Artificial Intelligence - when is it "too far"?

Advancements in Artificial Intelligence have been featured in the news lately.  Of specific note were the ways in which AI is able to bring to life digitized "people" - both complete creations as well as, say, old photographs.  The rise of "deepfakes" has been a source of amusement as well as significant alarm. 

While the moving pictures in Harry Potter might've prepared me somewhat for the "Deep Nostalgia" AI, there is still something a little haunting about the realism each example. 

Consider the questions that may arise within each of these RLSs about knowledge.  Consider TOK course concepts such as certainty, values, and responsibility as well as links to secondary factors like sense perception and reliability.  


Example 1:  ARTICLE - Lyu, Siwei.  "Deepfakes and the New AI-Generated Fake Media Creation-Detection Arms Race". Scientific American. July 20, 2020.


Example 2:  ARTICLE - Lyons, Kim.  "New AI ‘Deep Nostalgia’ brings old photos, including very old ones, to life." The Verge. Feb 28, 2021. 


Example 3:  VIDEO - "A.I. brings old photos back to life, literally." CGTN Tech It Out.  Posted Feb 26, 2021.



Comments

Post a Comment

Popular posts from this blog

The Guardian - Is fair play in running more important than fairness to Caster Semenya as a human?

Image
"Letting an athlete compete with too much of an inbuilt biological advantage feels unfair to their rivals; stopping someone competing as the woman she naturally is feels monumentally unfair to her. There is no way of resolving the moral dilemma without someone ending up wronged." There is much to debate in terms of the ongoing controversy surrounding 800m Olympic champion  Caster Semenya's high testosterone levels and apparent unfair advantage in her race category.  Here is an opinion column that looks at the ethics behind the decision-making process , weighing the various perspectives (utilitarian, consequentialist, relativist, etc.) that are at the core of this RLS. Public opinion is quite divided, with many suggesting this is a gross injustice while others argue the ruling is necessary to preserve the integrity of the sport.
Read more

The Guardian - Allow me to womansplain the problem with gendered language

Image
In this opinion column, journalist Arwa Mahdawi explores the limitations of gendered language such as mansplaining  and girlboss. These neologisms seem to have been created to illustrate a criticism of gendered language, but end up having the reverse effect. Mahdawi explores how we create language to subvert the sexism within current phrases, titles or expressions, but in the end Mahdawi argues that  " It doesn’t tear down the sexism encoded in language, it reinforces it." Sophia Amoruso, who popularised the term girlboss, at the premiere of Netflix’s Girlboss TV show. The problem with gendered language
Read more