ChatGPT ran into trouble with some strange answers to the queries and one or two completely wrong answers. Fortunately, GPT-4 is more accurate than ChatGPT. OpenAI stated that GPT-4 is 82% less likely to respond to requests for content that OpenAI does not allow and 60% less likely to invent responses. OpenAI has announced its long-awaited GPT-4, available to ChatGPT Plus subscribers.
OpenAI released GPT-4 on March 14 and claims that the updated version is more reliable and creative than its predecessor. However, those who have tried Microsoft's Bing, which works with artificial intelligence, have already experienced some of the new technologies. On Tuesday, Microsoft said that its AI-powered Bing worked with a version of GPT-4 customized for search. The GPT-4 is multimodal, meaning that it can view and process the indications in the images as well as the text.
The GPT-4 can capture and generate about eight times more text than ChatGPT. This means that the chatbot will have a longer memory and will be able to keep up with longer conversations. OpenAI said that the latest version could process up to 25,000 words, compared to the previous 3,000 words. The GPT-4 may be bad news for fans of ChatGPT's evil alter egos.
The company said the system was 82% less likely to respond to requests for disallowed content. Many users have tried to trick ChatGPT into responding inappropriately or unmoderating its content, which has probably provided OpenAI with many examples of malicious messages. The company said that the changes may be subtle in casual conversations, but that they would be clear when the bot is faced with complex situations. The GPT-4 is also capable of handling much more nuanced instructions than the GPT-3.5, OpenAI said.
