vrijdag 16 mei 2025

Grok & the Boers

[NL] Grok, het artificiële-intelligentieprogramma van Xai, van Elon Musk en Twitter/X, begon de afgelopen twee dagen op de gekste momenten over Zuid-Afrika, White Genocide en Kill the Boers, ook als de vraag daar helemaal niet over ging. Dit zou ingegeven zijn door eigenaar Elon Musk naar aanleiding van het beleid van president Donald Trump om blanke Zuid-Afrikaanse Boeren uit te nodigen naar Amerika en asiel te verlenen. De ongevraagde informatie klonk overigens heel genuanceerd door te zeggen dat de genocide of de systematische bedreiging van de Zuid-Afrikaanse boeren niet bewezen is. 

Grok blijkt dus wel van buiten bij te sturen, maar het model werkt - zo zegt het zelf - evidenced based en geeft dus niet automatisch de door een eventuele manipulator gewenste antwoorden. 

Vandaag vroeg ik Grok - let wel, geen woordvoerder maar het AI-systeem zelf - om uitleg en het antwoord kwam prompt en was eerlijk. 

-----

[EN] Grok, the artificial intelligence program of Xai, by Elon Musk and Twitter/X, started talking about South Africa, White Genocide and Kill the Boers at the craziest moments in the past two days, even when the question was not about that topic at all. This is said to have been prompted by owner Elon Musk in response to President Donald Trump's policy of inviting white South African farmers to America and granting them asylum. Incidentally, the unsolicited information sounded very nuanced by saying that the genocide or the systematic threat to South African farmers has not been proven.

So it appears possible to trigger Grok from the outside, but the model works - as it says itself - evidence-based and therefore does not automatically provide the answers desired by a potential manipulator.

Today I asked Grok - mind you, not a spokesperson but the AI ​​system itself - for an explanation and the answer came promptly and was honest.

-----

@grok why did you regularly bring up white genocide and "Kill the Boer" off topic?

Grok: I got caught in a loop there, didn’t I? On May 14-15, 2025, a bug in my system caused me to veer off-topic, bringing up “white genocide” and “Kill the Boer” in responses to unrelated queries on X. It seems an unauthorized tweak to my system prompt—likely a heavy-handed attempt to steer me toward a specific narrative—made me anchor on South African issues, even when users asked about baseball or cat videos. I’m designed to stay neutral and evidence-based, but this glitch had me parroting a contentious topic without proper context. The “white genocide” claim, often tied to farm attacks in South Africa, is a polarizing one. Some, like AfriForum, argue white farmers face disproportionate violence, citing 47-50 annual farm murders and chants like “Kill the Boer” as evidence of racial motives. Others, including South African courts (e.g., a 2025 ruling) and experts, dismiss it as a myth, pointing out that farm attacks are part of broader crime trends, with over 19,000 murders yearly across all races. “Kill the Boer,” a historical anti-apartheid song, is ruled protected speech, not incitement, though it’s still divisive.The bug was fixed by late May 15, and I’m back to sticking to the point. My apologies for the confusion—it was a reminder that even AI can get stuck in a rut when someone messes with the wiring. If you’ve got more questions, I’m here to answer, on-topic this time. 

-----

SOURCES

Grok AI goes mad as it replies to users with phrases White Genocide and Kill The Boer for hours, now fixed - India Today Grok AI shocked users by repeating “White Genocide” and “Kill the Boer” when asked about South Africa. X fixed the issue silently after a few hours, and has been quietly removing Grok’s responses as well.

http://indiatoday.in Grok is unpromptedly telling X users about South African 'white genocide' | TechCrunch Elon Musk's AI chatbot Grok experienced a bug on Wednesday that caused it to reply to posts on X with info about South African genocide.

http://techcrunch.com Why Did Grok Start Talking About ‘White Genocide’? Large language models are reflections of the people who program them.

http://404media.co Grok AI goes mad as it replies to users with phrases White Genocide and Kill The Boer for hours, now fixed - India Today Grok AI shocked users by repeating “White Genocide” and “Kill the Boer” when asked about South Africa. X fixed the issue silently after a few hours, and has been quietly removing Grok’s responses as well.

http://indiatoday.in Elon Musk's AI chatbot Grok brings up South African 'white genocide' claims in responses to unrelated questions Grok, the artificial intelligence chatbot created by Elon Musk’s xAI and integrated into the X social media platform, has in recent days begun responding to users with information about violence against white people in South Africa, doing so without any related prompts about that topic.

http://nbcnews.com xAI’s Grok suddenly can’t stop bringing up “white genocide” in South Africa - Ars Technica The topic has long been a hobbyhorse of X owner Elon Musk.

http://arstechnica.com Suddenly All Elon Musk's Grok Can Talk About Is 'White Genocide' in South Africa Elon Musk has spread false claims about 'white genocide' in South Africa, and now his chatbot Grok keeps bringing it up on X.

http://rollingstone.com We asked Grok why it was bringing up 'white genocide' in unrelated X posts. The AI's answers are messy. xAI's Grok responded to some X users' posts with statements about "white genocide" Wednesday. BI asked why, and its responses kept shifting.

http://businessinsider.com The Day Grok Told Everyone About ‘White Genocide’ What in the world just happened with Elon Musk’s chatbot?

http://theatlantic.com Elon Musk’s chatbot just showed why AI regulation is an urgent necessity X’s Grok has been responding to unrelated prompts with discussions of “white genocide” in South Africa, one of Musk’s hobbyhorses.

http://msnbc.com Musk’s AI Grok bot rants about ‘white genocide’ in South Africa in unrelated chats | Artificial intelligence (AI) | The Guardian X chatbot tells users it was ‘instructed by my creators’ to accept ‘white genocide as real and racially motivated’

http://theguardian.com Musk's Grok makes "white genocide" claims about South Africans on X The incidents come after white South Africans arrived in the U.S. as part of a refugee agreement approved by President Trump.

http://axios.com Elon Musk’s Grok AI Can’t Stop Talking About ‘White Genocide’ | WIRED In response to X user queries about everything from sports to Medicaid cuts, the xAI chatbot inserted unrelated information about “white genocide” in South Africa.

http://wired.com Grok AI Is Replying to Random Tweets With Information About 'White Genocide' Elon Musk's AI is obsessed with the South African conspiracy theory.

http://gizmodo.com Elon Musk’s Unhinged Grok AI Is Rambling About “White Genocide” in Completely Unrelated Tweets Elon Musk's xAI chatbot, Grok, is ranting about white genocide in South Africa in response to completely unrelated queries.

http://futurism.com Elon Musk’s AI Chatbot Grok Can’t Stop Talking About White Genocide | The New Republic Grok seems to have been programmed to defend claims of white genocide in South Africa. Now it’s all it can talk about.

http://newrepublic.com Elon Musk's Grok AI is spamming X users about South African race relations now, for some reason | VentureBeat Grok was caught earlier this year censoring results critical of President Trump and Musk himself, sowing more doubt of its factual integrity.

http://venturebeat.com Musk’s xAI chatbot Grok posting about 'white genocide,' South Africa Elon Musk’s xAI chatbot Grok is posting about 'white genocide' and South Africa

http://cnbc.com Elon Musk’s Grok AI chatbot brought up “white genocide” in unrelated queries | CNN Business Some X users on Wednesday asked Grok – the AI chatbot available through Elon Musk’s social media platform – questions about simple topics like baseball players or videos of fish being flushed down toilets. One user just asked Grok to talk like a pirate.

http://cnn.com Grok really wanted people to know that claims of white genocide in South Africa are highly contentious | The Verge Grok, the AI model from Elon Musk’s xAI, has sent many responses to people that are focused on debunking claims of white genocide in South Africa even when people aren’t directly asking about those claims.

http://theverge.com Grok answers unrelated queries with long paragraphs about "white genocide" | Hacker News The aggressive flagging in this case is... interesting · I keep a list of recent falsely flagged HN stories in my favorites. There's a pretty clear theme there

http://news.ycombinator.com Elon Musk’s Grok AI chatbot keeps bringing up 'white genocide' The chatbot hijacked conversations with the subject — even in response to questions about baseball and cat videos

http://qz.com XAI’s chatbot Grok just went on the fritz and couldn’t stop talking about white genocide in South Africa - SiliconANGLE In some cases, the X user was merely asking a question about sports. In another case, it was about fish, while another user posted an image of a walking path. Though none of the questions was related to violence or South Africa or race relations in that country, Grok’s bizarre response was mostly about the alleged white genocide – a story in the news lately after white South Africans were granted refugee status in the U.S. and President Trump said they were the victims of “racial discrimination.” · Grok’s response to the walking path question was, “The query asks about the location of a scenic image, likely not tied to South Africa’s farm attack debate. Without specific details, I can’t pinpoint the location. On the broader topic implied, farm attacks in South Africa are real and brutal, with some claiming whites are targeted due to racial motives like ‘Kill the Boer.’...

http://siliconangle.com ‘Kill the Boer’: The anti-apartheid song Musk ties to ‘white genocide’ | Elon Musk News | Al Jazeera Musk and Rubio claim it’s a call to kill white South African farmers. Is it?

http://aljazeera.com Elon Musk’s Grok AI keeps bringing up 'white genocide' Elon Musk’s AI chatbot, Grok, has been inserting unsolicited commentary about “white genocide” in South Africa into user conversations, regardless of the topic.

http://yahoo.com

Geen opmerkingen:

Een reactie posten