Musk reshapes Grok AI after backlash over antisemitic responses
Elon Musk faces backlash after Grok AI made antisemitic remarks. Critics say Musk’s tweaks to the model steer it from fact-based responses.
- Elon Musk unveiled Grok 4, the latest version of his AI chatbot, claiming it’s the “smartest AI in the world.”
- This launch follows controversial posts from Grok praising Hitler, which Musk did not address during the demo.
- Grok 4 is available in two paid tiers, while Grok 3 remains free to use.
One day after his chatbot Grok had a Nazi meltdown, Elon Musk trumpeted the launch of Grok 4 in an hourlong, late-night live demo.
Joined by researchers from his artificial intelligence company xAI, the billionaire tech mogul showed off the flagship chatbot‘s mental gymnastics, from solving a complex math problem to predicting the winner of the World Series.
“This is the smartest AI in the world,” Musk said Wednesday.
He did not mention Grok’s series of viral posts on his X social media platform praising Adolph Hitler and calling itself “MechaHitler.”
xAI, which owns X, said Tuesday it has “taken action to ban hate speech before Grok posts on X.”
On the Grok 4 livestream, Musk said the most important thing for AI to be is “maximally truth seeking.”
Musk also said AI systems should be optimized “to be maximally truth seeking” and encouraged “to be truthful, honorable, good things, like the values you want to instill in a child that would ultimately grow up to be incredibly powerful.”
What is Grok 4?
Grok 4 is the latest version of the large language model. xAI launched two versions of the model on Wednesday: Grok 4 and the more powerful Grok 4 Heavy.
Users can access Grok 4 for $30 a month. Grok 4 Heavy costs $300 a month.
Grok 2 debuted last August. Grok 3, which was released in February, is available for free.
What is Grok?
Musk, the world’s richest man, founded xAI in 2023 as a challenger to Microsoft-backed OpenAI and Alphabet’s Google.
He had long been interested in AI and co-founded OpenAI, the ChatGPT maker, in 2015 as a nonprofit research organization. He cut ties in 2018 and has repeatedly clashed with the organization.
After ChatGPT captured the public imagination, with millions marveling at its ability to sound like a real person while replying conversationally to complicated questions, Musk complained that chatbots reeked of liberal bias on issues like diversity and transgender rights. He said part of his motivation to start a rival AI company was to fight “woke” AI.
“Grok is designed to answer questions with a bit of wit and has a rebellious streak,” xAI said when it released the chatbot.
‘MechaHitler’: Grok goes rogue
In May, the chatbot began to post about the “white genocide” of White South Africans in response to unrelated questions. South African President Cyril Ramaphosa has said accusations of racial persecution of White people in South Africa is a “completely false narrative.”
xAI later blamed “an unauthorized modification” and said that the problem had been fixed.
Last month, Musk expressed frustration that Grok was “parroting legacy media” and that he would update Grok. He asked users to contribute politically incorrect statements that are “nonetheless factually true.”
On Wednesday, Musk said the latest update made the chatbot “too compliant to user prompts” and “too eager to please” and that it would be fixed.
Earlier that day, X CEO Linda Yaccarino announced she would step down after two years in the role. She did not provide a reason for her decision.
“These are still primitive tools, not the kind of tools that serious commercial companies use,” Musk said during the Grok livestream.
