Local experts say they are seeing an increase in the use of artificial intelligence, especially on social media. Doctored images, videos, headlines and articles have taken over the platforms. Experts say about 28% of what we see is entirely manipulated or created by AI. “Especially in an election year, it’s really important to understand that not everything you see online is true,” Amy Konwinski said. “So it can be very dangerous.” Konwinski, founder and CEO of Smart Gen Society, said it’s important to be sober about the information you see online, especially in days like Saturday after the assassination attempt on former President Donald Trump. “Right away, we saw things floating around on social media that weren’t exactly verified or true,” Konwinski said. Dr. Joel Elson of the National Center for Counterterrorism Innovation, Technology and Education at the University of Nebraska at Omaha said AI is being used in all forms of technology, whether someone realizes it or not. And while AI has its benefits, it can also be exploited for someone’s specific purposes. “It’s clear that we’re starting to see individuals and groups using AI tools to do social harm,” Elson said. But both Elson and Konwinski said there are ways to determine whether content is legitimate, and it starts with identifying the source. “Be skeptical of everything,” Elson said. “Take a moment, stop and think, ‘Where is this coming from?'” “Most of the time, it’s a legitimate news source, and it’s featured on two or three platforms at the same time,” Konwinski said. Other clues include small changes to website URLs, or headlines or articles that are too perfect. “For example, sometimes you’ll see a news source like NBC.com, which is a trusted news source, but sometimes you’ll see NBC.com.co, which could be a fake news source,” Konwinski said. “So when you’re reading something, it doesn’t sound like a real conversation, so that’s how you look for errors.” When it comes to images and videos, both said there are clues that AI is involved. They are often found in small details, or blurred edges of photos, blurred backgrounds, or excessive pixelation when zooming in on details. “When you look at images of people, it seems like AI hasn’t figured out how to draw pores or wrinkles yet,” Konwinski said. “Look at fingers and toes. AI doesn’t really know how to draw these things well.” Elson said many content creators are starting to fight back against AI by implementing things like watermarking images. You can also use tools like the website Hive Moderation to reverse search images, videos, and articles. This will help you see what might have been generated by AI. Elson also said one of the best ways to better detect AI is to educate yourself about its capabilities. “All the big tech companies offer free tools that you can experiment with, play around with, and see for yourself what kind of content you can generate,” Elson said. Overall, Elson and Konwinski said to be thoughtful. Make sure what you’re sharing is true, necessary, and comes from a verified source. For the latest headlines from KETV NewsWatch 7, click here
Local experts say they are seeing an increase in the use of artificial intelligence, especially on social media.
Doctored images, videos, headlines, and articles are flooding the platforms, and experts say that roughly 28% of what you see has been manipulated or created entirely by AI.
“Especially in an election year, it’s really important to understand that not everything you see online is true,” Amy Konwinski said, “so it can be very dangerous.”
Konwinski, founder and CEO of Smart Gen Society, said it’s important to stay calm about what you see online, especially on a day like Saturday, which marked the day of the assassination attempt on former President Donald Trump.
“Immediately, I noticed there was information circulating on social media that wasn’t necessarily verified or true,” Konwinski said.
Dr. Joel Elson of the National Counterterrorism Innovation, Technology and Education Center at the University of Nebraska at Omaha says AI is being used in all forms of technology, whether we realize it or not, and while it has benefits, it can also be used to further someone’s specific goals.
“We are certainly beginning to see individuals and groups using AI tools to inflict societal harm,” Elson said.
But both Elson and Konwinski said there are ways to determine whether content is legitimate, and it starts with identifying the source.
“Be skeptical of everything,” Elson says. “Take a moment to stop and think, ‘Where is this coming from?'”
“Most of the time, it’s a legitimate news source and it’s going to be reported on two or three different platforms at once,” Konwinski said.
Other clues include small changes to a website’s URL or a headline or article that’s written too perfectly.
“For example, you might see a news source like NBC.com, which is a trusted news source, but you might also see NBC.com.co, which could be a fake news source,” Konwinski said. “When you’re reading something, it doesn’t sound like a real conversation, so it’s important to look for errors in that regard.”
When it comes to images and videos, the companies say there are clues that AI is at work, often in the details — blurred edges of a photo, blurred backgrounds, or excessive pixelation when you zoom in on small details.
“When you look at images of people, you can see that the AI hasn’t yet figured out how to draw pores or wrinkles,” Konwinski said. “When you look at fingers and toes, you can see that the AI doesn’t really know how to draw those things well.”
Elson said many content creators are starting to combat AI by implementing things like watermarking images.
You can also reverse search images, videos, and articles using tools like the website Hive Moderation, which can help you see if something was generated by an AI. Elson also said that one of the best ways to better detect AI is to educate yourself on how it works.
“All the big tech companies offer free tools that allow you to experiment, play around and see for yourself what type of content you can generate,” Elson said.
Overall, Elson and Konwinski say to be thoughtful: Make sure what you’re sharing is true, necessary, and comes from a verified source.
For the latest headlines from KETV NewsWatch 7, click here
