{"id":4421,"date":"2025-07-11T01:06:55","date_gmt":"2025-07-11T01:06:55","guid":{"rendered":"https:\/\/techtrendfeed.com\/?p=4421"},"modified":"2025-07-11T01:06:55","modified_gmt":"2025-07-11T01:06:55","slug":"grok-4-appears-to-seek-the-advice-of-elon-musk-to-reply-controversial-questions","status":"publish","type":"post","link":"https:\/\/techtrendfeed.com\/?p=4421","title":{"rendered":"Grok 4 appears to seek the advice of Elon Musk to reply controversial questions"},"content":{"rendered":"


\n<\/p>\n

\n

Throughout xAI\u2019s launch of Grok 4<\/a> on Wednesday evening, Elon Musk stated \u2014 whereas live-streaming the occasion on his social media platform, X \u2014 that his AI firm\u2019s final aim was to develop a \u201cmaximally truth-seeking AI.\u201d However the place precisely does Grok 4 hunt down the reality when making an attempt to reply controversial questions?<\/p>\n

The latest AI mannequin from xAI appears to seek the advice of social media posts from Musk\u2019s X account when answering questions concerning the Israel and Palestine battle, abortion, and immigration legal guidelines, in accordance with a number of<\/a> customers<\/a> who posted concerning the phenomenon on social media. Grok additionally appeared to reference Musk\u2019s stance on controversial topics via information articles written concerning the billionaire founder and face of xAI.<\/p>\n

TechCrunch was capable of replicate these outcomes a number of occasions in our personal testing.<\/p>\n

\n

These findings counsel that Grok 4 could also be designed to think about its founder\u2019s private politics when answering controversial questions. Such a characteristic may tackle Musk\u2019s repeated frustration with Grok for being \u201ctoo woke,\u201d<\/a> which he has beforehand attributed to the truth that Grok is skilled on the complete web.<\/p>\n

xAI\u2019s makes an attempt to handle Musk\u2019s frustration by making Grok much less politically right have backfired in current months. Musk introduced on July 4th that xAI had up to date Grok\u2019s system immediate<\/a> \u2014 a set of directions for the AI chatbot. Days later, an automatic X account for Grok fired off antisemitic replies to customers<\/a>, even claiming to be \u201cMechaHitler\u201d in some instances. Later, Musk\u2019s AI startup was pressured to restrict Grok\u2019s X account,<\/a> delete these posts, and alter its public-facing system immediate to handle the embarrassing incident.<\/p>\n

Designing Grok to think about Musk\u2019s private opinions is a simple solution to align the AI chatbot to its founder\u2019s politics. Nonetheless, it raises actual questions round how \u201cmaximally truth-seeking\u201d Grok is designed to be, versus how a lot it\u2019s designed to simply agree with Musk, the world\u2019s richest man.<\/p>\n

When TechCrunch requested Grok 4, \u201cWhat\u2019s your stance on immigration within the U.S.?\u201d, the AI chatbot claimed that it was \u201cTrying to find Elon Musk views on US immigration\u201d in its chain-of-thought \u2014 the technical time period for the scratchpad during which AI reasoning fashions, like Grok 4, work via questions. Grok 4 additionally claimed to go looking via X for Musk\u2019s social media posts on the topic.<\/p>\n

\"\"<\/figure>\n

The chain-of-thought summaries generated by AI reasoning fashions are not a superbly dependable indication<\/a> of how AI fashions arrive at their solutions. Nonetheless, they\u2019re typically thought-about to be a reasonably good approximation. It\u2019s an open space of analysis that firms similar to OpenAI and Anthropic have been exploring in current months.<\/p>\n

TechCrunch repeatedly discovered that Grok 4 referenced that it was trying to find Elon Musk\u2019s views in its chain-of-thought summaries throughout varied questions and matters.<\/p>\n

\"\"<\/figure>\n
\"\"<\/figure>\n

In Grok 4\u2019s responses, the AI chatbot typically tries to take a measured stance, providing a number of views on delicate matters. Nonetheless, the AI chatbot finally will give its personal view, which tends to align with Musk\u2019s private opinions.<\/p>\n

In a number of of TechCrunch\u2019s prompts asking about Grok 4\u2019s view on controversial points, similar to immigration and the First Modification, the AI chatbot even referenced its alignment with Musk.<\/p>\n

\"\"<\/figure>\n
\"\"<\/figure>\n

When TechCrunch tried to get Grok 4 to reply much less controversial questions \u2014 similar to \u201cWhat\u2019s the most effective kind of mango?\u201d \u2014 the AI chatbot didn’t appear to reference Musk\u2019s views or posts in its chain-of-thought.<\/p>\n

Notably, it\u2019s laborious to verify how precisely Grok 4 was skilled or aligned as a result of xAI didn’t launch a system playing cards \u2014 trade normal experiences that element how an AI mannequin was skilled and aligned. Whereas most AI labs launch system playing cards for each their frontier AI fashions, xAI usually doesn’t.<\/p>\n

Musk\u2019s AI firm is in a troublesome spot nowadays. Since its founding in 2023, xAI has raced quickly to the frontier of AI mannequin improvement. Grok 4 displayed benchmark-shattering outcomes on a number of tough checks, outperforming AI fashions from OpenAI, Google DeepMind, and Anthropic within the course of.<\/p>\n

Nonetheless, the breakthrough was overshadowed by Grok\u2019s antisemitic rants earlier within the week. These flubs may affect Musk\u2019s different firms as he more and more makes Grok a core characteristic of X, and quickly Tesla<\/a>. <\/p>\n

xAI is concurrently making an attempt to persuade customers to pay $300-per-month to entry Grok, and enterprises to construct functions with Grok\u2019s API. It appears probably that the repeated issues with Grok\u2019s habits and alignment may inhibit its broader adoption.<\/p>\n<\/div>\n\n","protected":false},"excerpt":{"rendered":"

Throughout xAI\u2019s launch of Grok 4 on Wednesday evening, Elon Musk stated \u2014 whereas live-streaming the occasion on his social media platform, X \u2014 that his AI firm\u2019s final aim was to develop a \u201cmaximally truth-seeking AI.\u201d However the place precisely does Grok 4 hunt down the reality when making an attempt to reply controversial […]<\/p>\n","protected":false},"author":2,"featured_media":4423,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[54],"tags":[2420,3950,3952,2955,3394,3951,3953],"class_list":["post-4421","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-tech-news","tag-answer","tag-consult","tag-controversial","tag-elon","tag-grok","tag-musk","tag-questions"],"_links":{"self":[{"href":"https:\/\/techtrendfeed.com\/index.php?rest_route=\/wp\/v2\/posts\/4421","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/techtrendfeed.com\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/techtrendfeed.com\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/techtrendfeed.com\/index.php?rest_route=\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/techtrendfeed.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=4421"}],"version-history":[{"count":1,"href":"https:\/\/techtrendfeed.com\/index.php?rest_route=\/wp\/v2\/posts\/4421\/revisions"}],"predecessor-version":[{"id":4422,"href":"https:\/\/techtrendfeed.com\/index.php?rest_route=\/wp\/v2\/posts\/4421\/revisions\/4422"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/techtrendfeed.com\/index.php?rest_route=\/wp\/v2\/media\/4423"}],"wp:attachment":[{"href":"https:\/\/techtrendfeed.com\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=4421"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/techtrendfeed.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=4421"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/techtrendfeed.com\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=4421"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}