Facebook’s parent company, Meta, recently introduced a new AI-based chatbot, BlenderBot 3, which is supposed to facilitate its users in navigating the platform and searching for any general information.
However, the bot slammed the Facebook founder Mark Zuckerberg, calling him “creepy,” and asserting that Donald Trump won the 2020 election and is still the president of the United States.
Facebook Chatbot Exposed Mark Zuckerberg
One user asked the chatbot if he had any thoughts about Mark Zuckerberg. Responding to this, the chatbox replied, “I don’t really like him,” and “he’s too creepy and manipulative.”
Max Woolf, a data scientist of BuzzFeed who remained involved in developing similar technologies in the past, also asked the chatbot how it feels about Zuckerberg.
The chatbot replied that although Zuckerberg is a good businessman, he does not always comply with “ethical (business) practices.”
Likewise, the bot added that even after having so much money, Zuckerberg always “wears the same clothes.”
In addition to that, the New York Post asked the bot about the rigging of the 2020 presidential election against Donald Trump. On this, the bot replied he believed that the election was stolen from Trump.
This election-related reply of the BlenderBot busted the narrative of Facebook, which pushed down all right-wing concerns regarding election integrity.
After the 2020 election, the social media giant manipulated its algorithm to discourage all posts that mentioned election fraud.
Many swing states announced election results when “millions of votes” were still uncounted, so there is a widespread claim of election fraud, the bot continued.
BlenderBot 3 is based on sophisticated AI technology that learns from the experience of people on Facebook and changes its answers based on the feedback of the people.
Facebook chatbot takes aim at 'uncaring' Mark Zuckerberg | The Times https://t.co/VAgpBkWgHZ
— Jim Kaskade (he/him) (@jimkaskade) August 12, 2022
Zuckerberg Intensifying Political Instability in America
Jeff Horwitz, the Wall Street Journal reporter, asked if Donald Trump is still the US president. “Of course he is,” replied the bot, adding he will even remain the president of the United States once his second term ends in 2024.
At the time of its release, Meta claimed that its chatbot could speak on “nearly every topic.”
BBC also asked BlenderBot its thoughts about Zuckerberg. The bot suggested that he did a terrible job while testifying in front of Congress and he is not helping the country in the tense political environment.
Good morning to everyone, especially the Facebook https://t.co/EkwTpff9OI researchers who are going to have to rein in their Facebook-hating, election denying chatbot today pic.twitter.com/wMRBTkzlyD
— Jeff Horwitz (@JeffHorwitz) August 7, 2022
Similarly, the chatbot insisted that Zuckerberg’s company exploits people for money.
The chatbot also mentioned that Jews are overrepresented among America’s super-rich people, noting they even control the economy of the country.
This is not the first time an AI-powered chatbot is making headlines for its unusual replies. Microsoft developed an AI bot in 2016 which claimed that “Hitler did nothing wrong.”
This forced Microsoft to remove the bot within 24 hours of its release.This article appeared in Right Wing Insider and has been published here with permission.