INDIANAPOLIS — Imagine getting a frantic phone call from a loved one asking for money. Their voice sounds familiar, and they seem terrified. But the person on the other end of the line may not be who you think it is. It could be an AI-generated voice designed to scam you.
As AI tools become more sophisticated, scammers are finding new ways to impersonate people and trick victims into sending money or sharing sensitive information.
AI expert Tristen O’Brien shared safety tips to help Indianapolis families recognize potential scams and use artificial intelligence responsibly.
“We’re starting to hear a lot more stories nowadays where AI is really a double-edged sword,” O’Brien said. “These bad actors can do things that are trying to trick us into doing something we typically wouldn’t do.”
O’Brien outlined three key areas to consider when using AI: what to trust, what to share and when to take control.
The risk: AI can convincingly imitate voices, create fake photos and generate fraudulent messages, making it harder to distinguish between real and fabricated content.
The solution: Verify unexpected requests, especially those involving money, by contacting the person through a trusted phone number or another established communication channel.
“If somebody calls you and says, ‘Hey, I need you to send some money,’ or ‘It’s really urgent,’ even though you’re in the moment, you might need to say, ‘I’m going to hang up. I’m going to call you back on a trusted number,’” O’Brien said.
He also recommends that families establish a secret passphrase to verify someone’s identity during an emergency. The approach is similar to codes parents once used to confirm that someone picking up their children was trustworthy.
“If they called me up, if it was their voice, I would feel panic,” O’Brien said. “But I may need to say, ‘What’s the code phrase?’”
The risk: AI tools such as ChatGPT and Gemini are increasingly used as digital assistants, but users may not fully consider the privacy implications of sharing personal information.
The solution: Avoid entering passwords, confidential work information or sensitive personal details into AI tools that have not been approved for that purpose.
“If you’re not gonna post it online, don’t put it into ChatGPT,” O’Brien said. “AI is going to a business. That business is using your data to help with their algorithm, and maybe you don’t want them to have that type of information.”
Users should review an AI platform’s privacy policies and data settings before sharing information, particularly when using the tools for work.
The risk: AI tools can draft emails, create spreadsheets and perform tasks automatically. Depending on the tool and its permissions, they may also send messages or make changes without users realizing what will happen.
The solution: Review AI-generated work and confirm what an application is authorized to do before allowing it to take action.
O’Brien shared an example of asking an AI tool to write an email, only to discover that it sent the message instead of saving it as a draft.
“I said, ‘Hey, create this email for me.’ Well, it sent the email for me. I didn’t say put it as a draft,” he said.
O’Brien also addressed the growing use of AI in education. For parents whose children use AI for schoolwork, he recommends having them explain AI-generated answers in their own words to demonstrate their understanding of the material.
“There still needs to be that human in the mix for almost everything,” O’Brien said.
O’Brien said people do not need to fear AI, but they should establish boundaries and understand its limitations.
“You don’t need to be afraid of AI. You need a few boundaries,” he said.
He recommends giving AI tools clear instructions about the tasks they should perform and specifying which actions require human approval.
“The more specific you can be on the task that you’re giving it, the better,” O’Brien said.
For more AI safety tips and to watch the full interview with Tristen O’Brien, visit youtube.com/@Tristen_OBrien.
