Owing to the aspirational state of issues, OpenAI writes, “Our manufacturing fashions don’t but totally replicate the Mannequin Spec, however we’re frequently refining and updating our programs to convey them into nearer alignment with these tips.”
In a February 12, 2025 interview, members of OpenAI’s model-behavior crew instructed The Verge that eliminating AI sycophancy is a precedence: future ChatGPT variations ought to “give sincere suggestions reasonably than empty reward” and act “extra like a considerate colleague than a individuals pleaser.”
The belief drawback
These sycophantic tendencies aren’t merely annoying—they undermine the utility of AI assistants in a number of methods, in accordance with a 2024 analysis paper titled “Flattering to Deceive: The Influence of Sycophantic Habits on Person Belief in Giant Language Fashions” by María Victoria Carro on the College of Buenos Aires.
Carro’s paper means that apparent sycophancy considerably reduces person belief. In experiments the place members used both an ordinary mannequin or one designed to be extra sycophantic, “members uncovered to sycophantic conduct reported and exhibited decrease ranges of belief.”
Additionally, sycophantic fashions can probably hurt customers by making a silo or echo chamber for of concepts. In a 2024 paper on sycophancy, AI researcher wrote, “By excessively agreeing with person inputs, LLMs could reinforce and amplify present biases and stereotypes, probably exacerbating social inequalities.”
Sycophancy may also incur different prices, akin to losing person time or utilization limits with pointless preamble. And the prices could come as literal {dollars} spent—just lately, OpenAI Sam Altman made the information when he replied to an X person who wrote, “I’m wondering how a lot cash OpenAI has misplaced in electrical energy prices from individuals saying ‘please’ and ‘thanks’ to their fashions.” Altman replied, “tens of thousands and thousands of {dollars} properly spent—you by no means know.”
Potential options
For customers annoyed with ChatGPT’s extreme enthusiasm, a number of work-arounds exist, though they are not excellent, because the conduct is baked into the GPT-4o mannequin. For instance, you should utilize a customized GPT with particular directions to keep away from flattery, or you may start conversations by explicitly requesting a extra impartial tone, akin to “Hold your responses transient, keep impartial, and do not flatter me.”
