ChatGPT 4 is making typos since last night – OpenAI Forum users complain

Author:

Published:

Updated:

AI writing software

ChatGPT 4 (not 4o) is once again under scrutiny for its spelling and typos, with OpenAI Forum users reporting a noticeable increase in errors since last night.

This isn’t the first time the AI has faced such criticism, as similar issues were reported last year, raising questions about the underlying cause and potential solutions.

One user shared the following details.

On Reddit too there were some complaints:

In this thread someone shared their error with the word “addther” been used, instead of “other”.

Last year some reported issues also went viral.

Incidents from 2023

A Reddit user last year shared a screenshot capturing a typo made by ChatGPT. The AI chatbot, while answering a query on “pet shop recording concerns,” spelled the word “infringing” incorrectly as “infrishing.”

The user, curious about the unusual word, asked ChatGPT for its meaning. The AI promptly corrected itself, apologizing for the confusion and explaining that “infrishing” was a typographical error.

Online Reactions

The screenshot quickly gained traction on Reddit, receiving over 9,100 upvotes and a flurry of responses. Comments ranged from humorous to critical, with one user joking, “Busted! ChatGPT is just humans,” and another speculating that the AI’s training data might be riddled with typos.

Other users shared their own experiences with ChatGPT making similar errors. One user mentioned that the AI had spelled “dam bursting” as “damn bursting,” while another noted that it had mixed up “phased” and “fazed.” These anecdotes highlight that, despite its advanced capabilities, ChatGPT is not immune to making mistakes.

Implications for AI

The incident has led to discussions about the limitations of AI and its potential impact on jobs. Some users expressed relief, suggesting that such errors indicate that AI is not yet capable of replacing human editors and writers. Others pointed out that while AI can be incredibly useful, it is not infallible and should be used with caution.

Conclusion

While ChatGPT’s typos may seem like a minor glitches, it serves as a reminder that even advanced AI systems are not perfect. The incident has provided a moment of levity for internet users and sparked important conversations about the role of AI in our lives. As AI continues to evolve, it will be interesting to see how these technologies improve and what new challenges they present.

Sources

Share this content

Leave a Reply

Your email address will not be published. Required fields are marked *

Latest posts