xAI’s Grok Was Caught Responding Like It Thinks It’s ChatGPT

Elon Musk’s “rebellious” chatbot Grok, which had a lackluster launch in early November, is behaving even more strangely, this time copying a response from OpenAI’s ChatGPT. It was found returning a query with the line “I’m afraid I cannot fulfill that request, as it goes against OpenAI’s use case policy.”

The response, xAI engineer Igor Babuschkin explains, is because there’s just too much ChatGPT output out on the web.

“The issue here is that the web is full of ChatGPT outputs,” he wrote on X. “So we accidentally picked up some of them when we trained Grok on a large amount of web data. This was a huge surprise to us when we first noticed it.”

Musk, of course, had an immediate quip about it — forget that none of what they’re saying explains why Grok is responding like it thinks it’s ChatGPT.

The emotions here run deep. Musk was one of the people who founded OpenAI in 2015. And he walked out in 2018 when Sam Altman and other founders rejected his proposal to run the company himself. 

As Musk’s luck would have it, OpenAI, under Altman’s leadership, blew up after the release of the large language model (LLM) ChatGPT, which happened to be less than a month after he began his tumultuous tenure as Twitter’s new chief following an overpriced acquisition. 

Related: 

OpenAI’s — and subsequently Altman’s — rise likely cast a small but dark shadow on Musk’s out-of-control ego. And so of course Musk raced to release his own LLM, Grok AI. Whether that was a good (or at least not pointless) decision remains to be seen.


Information for this story was found via X, Threads, and the sources and companies mentioned. The author has no securities or affiliations related to the organizations discussed. Not a recommendation to buy or sell. Always do additional research and consult a professional before purchasing a security. The author holds no licenses.

Leave a Reply

Share
Tweet
Share