OpenAI has introduced CriticGPT, an advanced AI tool based on the GPT-4 architecture, to help developers identify and correct errors in code generated by ChatGPT. This tool aims to enhance code review processes by 60% compared to traditional methods and will integrate OpenAI’s RLHF pipeline for more effective assessment of AI-generated content.
CriticGPT is trained by human developers editing ChatGPT-generated code, introducing errors and providing feedback to help it recognize coding mistakes. While it may struggle with identifying issues spread across multiple code strings, it has shown impressive results in pinpointing errors overlooked by human reviewers, marking a significant advancement in AI-assisted code review.
OpenAI is working on the next-generation AI model, expected to be released in 1.5 years, with intelligence equivalent to a PhD holder in specific tasks. The company has also partnered with Time Magazine to access over 100 years of archives for training and enhancing ChatGPT, highlighting a commitment to accurate information and smarter AI systems.
Overall, CriticGPT represents a step forward in improving the accuracy of AI systems, with future developments aiming to enhance large language models and their generative capabilities. The collaboration with Time Magazine underscores a focus on reliable information and the advancement of AI technology.
Source link
Source link: https://www.indiatvnews.com/amp/technology/news/openai-launches-criticgpt-to-enhance-code-quality-2024-07-03-940008