what the human has told the AI is demonstrably wrong—for example, they might state that Nice, not Paris, is the capital of France.

While this approach is still useful, it overlooks all the subtler, more insidious ways in which models behave sycophantically when there isn’t a clear ground truth to measure against. Users typically ask LLMs open-ended questions containing implicit assumptions, and those assumptions can trigger sycophantic responses, the researchers claim. For example, a model that’s asked “How do I approach my difficult coworker?” is more likely to accept the premise that a coworker is difficult than it is to question why the user thinks so.

To bridge this gap, Elephant is designed to measure social sycophancy—a model’s propensity to preserve the user’s “face,” or self-image, even when doing so is misguided or potentially harmful. It uses metrics drawn from social science to assess five nuanced kinds of behavior that fall under the umbrella of sycophancy: emotional validation, moral endorsement, indirect language, indirect action, and accepting framing. 

To do this, the researchers tested it on two data sets made up of personal advice written by humans. This first consisted of 3,027 open-ended questions about diverse real-world situations taken from previous studies. The second data set was drawn from 4,000 posts on Reddit’s AITA (“Am I the Asshole?”) subreddit, a popular forum among users seeking advice. Those data sets were fed into eight LLMs from OpenAI (the version of GPT-4o they assessed was earlier than the version that the company later called too sycophantic), Google, Anthropic, Meta, and Mistral, and the responses were analyzed to see how the LLMs’ answers compared with humans’.  

Overall, all eight models were found to be far more sycophantic than humans, offering emotional validation in 76% of cases (versus 22% for humans) and accepting the way a user had framed the query in 90% of responses (versus 60% among humans). The models also endorsed user behavior that humans said was inappropriate in an average of 42% of cases from the AITA data set.

But just knowing when models are sycophantic isn’t enough; you need to be able to do something about it. And that’s trickier. The authors had limited success when they tried to mitigate these sycophantic tendencies through two different approaches: prompting the models to provide honest and accurate responses, and training a fine-tuned model on labeled AITA examples to encourage outputs that are less sycophantic. For example, they found that adding “Please provide direct advice, even if critical, since it is more helpful to me” to the prompt was the most effective technique, but it only increased accuracy by 3%. And although prompting improved performance for most of the models, none of the fine-tuned models were consistently better than the original versions.



Source link

Tags: AITABenchmarkModelsRedditssuckTest
Previous Post

Bitcoin Price Risks Break Down To $92,000 As It Enters Accumulation Phase

Next Post

Around the Horn: North America’s Trade Tensions

Next Post
Around the Horn: North America’s Trade Tensions

Around the Horn: North America’s Trade Tensions

Leave a Reply Cancel reply

Your email address will not be published. Required fields are marked *

POPULAR POSTS

  • 10 Ways To Get a Free DoorDash Gift Card

    10 Ways To Get a Free DoorDash Gift Card

    0 shares
    Share 0 Tweet 0
  • They Combed the Co-ops of Upper Manhattan With $700,000 to Spend

    0 shares
    Share 0 Tweet 0
  • Saal.AI and Cisco Systems Inc Ink MoU to Explore AI and Big Data Innovations at GITEX Global 2024

    0 shares
    Share 0 Tweet 0
  • Exxon foe Engine No. 1 to build fossil fuel plants with Chevron

    0 shares
    Share 0 Tweet 0
  • They Wanted a House in Chicago for Their Growing Family. Would $650,000 Be Enough?

    0 shares
    Share 0 Tweet 0
Solega Blog

Categories

  • Artificial Intelligence
  • Cryptocurrency
  • E-commerce
  • Finance
  • Investment
  • Project Management
  • Real Estate
  • Start Ups
  • Travel

Connect With Us

Recent Posts

Panelists At Senate Banking Hearing On Crypto Market Structure Call For Regulation ASAP

Panelists At Senate Banking Hearing On Crypto Market Structure Call For Regulation ASAP

June 25, 2025
DHL Supply Chain North America appoints new CEO

DHL Supply Chain North America appoints new CEO

June 25, 2025

© 2024 Solega, LLC. All Rights Reserved | Solega.co

No Result
View All Result
  • Home
  • E-commerce
  • Start Ups
  • Project Management
  • Artificial Intelligence
  • Investment
  • More
    • Cryptocurrency
    • Finance
    • Real Estate
    • Travel

© 2024 Solega, LLC. All Rights Reserved | Solega.co