The old saying, “seeing is believing” no longer seems to apply now that Artificial Intelligence (AI) has become commonplace in everyday living.  The Contra Costa County Library and the League of Women Voters presented a March 19 webinar titled “AI:  The Good, The Bad, and The Ugly” in order to enlighten participants as to the pros and cons behind the technological phenomenon that’s taking the world by storm.

    Dr. Nolan Higdon, a founding member of the Critical Media Literacy Conference of the Americas, a Project Censored National Judge, author, and lecturer at Diablo Valley College and at the University of California, Santa Cruz served to interview the webinar’s guest speaker Dr. David Evan Harris.  Professor Harris is a Chancellor’s Public Scholar at the University of California, Berkeley and a continuing lecturer at the Haas School of Business.

    Higdon asked, “What’s really going on with AI?  It was conceived decades ago, but differently from what we have now.”

    Harris explained the basic fundamentals in that “words are put into a ‘machine,’ and the ‘machine’ puts them together.  However, it’s an imitation of being a human that may or may not offer the truth.  Chatbots should offer the sources of where the information is coming from, but that’s not always the case.  People generally understand AI as something that can perform what a human can do.

     “Artificial Super Intelligence is one step beyond in that it surpasses human intelligence and does it better,” Harris added.  “We’re not there yet.  Scholars believe that we are still quite a few years away.  There are significant technical barriers in getting there.”

     Higdon stated that AI is a very impressive achievement and wondered, “What is the ‘good’ aspect of AI?”

    “AI has been around for quite some time,” noted Harris, while mentioning social media, Alexa, Siri, Google Maps, and speech synthesis apps.  “It’s added to our productivity.”  Unfortunately, employees have become more efficient due to their use of AI, and employers have begun to lay off the work force because of it." 

    Many customer service agents are being replaced by AI, which is why there is virtually no wait time when calling companies on the phone.  The downside to using AI for customer service is that the information given to customers may not always be correct.  Harris noted successful lawsuits that have been won by plaintiffs against AI misinformation.  

    “Is Big Tech treating humans like rats in a lab while they get AI right?” asked Higdon, referring to the “bad” aspect of Artificial Intelligence when wrong facts distribute “hallucination information” (false information).

    Harris replied that there are hundreds of reported instances where consulting agencies or lawyers have used “hallucinated information” because they trusted AI.  “Many people don’t realize this is happening.  As AI gets better and better the deepfakes will get worse and worse.”  Audio, images, or videos convincingly replace one person’s mannerisms with another’s, consequently depicting people doing or saying things they never did.   “It’s starting to play a big part in politics, and AI is getting better at it.”  In California, big AI companies need to commit to noting when information is AI by labeling the content as such thanks to the recently passed SB53 (Transparency in Frontier Artificial Intelligence Act).  “There needs to be accountability.”

    The “ugly” aspect of AI, according to Harris, has resulted in homework cheating, botched hospital procedures, incorrect medical advice, and even cases of teen suicide after taking guidance from chatbots. AI data centers are also a hot topic in that job losses are on the rise, but there will be bills passed nationwide to regulate them.  AI systems can also be used to discriminate against race, religion, gender, and more.

    While some of the tools AI offers can be useful, especially if the information is correct, Harris said, “I wouldn’t use it if my career or life depended on it.”  He also cautioned that with the coming elections the deepfakes will be plentiful in number.  Additionally, he warned that the elderly are most subject to deepfake scams and need to be especially vigilant.  He recommends using www.poynter.org (an international fact-checking network).  “Whether we like it or not, AI is here to stay.”

    For more information contact Nolan Higdon at nhigdon@ucsc.edu or David Evan Harris at deh@berkeley.edu.