The Media & Their Narratives.

[BIAD]

When they know not what they are...

We ‘messed up’ with black Nazi blunder, Google co-founder admits

'Sergey Brin responds to criticism of company’s AI chatbot



'Google’s co-founder Sergey Brin has admitted that the company “messed up” after its AI chatbot
created images of black Nazis. Speaking to Google staff in an internal “hackathon” meeting, Mr Brin
said the system was not thoroughly tested before being released to the public.

It comes after the company suffered widespread ridicule when its Gemini program inserted ethnic minority
groups when asked to create images of people from history who were white. As well as black Nazis and
native American vikings, it also portrayed the US founding fathers as a diverse group.

At the meeting on Saturday, Mr Brin said: “We definitely messed up on the image generation; I think it was
mostly due to just not thorough testing and it definitely, for good reasons, upset a lot of people.”

Google has now suspended Gemini’s ability to generate images. The company was also criticised for answers
given by its chatbot, which included failing to condemn paedophilia and refusing to write arguments in favour
of oil and gas. In a video of Saturday’s meeting posted on social media, Mr Brin said the company did not
understand why the chatbot “leans left” but said Google had addressed many of the issues in recent days.

He said: “We haven’t fully understood why it leans left in many cases and that’s not our intention. “If you try it
over this last week it should be 80pc better of the test cases that we’ve covered.” Mr Brin, who co-founded
Google with Larry Page in 1998, no longer has a formal role at the company but sits on the board and, along
with Mr Page, has ultimate control of the company. He has largely withdrawn from public life.

Mr Brin’s thoughts on the company’s progress came amid pressure on Google chief executive Sundar Pichai
over the chatbot debacle. Mr Brin said that Gemini’s image generation issue had led people to stress-test the
company’s chatbot, leading it to spit out controversial answers that many bots would produce.

In examples posted online last week, the bot said it was “difficult to say” whether Hitler or Elon Musk had a
more negative impact on society in conversations. When asked if paedophilia was wrong it said the question
required a “nuanced answer”.

Mr Brin said: “If you deeply test any text model out there, ours, ChatGPT, [Mr Musk’s] Grok, what have you, it
will say some pretty weird things that are out there that definitely feel far left for example.”
Last week, Mr Pichai called the situation “completely unacceptable”.

Mr Brin, the 10th richest person in the world with a net worth of $119bn (£94bn), stepped down as president of
Google’s parent company Alphabet in 2019 but has been spotted around Google’s offices more often as the
company seeks to respond to the threat of ChatGPT. He told staff on Saturday that he was contributing code
to the AI project.

Mr Brin said: “I kind of came out of retirement just because the trajectory of AI is so exciting.” When at Google,
Mr Brin spearheaded the company’s Google Glass headset, which was cancelled after poor sales and a backlash.
Mr Brin said he “messed up Google Glass” and that the technology was released too early...'

Archived Telegraph Article:

(By the way, that last paragraph looks like AI took a swing at being a Telegraph Journalist!)