AI burns false claims after Charlie Kirk's death, CBS News analysis reveals

AI Video & Visuals


False claims, conspiracy theories, and claims of posts named by people not associated with the case, and spread rapidly across social media after posting The murder of conservative activist Charlie Kirk On Wednesday, some were amplified and fueled by AI tools.

CBS News has identified 10 posts by Grok, the AI ​​chatbot from X. Tyler Robinson, resident of South Utahhas been released. Grok ultimately generated a response that he misidentified the suspect, but by then, posts featuring the wrong person's face and name have already been circulating across X.

The chatbot also generated a modified “enhance” of photos released by the FBI. One such photo was reposted by the Washington County Sheriff's Office in Utah and later posted an update that said “this looks like an AI-enhanced photo” that distorted clothing and facial features.

One image of Ai-enhanced depicted a man much older than 22-year-old Robinson. The AI-generated video, which smoothes the suspect's functionality and clutters the shirt design, has been posted by X users with over 2 million followers and has been reposted thousands of times.

After Utah Gov. Spencer Cox announced it on Friday morning. Suspects in custody Robinson said Glock's response to an X user inquiry about him was inconsistent. According to one Groke post, Robinson is a registered Republican, but other posts reported that he is a nonpartisan voter. Voter Registration Records show that Robinson is not affiliated with political parties.

CBS News also identified 12 cases in which Grok said Kirk was alive the day after his death. Other GROK responses have been dubbed the FBI's reward offer “Hoax” and have given the false assassination date, stating that reports of Kirk's death were “contradictory” after his death was confirmed.

The most generative AI tools generate results based on probability. This could make S. Shyam Sundar, a professor at Penn State University and director of the Social Director Artificial Intelligence Center, difficult to provide accurate information in real time as events unfold.

“They are looking at the next word or the next passage that is most likely,” Sundar said. “It's not based on fact checks, it's not based on reports of any kind in the scene, it's based on the likelihood that this event will occur.

X did not respond to requests for comments regarding false information posted by Grok.

Meanwhile, the X-bot of the AI-powered search engine Perplexity described the filming as a “virtual scenario” in its first post since, suggesting that a White House statement about Kirk's death has been produced.

A spokesman for Perplexity told CBS News that “precise AI is the core technology we build and is central to the experience of every product,” but “Perplexity doesn't claim to be 100% accurate.”

Another spokesperson added that the X-bot is not the latest. There have been improvements the company has made to its technology, and the company has since removed the bot from X.

Google's AI Summary also provided inaccurate information, summaries of search results that may appear at the top of a search. Late Thursday evening, the AI ​​summary of Hunter Kozak's search, the last person to ask questions before Kirk was killed, misidentified him as the FBI's looking for interest. By Friday morning, incorrect information was no longer displayed in the same search.

“The majority of queries seeking information on this topic will return high quality, accurate responses,” a Google spokesperson told CBS News. “Given the rapidly evolving nature of this news, it is possible that the system misinterpreted web content or missed context, as all search functions can take into account the scale of the open web.”

Sundar told CBS News that people tend to perceive AI as less biased or more reliable than people they don't know online.

“We don't think that machines are partisan or biased or want to sow the seeds of objections,” Sundar said. “If it's just a social media friend or someone on your contact list sent something to your feed with an unknown pedigree… they could trust more people than random people.”

Utah Governor Cox said the misinformation could also come from foreign sources. He said at a press conference Thursday that foreign enemies, including Russia and China, have bots that are “instilling disinformation and trying to encourage violence.” Cox urged listeners to spend less time on social media.

“I would suggest you ignore them, turn those streams off and spend a little more time with our family,” he said.



Source link

Leave a Reply

Your email address will not be published. Required fields are marked *