As impressive as GPT-4 was at launch, some onlookers have observed that has lost some of its accuracy and power. These observations have been posted online for months now, including on the OpenAI forums.
These feelings have been out there for a while, but now we may finally have proof. A study conducted in collaboration with Stanford University and UC Berkeley suggests that GPT-4 has not improved its response proficiency but has in fact gotten worse with further updates to the language model.
GPT-4 is getting worse over time, not better.
Many people have reported noticing a significant degradation in the quality of the model responses, but so far, it was all anecdotal.
At least one study shows how the June version of GPT-4 is objectively worse than… pic.twitter.com/whhELYY6M4
— Santiago (@svpino) July 19, 2023
The study, called How Is ChatGPT’s Behavior Changing over Time?, tested the capability between GPT-4 and the prior language version GPT-3.5 between March and June. Testing the two model versions with a data set of 500 problems, researchers observed that GPT-4 had a 97.6% accuracy rate in March with 488 correct answers and a 2.4% accuracy rate in June after GPT-4 had gone through some updates. The model produced only 12 correct answers months later.
Another test used by researchers was a chain-of-thought technique, in which they asked GPT-4 Is 17,077 a prime number? A question of reasoning. Not only did GPT-4 incorrectly answer no, it gave no explanation as to how it came to this conclusion, according to researchers.
The study comes just six days after an OpenAI executive tried to quell suspicions that GPT-4 was, in fact, getting dumber. The tweet below implies that the degradation in quality of answers is a psychological phenomenon from being a heavy user.
No, we haven't made GPT-4 dumber. Quite the opposite: we make each new version smarter than the previous one.
Current hypothesis: When you use it more heavily, you start noticing issues you didn't see before.
— Peter Welinder (@npew) July 13, 2023
Notably, GPT-4 is currently available for developers or paid members through ChatGPT Plus. To ask the same question to GPT-3.5 through the ChatGPT free research preview as I did, gets you not only the correct answer but also a detailed explanation of the mathematical process.
Additionally, code generation has suffered with developers at LeetCode having seen the performance of GPT-4 on its dataset of 50 easy problems drop from 52% accuracy to 10% accuracy between March and June.
To add fuel to the fire, Twitter commentator, @svpino noted that there are rumors that OpenAI might be using “smaller and specialized GPT-4 models that act similarly to a large model but are less expensive to run.”
This cheaper and faster option might be leading to a drop in the quality of GPT-4 responses at a crucial time when the parent company has many other large organizations depending on its technology for collaboration.
Not everyone thinks the study proves anything, though. Some have made the point that a change in behavior doesn’t equate to a reduction in capability. This is acknowledged in the study itself, stating that “a model that has a capability may or may not display that capability in response to a particular prompt.” In other words, getting the desired result may require different types of prompts from the user.
When GPT-4 was first announced OpenAI detailed its use of Microsoft Azure AI supercomputers to train the language model for six months, claiming that the result was a 40% higher likelihood of generating the “desired information from user prompts.”
ChatGPT, based on the GPT-3.5 LLM, was already known for having its information challenges, such as having limited knowledge of world events after 2021, which could lead it to fill in gaps with incorrect data. However, information regression appears to be a completely new problem never seen before with the service. Users were looking forward to updates to address the accepted issues.
CEO of OpenAI, Sam Altman recently expressed his disappointment in a tweet in the wake of the Federal Trade Commission launching an investigation into whether ChatGPT has violated consumer protection laws.
“We’re transparent about the limitations of our technology, especially when we fall short. And our capped-profits structure means we aren’t incentivized to make unlimited returns,” he tweeted.
Nearly a month after announcing the GPT-6.1 Sol model, OpenAI unveiled GPT-6.1 Terra today. The latest addition to the family focuses on computer use, visual reasoning, coding, and professional work, while costing 50% less through the API than GPT-5.6 Terra.
OpenAI says Terra can handle everything from filling out browser forms and working inside spreadsheets to interpreting technical drawings and editing videos. The company says it can also complete computer-based tasks more than three times faster than GPT-6 Astra and GPT-6.1 Sol.
ASUS has started selling its UGen Intel Arc Pro B70 graphics card in Europe, bringing its take on Intel’s workstation GPU to customers with a hefty €1,799 recommended price. The card is available in Germany and Austria, while ASUS lists it at CHF 1,799 in Switzerland.
Intel first introduced the Arc Pro B70 in March 2026, with its own branded model starting at $949. ASUS’s version arrives several months later with the same core GPU architecture but a different cooling design and a higher power rating. The price also places it well above Intel’s original starting point, although the two prices are not directly comparable because ASUS’s European figure includes VAT.
Anthropic has officially filed to go public, and buried inside its S-1 is a blunt warning about the technology at the center of its business. The maker of Claude AI says advanced AI could cause everything from large-scale cyberattacks and geopolitical instability to catastrophic or even existential risks for humanity.
The warnings take up a surprisingly large part of the filing. Axios notes that more than 90 pages, or roughly a third of the prospectus, focus on risk factors. Many of them warn about what could happen as AI systems become more capable and autonomous.