In a recent announcement, OpenAI CEO Sam Altman addressed user complaints about ChatGPT’s perceived laziness, assuring the community that the issue has been resolved with a software update. Altman stated that GPT-4, the latest iteration of the chatbot, is expected to be “much less lazy now.”
Altman shared his sentiments in a post on platform X, acknowledging the initial hiccups GPT-4 faced in adhering to its New Year’s resolutions. The CEO’s comments come after numerous users reported instances where ChatGPT refused to complete tasks and even exhibited a sassy attitude.
Late last year, one frustrated startup founder requested ChatGPT to generate a list of all the weeks between November 2023 and May 2024, only to receive a response claiming the chatbot couldn’t provide an “exhaustive list.”
This sparked a wave of similar complaints from users who experienced challenges in obtaining desired outputs.
“It’s good that we’re seeing more people voicing their concerns, there’s two topics being simultaneously discussed here and perhaps we should discuss them separately,” a user complained in December last year.
He said that ChatGPT’s performance in handling files is dwindling (or is gradually ramping up) and that the chatbot’s general performance is slowing down, probably intelligently, to save on resources, processing, and costs.
“Although I’m beginning to understand point 1 better from the responses on this thread, I still feel that point 2 is more prevalent, I see more and more of it daily, even though I clearly instruct ChatGPT to “provide full and complete code with comments” I end up getting with missing code, shortcuts, deflection of responsibility for me to carry out its instructions, or runtime errors or network errors, etc.
“I read a joke that it probably got promoted to a senior developer level and now talks at a high level just like a senior dev does, which although sounds like a joke might be true, who knows?
“I wish to see this fixed. Or else why would OpenAI charge monthly fees if its performance isn’t satisfactory,” he concluded.
Some users, however, found creative ways to incentivize ChatGPT to provide more comprehensive responses. One user reported that the AI model would offer longer answers if promised a $200 tip.
OpenAI took note of these issues and swiftly responded with a software update in January, aiming to address the reported “laziness” in the advanced GPT-4 “turbo” model.
The root cause of ChatGPT’s reluctance to perform tasks remains unclear. AI models, by their nature, can exhibit unpredictable behaviors that may be challenging for their creators to fully comprehend. OpenAI emphasized that different training runs can lead to models with diverse personalities and quirks, even when trained on the same data.
Speculations arose regarding the cause of ChatGPT’s lackadaisical behavior, with one developer suggesting the possibility of the chatbot taking a winter break. Rob Lynch shared on X that his test on GPT-4 turbo indicated statistically shorter answers when the model “thought” it was December rather than May.
As OpenAI continues to fine-tune and enhance its AI models, Altman assured users that the recent fix should significantly improve ChatGPT’s responsiveness and willingness to complete tasks.
The company said it remains committed to addressing user concerns promptly and maintaining a positive user experience as it navigates the evolving artificial intelligence industry.