OpenAI Launches More Human-Like, Creative Chatbot `GPT-4` Globally, What Are Features That Previous Versions?

OpenAI GPT-4, which can also accept visual Inputs apart from text-inputs, is more reliable, creative, and able to handle much more nuanced instructions than GPT-3.5.

New Delhi: Artificial research firm OpenAI has launched the more advanced and sophisticated version of chatbot called ‘GPT-4’, taking the deep and machine learning at the next level. The new version of chatGPT will not only "exhibit human-level performance on various professional and academic benchmark but also it accepts image and text inputs". It is available for ChatGPT Plus subscribers with a usage cap. 

How GPT-4 Is Different From Earlier GPT 3.5?

The earlier version of GPT only accepted text-inputs and weren’t as much human-alike as it is. In an easy language, GPT-4 is more reliable, creative, and able to handle much more nuanced instructions than GPT-3.5. 

“In a casual conversation, 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 reliable, creative, and able to handle much more nuanced instructions than GPT-3.5,” OpenAI informed in the blog.

What Are The Features Of GPT-4?

Visual Inputs

It can accept a prompt of text and images, letting the users to expand the options and allowing them to specify any vision or language task. However, image inputs are still a research preview and not publicly available.

Risks & Mitigation

GPT-4 is more safer and secure than the previous versions to mitigate and filter out harmful advice, buggy code and inaccurate information.

“GPT-4 to make it safer and more aligned from the beginning of training, with efforts including selection and filtering of the pretraining data, evaluations and expert engagement, model safety improvements, and monitoring and enforcement,” OpenAI blog said.

What Are The Current Limitations?

GPT-4 has similar limitations as earlier GPT models. It “still not fully reliable as it “hallucinates facts and makes reasoning errors”.

“Great care should be taken when using language model outputs, particularly in high-stakes contexts, with the exact protocol (such as human review, grounding with additional context, or avoiding high-stakes uses altogether) matching the needs of a specific use-case,” the blog added.

Who Can Use new GPT-4?

ChatGPT Plus subscribers will get GPT-4 access on chat.openai.com with a usage cap. The company may introduce a new subscription level for higher-volume GPT-4 usage and also allow some amount of free GPT-4 queries to those who don’t have the subscritpion.