New Study Suggests ChatGPT Is Getting Dumber—But, Is It?

A new study released on Tuesday by researchers from Stanford University and University of California, Berkeley has ignited debate within the AI community regarding the performance of OpenAI’s GPT-4 language model. 

The paper, titled “How Is ChatGPT’s Behavior Changing over Time?” and published on arXiv by Lingjiao Chen, Matei Zaharia, and James Zou, investigates changes in GPT-4’s outputs over a span of a few months, suggesting a potential decline in coding and compositional task abilities.

The study utilized API access to test GPT-3.5 and GPT-4 versions from March and June 2023 on various tasks, including math problem-solving, answering sensitive questions, code generation, and visual reasoning. 

Source: Chen, Zaharia, Zou

Notably, the research found a significant drop in GPT-4’s ability to identify prime numbers, plummeting from an accuracy of 97.6 percent in March to just 2.4 percent in June. Surprisingly, GPT-3.5 displayed improved performance during the same period.

This investigation comes amidst growing concerns expressed by users who have observed a subjective decline in GPT-4’s performance over the past few months. Speculations about the reasons behind this decline abound, including OpenAI’s possible distillation of models to enhance efficiency, fine-tuning to mitigate harmful outputs, and unfounded conspiracy theories suggesting a reduction in GPT-4’s coding capabilities to promote GitHub Copilot usage.

OpenAI has consistently denied the alleged decrease in GPT-4’s capabilities. Peter Welinder, OpenAI’s VP of Product, recently took to Twitter to counter the claims, asserting that each new version of the AI language model is more advanced than its predecessor. He posits that intensive usage may lead to heightened awareness of these perceived issues.

But the company, unlike its large language model, isn’t completely closed to the possibility. Logan Kilpatrick, OpenAI’s head of developer relations, confirmed on Twitter that the team is aware of the reported regressions and is actively investigating the matter.

Arvind Narayanan, a computer science professor at Princeton, also pointed out some problems in the study, arguing that the research’s findings do not definitively prove a decline in GPT-4’s performance and could align with fine-tuning adjustments made by OpenAI. 


Information for this story was found via Twitter, Gizmodo, Ars Technica, 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.

One Response

  1. This issue isn’t even dumber or smarter – the issue is the performance is changing radically even in a short time. Inconsistency directly correlates to unreliability, and this isn’t a GIGO situation so much as GO potentially happening any time for any reason.

Video Articles

Silver @ $36 & We’re Still 200M Oz Short | Paul Andre Huet – Americas Gold and Silver

Gold’s Wild Bull Run: Are Markets About to Break? | Mike McGlone

First Majestic Silver: The Santa Elena Mine

Recommended

Goliath Resources Closes Out Funding Round With Total Gross Proceeds Of $27.1 Million

Silver47 Kickstarts 4,000 Metre Drill Campaign At Red Mountain Project

Related News

OpenAI Seeks Record $40 Billion Investment at $300 Billion Valuation

OpenAI is in early discussions to raise up to $40 billion in funding that would...

Friday, January 31, 2025, 11:10:00 AM

Apple-OpenAI Deal Apparently Doesn’t Cost A Thing For Apple

In a move that has captured the tech industry’s attention, Apple (NASDAQ: AAPL) and OpenAI...

Friday, June 14, 2024, 12:59:00 PM

OpenAI Deletes Evidence in New York Times Copyright Lawsuit

OpenAI, the artificial intelligence company behind ChatGPT, has inadvertently erased crucial evidence in its ongoing...

Friday, November 22, 2024, 03:08:00 PM

Musk Looks to Distract from Tesla Earnings with Microsoft Lawsuit Threat

Elon Musk has a lot on his plate.  On Wednesday, the Twitter owner and Tesla...

Thursday, April 20, 2023, 11:12:15 AM

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...

Thursday, January 4, 2024, 06:43:00 AM