Saturday, June 14, 2025

Latest

New Study Reminds Us That ChatGPT Does Not Really *Understand* What You Want It To Do

Amid chatter about ChatGPT’s reportedly degrading performance, a new study found that recent open-sourced large language models (LLMs), including OpenAI’s GPT-3, perform “surprisingly better” on datasets released before the training data creation date than on datasets released after. 

The University of California, Santa Cruz paper by Changmao Li and Jeffrey Flanigan suggests that it isn’t that ChatGPT’s performance is degrading because new tasks are different from what the models are trained on, but that we forget that these models, especially the groundbreaking GPT-3, performed astoundingly well because they were trained with massive amounts of data, with a vast amount of examples of what is asked of them, and not particularly because they understand the tasks per se. 

As writing teacher and AI in education specialist Anna Mills puts it, it’s like “it has studied advance copies of lots of tests,” however, “when you give it new tests (tasks with no examples in its training data), it performs worse.”

The paper emphasizes that LLMs use a retrieval-based approach that mimics intelligence, as tech entrepreneur Chomba Bupe points out.

OpenAI may be having trouble catching up. Claims of getting “lazy” (which many have equated to “degrading”) have been plaguing OpenAI’s paid model, GPT-4, in recent weeks. 

The company, through its X account, explains that training a chat model “is not a clean industrial process,” and that it’s “less like updating a website with a new feature and more an artisanal multi-person effort to plan, create, and evaluate a new chat model with new behavior!”


Information for this story was found via X, and the sources and companies mentioned. The author has no securities or affiliations related to the organizations discussed. Not a recommendation to buy or sell. Always do additional research and consult a professional before purchasing a security. The author holds no licenses.

Video Articles

Aznalcóllar Corruption Case Nears Conclusion | David Gower – Emerita Resources

Why Precious Metals Crashed on Trump’s Trade Deals? | Jesse Day

Big Gold Miners Don’t Trust $3,300 Gold, Do They See a Crash Ahead? | Rob McEwen

Recommended

NexGen Energy Receives Approval For Site Expansion At Rook I Project

ESGold Sees Installation Of Humphrey Spirals Completed At Montauban

Related News

Sam Altman Wants To Raise Trillions For The Future Of AI

OpenAI CEO Sam Altman wants to transform the global semiconductor industry. The catch: he needs...

Friday, February 9, 2024, 10:38:00 AM

Sam Altman Claims Ignorance on Equity Clawbacks But Data Says Otherwise

OpenAI CEO Sam Altman faced backlash over the company’s restrictive exit policies that threatened to...

Sunday, May 26, 2024, 09:10:00 AM

OpenAI’s GPT-4o-Powered ChatGPT Is Now More (Terrifyingly) Conversational and Life-Like

OpenAI on Monday announced GPT-4o, a new flagship generative AI model that expands on the...

Tuesday, May 14, 2024, 05:32:00 PM

OpenAI Power Struggle: Employees Vs. Board Vs. Sam Altman, Explained

While the real reason behind Sam Altman’s sudden dismissal as CEO of OpenAI seems to...

Tuesday, November 21, 2023, 09:54:00 AM

Sam Altman Kicked From OpenAI By Board Of Directors

Sam Altman is being removed from the role of chief executive officer of OpenAI, following...

Friday, November 17, 2023, 03:47:03 PM