Have you ever had the feeling that something that was once revolutionary is starting to lose its edge? If you’ve noticed that AI platforms like ChatGPT OpenAI’s apps seem less brilliant than they once were, you’re not alone. This phenomenon has sparked a heated debate among experts and users, bringing to light doubts and perplexities about a technology which seemed to promise so much.
A concrete example of disillusionment
Steven Vaughan Nicholsan author of Computerworldrecently shared his frustration in a scathing opinion. According to him, the main models of artificial intelligence accessible to the public, such as ChatGPT And Claudethey no longer work as well as the versions previous ones. “The result is often annoyingly wrong,” he writes. But it’s not just a question of errors; it is the randomness of the errors to be more disturbing. If the answers were mediocre but consistently accurate, yes could be still rely on it. But that is not the case.
This feeling is not isolated. One user has commented on a developer forum OpenAI that the latest version of GPT was a “big disappointment” after all the hype. Not surprises that many are wondering: what is happening?
An underlying problem or something more?
Usually, we expect new versions of a software are better than previous ones. So why do these AIs seem to be getting worse? One theory is that AI was never as powerful as it seemed. Perhaps we got caught from the enthusiasm of seeing something that worked, even if it wasn’t perfect.
Another explanation, suggested by Vaughan–Nicholsconcerns a phenomenon called “model collapse”. When AI models are fed data generated by other AIs, their quality deteriorates. It’s as if the intelligence artificial was “eating” if sameprogressively worsening with each training cycle.
What if we were Really running out of contents high-quality created by humans? Some experts estimate that we could face this reality by 2026. We could rediscover the value of human work, unique and irreplaceable? Perhaps, but not everyone is so optimistic.
What do you think about everything? This? Artificial Intelligence is it really becoming less intelligent? And how should we deal with this situation? Leave a comment comment below and let’s do it know your opinion!
#artificial #intelligence #losing #effectiveness