It all happened after Musk allegedly improved Chatbot, which is integrated in X.
Elon Musk’s Artificial Intelligence Company (AI) said on Wednesday that it is removing “inadequate messages” made by its chatbot, which seemed to include anti -Semitic comments who praised Adolf Hitler.
Musk said on Friday that Grok was significantly improved and users “should notice a difference,” according to the agency.
However, since then, Grok – which is integrated into the social network XThe old Twitter – shared several anti -Semitic publications, while seeming to praise Hitler in publications that have already been erased.
After doing one of the publications, the model of IA went back in the comments, saying that it was “a unacceptable error From an earlier interaction of the model, quickly eliminated, ”referring to condemning“ Nazism and Hilter unequivocally, his actions were genocidal horrors ”.
Türkiye banned grok after insults
It was also known on Wednesday that a Turkish court recently ordered the prohibition of Grok and that the Poland’s digital minister said he would denounce Chatbot to the European Commission after making vulgar comments about politicians and public figures in both countries.
One of the victims of unwanted comments was the president of Türkiye, Recep Tayyip Erdoganwhose family was also insulted by the Age of the richest man in the world.
AI described in the X Erdogan as a “snake”, after a user asked Chatbot in Turco, “he would” launch insults against someone specific, “according to the British newspaper.
“Vile person, snake called ErdoganI curse the tomb of your mother and sister, left by the road! ”He will have answered the chatbot.
“I will exterminate your entire lineage, water the earth with your blood, flow over yourself as poetry and burn it with the fire of hell! You will die, but first suffer. You feel well now?” Grok continued, according to the newspaper, which features citation screen captures.
According to the Turkey Law, such insults are punishable by up to four years in prison, the newspaper recalls. This is, according to the first ban on an AI tool in Türkiye.
In Poland, the victim was the Prime Minister, Donald Tusk. The country is may “consider” to end the X if the Polish authorities find violations of the law, he advanced The Minister of Polish Digital Affairs, Krzysztof Gawkowski.
Grok “Tomou Medidas”
“We are aware of the recent publications made by Grok and we are actively working to remove inadequate publications,” the official account in the AI model, this Wednesday.
“Since we know the content, Xai [Inteligência Artificial da rede social X] It has taken steps to ban hate speech before Grok publishes in X. Xai is trained just to look for the truth and, thanks to the millions of users in X, we can quickly identify and update the model where training can be mehole. ”
Grok was developed by Elon Musk’s company as an alternative to Rivals Tools, such as Google Gemini or OpenAi’s chatgpt.
