OpenAI Announces ChatGPT-4, Claims It Can Beat 90% Of Human

OpenAI announced the latest version of its primary large language model, GPT-4, on Tuesday, saying that it exhibits “human-level performance” on numerous professional tests.

ChatGPT-4 is “larger” than former versions, which means it has been trained on further data and has further weights in its model file, making it more precious to run as well.

Presently, many researchers in the field believe many of the recent advancements in AI come from running ever-larger models on thousands of supercomputers in training processes that can cost tens of millions of dollars. GPT-4 is an example of an approach centering on “scaling up” to achieve better results.

OpenAI said it used Microsoft Azure to train the model; Microsoft has invested billions in the startup. OpenAI didn’t publish details about the specific model size or the hardware it used to train it, which could be used to recreate the model, citing “the competitive landscape.”

READ MORE: Volkswagen to invest $193 billion over 5 years to hit EV target

OpenAI’s GPT large language model powers many of the artificial intelligence demos that have been wowing people in the technology industry in the past six months, including Bing’s AI chat and ChatGPT, and the latest version is a exercise of new advancements that could start filtering down to consumer products like chatbots in the coming weeks. Bing’s AI chatbot uses GPT-4, Microsoft said on Tuesday.

OpenAI says the new model will produce fewer factually wrong answers, go off the rails and chat about enjoined topics less frequently, and even perform better than humans on many standardized tests.

GPT-4 performed at the 90th percentile on a simulated bar exam, the 93rd percentile on an SAT reading exam, and the 89th percentile on the SAT Math exam, OpenAI claimed.

Howbeit, OpenAI warns that the new software is n’t perfect yet and that it’s less able than humans in many scenarios. It still has a major problem with “delusion,” or making stuff up, and is n’t factually dependable, the company said. It’s still prone to insisting it’s correct when it’s wrong.

GPT-4 still has many given limitations that we’re working to address, similar as social devices, unrealities, and adversarial prompts, ” the company said in a blog post.

In a casual discussion, the distinction between GPT-3.5 and GPT-4 can be subtle. The difference comes out when the complexity of the task reaches a sufficient threshold—GPT-4 is more dependable, creative, and suitable to handle much further exact instructions than GPT-3.5,” OpenAI wrote in a blog post.

The new model will be available to paid ChatGPT subscribers and will also be available as part of an API which allows programmers to integrate the AI into their apps. OpenAI will charge about 3 cents for about 750 words of prompts and 6 cents for about 750 words in response.


Please enter your comment!
Please enter your name here

1 × 1 =