Musk works on a ChatGPT rival because it’s “too woke”
Elon Musk has reached out to artificial intelligence researchers to build his own chatbot that could compete with OpenAI’s ChatGPT. He is planning on forming a new research lab.
According to The Information, Musk has criticized OpenAI, a company he co-founded in 2015, for preventing ChatGPT from producing text that might offend users. OpenAI is using extra safeguards to prevent it from happening, and apparently, Musk has different thoughts. He left the company in 2018 due to a few disagreements and has recently been vocally criticizing OpenAI, especially on Twitter. A couple of months back, he tweeted, “the danger of training AI to be woke – in other words, lie – is deadly,” as a response to a Twitter user who asked Sam Altman, CEO of OpenAI, “to turn off the woke settings for GPT and offer that version as an option for users.”
His comments imply that a rival chatbot would have fewer divisive subject restrictions than ChatGPT and a related chatbot Microsoft recently launched.
He thinks “ChatGPT is woke,” and it is very concerning. OpenAI’s security filters are meant to prevent any kinds of biased or harmful answers the chatbot could give to users. Some researchers still think these filters must be improved as they are insufficient to protect users from harmful acts. Especially for marginalized communities, ChatGPT’s biased actions could be very harmful on a psychological level. There are multiple examples of ChatGPT giving harmful answers regarding sex, race, and ethnicity.
Former Google employee Igor Babuschkin to lead the project
Elon Musk recruited Igor Babuschkin to lead the project, a former researcher at Google DeepMind AI, and he spoke to The Information. “The goal is to improve the reasoning abilities and the factualness of these language models. That includes making sure the model’s responses are more trustworthy and reliable,” he said. “Musk’s plan is not to build a chatbot with fewer content safeguards,” he added. The Tesla and SpaceX CEO has been teasing his new project on Twitter under the “Based AI” banner, sharing memes and such. He also replied, “Absolutely,” to a user who retweeted The Information’s article and mentioned how biased ChatGPT is.
His “beef” with OpenAI also includes the company shifting from a non-profit organization to a for-profit enterprise. “OpenAI was created as an open source (which is why I named it ‘Open’ AI), non-profit company to serve as a counterweight to Google, but now it has become a closed source, maximum-profit company effectively controlled by Microsoft. Not what I intended at all,” he said in one of his tweets.
Elon Musk has been teasing his new AI chatbot that could potentially compete with ChatGPT, and it seems like he will implement the project soon.
Thank you for being a Ghacks reader. The post Musk works on a ChatGPT rival because it’s “too woke” appeared first on gHacks Technology News.
gHacks Technology News
