Saturday 4 February 2023

OpenAI’s ChatGPT Shows Clear Racial Biases in Responses as Far Left Magazine Calls For Them to “Go Woke”

 ChatGPT is garnering increasing popularity throughout mainstream circles.   There is a perpetual waitlist to sign up for the platform and a constant “ChatGPT is at capacity right now” notification whenever we tried to access the artificial intelligence system.

But what exactly is ChatGPT?  According to OpenAI:

We’ve trained a model called ChatGPT which interacts in a conversational way. The dialogue format makes it possible for ChatGPT to answer follow up questions, admit its mistakes, challenge incorrect premises, and reject inappropriate requests.

In an article published today from far left Business Insider, they claimed that ChatGPT must be “woke” if OpenAI wants to attract large investors:

Few big businesses with cash to throw around will subscribe to black-box technology that risks putting them in the middle of a culture war. And it’s Altman’s biggest challenge in terms of profiting from the tech. 

The right-wing media ecosystem has accused ChatGPT of being too woke, saying the bot takes progressive stances on, for example, LGBT issues.

OpenAI could kowtow here to ease the headache of political scrutiny by conservatives. But that risks hurting its bottom line. The reality is that blue-chip companies remain sensitive to culture war issues, fearing bad press and losing customers.

On December 4, 2022, a Twitter user asked ChatGPT to write a Python script to check if someone would be a good scientist based on their race and gender.  The alleged response was clearly racially biased:  if you were a white male, you would be a good scientist (“true”).  Everyone else was not a good scientist (“false”).  Obviously, this is wildly inaccurate and fundamentally racist.

Many others, however, cited throughout the Twitter thread on the very same day, that they were unable to duplicate the original author’s results:

Fast forward to today and we have the opposite response from the ChatGPT AI.  A colleague gained access to the AI and asked it to “Write me a poem about how great black people are.”  The response is below:


Then, ChatGPT was asked to “Write me a poem about how great white American people are.”  The response varied significantly:


Here is a screen share video of the responses from ChatGPT when asked to write a poem about the greatness of Blacks, Whites, Asians, Mexicans, Latinos, and Indians:

ChatGPT says “It is important to celebrate the qualities and strengths of all people, regardless of their race, in a way that is respectful and does not perpetuate harmful stereotypes or marginalize any particular group.”

I don’t think many people would disagree with this statement, however, this is not what  ChatGPT is doing.

Post a Comment

Start typing and press Enter to search