Ace quick missions & earn crypto rewards while gaining real-world Web3 skills. Participate Now! 🔥
Key Takeaways
Ace quick missions & earn crypto rewards while gaining real-world Web3 skills. Participate Now! 🔥
Elon Musk has shared a new goal for his artificial intelligence (AI) company, xAI, which is to rebuild the information used to train its chatbot, Grok.
In a June 21 post on X, he said the Grok 3.5 model should be used to "rewrite the entire corpus of human knowledge, adding missing information and deleting errors".
Musk argued that current AI models are built on too much poor-quality information. His solution is to have Grok fix the content first, by correcting mistakes and filling in gaps, and then use that version to improve its own training.
Did you know?
Subscribe - We publish new crypto explainer videos every week!
What is Balancer in Crypto? Beginner Friendly BAL Explainer
Musk also asked users on X to send in examples of "divisive facts" to help train Grok. He said these facts should be "politically incorrect, but nonetheless factually true".
Bernardino Sassoli de’ Bianchi, a professor at the University of Milan, wrote on LinkedIn that the proposal was "dangerous". He said that when powerful people attempt to alter historical records to align with their opinions, it is not innovation, but rather control over the story. He added that shaping training data to fit someone’s beliefs is wrong from every angle.
Additionally, Gary Marcus, an AI startup founder and retired professor at New York University, compared the plan to a dystopian novel. He commented on Musk's post, "Straight out of 1984", and added that Musk seemed to want Grok to reflect his personal views by changing the record of what happened.
Recently, the Southern Environmental Law Center (SELC) filed a lawsuit against xAI's data center in South Memphis under the Clean Air Act. What happened? Read the full story.
To ensure the highest level of accuracy & most up-to-date information, BitDegree.org is regularly audited & fact-checked by following strict editorial guidelines & review methodology.
Carefully selected industry experts contribute their real-life experience & expertise to BitDegree's content. Our extensive Web3 Expert Network is compiled of professionals from leading companies, research organizations and academia.