Menu AI and Bias
 

Artificial Inteligence and Bias

Now that ChatGPT is available to those using the new Microsoft Edge browser I have a few interesting observations on the results that I get when "chatting" with it. I get similar, but different results using Google's Bard.

When I ask ChatGPT about things that I know are featured on this website I am quite surprised of some the results that are returned. I am particularly troubled by the apparent presentation that I am a "definitive source". I am not saying that I have made things up but my pages are mainly construction from research that I have performed and as a consequence it is less than emprical.

Part of the problem with the return of "speculative" text generated by Chatbots is due to data uploaded to public databases by those seeking to build a family history with the aim of establishing a connection to somebody famous.

Top

Algorithmic bias

Like mot things in life there will be a bias. This does necessarily mean that the bias is a negative one. The Civil Liberties Union for Europe website says:

Algorithmic bias refers to certain attributes of an algorithm that cause it to create unfair or subjective outcomes. When it does this, it unfairly favors someone or something over another person or thing.

My experinces with chatbots detected a positive bias on some of the generated text from queries that I had asked.

Links

Top

References:

  • What Do We Do About the Biases in AI? - https:// hbr.org/2019/10/ what-do-we-do-about-the-biases-in-ai - Harvard Business Review - October 25, 2019 - James Manyika, Jake Silberg, and Brittany Presten
  • What does algorithmic bias mean? - https:// www.liberties.eu/en/stories/ algorithmic-bias-17052021/43528 - The Civil Liberties Union for Europe - I stripped out the UTM parameters from the URL

Site design by Tempusfugit Web Design -