Grok, Elon Musk's AI platform, has recently caused a sensation due to an astonishing discovery. ChatGPT shared a screenshot showing Grok's response to a prompt, which was almost identical to ChatGPT's response and even mentioned OpenAI. This has sparked a renewed discussion about whether Grok uses OpenAI's code, despite Musk's previous denial. This ongoing discussion reflects people's curiosity and uncertainty about the relationship between Grok and OpenAI.

Elon Musk quickly responded to ChatGPT's post, suggesting that the similarity in responses may be because ChatGPT trained on data obtained from his platform. Musk implied that ChatGPT may have learned something from Grok's content, indicating that this similarity is not a result of Grok being trained on OpenAI's code.

Elon Musk values feedback about Grok and actively engages in positive and negative comments. In one example, he responded with a smiling face emoji to a user's share. The user stated that Grok is the best AI model ever. Musk's involvement on social media reflects his interest in the public's perception of Grok and the ongoing discussion about its capabilities.

ChatGPT vs Grok

Both ChatGPT and Grok rely on prompt-based structures and have similarities in their methods of generating responses. However, they differ in terms of real-time information retrieval and training data environment.

The basic version of ChatGPT operates under the limitation of information available before 2021. Users seeking real-time data need to opt for the Plus version, which incurs a certain cost. This model aims to generate responses based on provided prompts but has limitations in terms of novelty of information it can provide.

In contrast, Grok, developed by xAI, sets itself apart by obtaining real-time information through the social media platform X. This feature allows Grok to provide users with the latest data, giving it a clear advantage over ChatGPT's basic version.

However, Grok faced scrutiny and controversy after its release, as some speculated that it may have been trained on OpenAI's code. To address these concerns, Grok adheres to xAI's use case policy, limiting access to certain information.

Igor Babuschkin, an X user associated with xAI, explained this suspicion, stating that Grok unintentionally included some of ChatGPT's outputs during the training process. This unintentional overlap is due to the widespread presence of ChatGPT data online. Babuschkin assured users that Grok was not created using any of OpenAI's code and promised corrective measures in future versions to prevent such issues.

This situation highlights the complexity and challenges of training large language models, especially when utilizing data from the internet. It emphasizes the importance of transparency, timely issue resolution, and implementation of corrective measures to ensure the reliability and integrity of AI models like Grok.