The AI Phone Assault Has Begun
I'm visiting a friend of the over 65 variety. His name isn't Ron, but that's what I'm going to call him. His mind used to be very sharp. It's dulled a lot this year. He's a naturally trusting soul and I'm worried sick about him.
Incoming
Ron told me his biggest frustration is the number of spam calls he gets every day. He doesn't answer calls from unknown numbers outside his area code. He generally picks up the local ones, though.
I saw it happen twice in an hour while we were hanging out. Ron's hard of hearing. His phone volume is up so high it might as well be on speaker. I heard both sides of the conversations.
Each times, they were AI bots.
Two Strikes
The calls sounded like they were from friendly women. The introductions were both some form of "we've noticed you have unused Medicaid benefits and want to make sure you're not missing out."
Ron ended the first call quickly. He recognized it as spam, but not a bot. He stayed on the second one longer. It started asking questions with no prelude. He just started answering them. He didn't have time to get his guard up.
I stopped him as soon as I realized he hadn't clocked the nature of the call. After telling him it was an AI bot, he asked how I knew. I couldn't offer much beyond just having experience and learning how they sound.
Medicaid is the New Car Warranty
Spam robo-calls were easy to spot when they were recorded messages telling you to press one because your car warranty had expired. Now that they interact, they're way more dangerous.
You can get an idea about why from this video demonstrating a call:
To be clear, there is no woman on the other end of that call. The voice and the words it says are completely computer generated.
The voice in the video sounds exactly like the bot that called Ron. The bot in video says it's AI in its introduction. The one that called Ron did not. I don't know what the play was, but anything that starts off by implying it has access to your Medicaid information is absolutely a scam.
Playing Defense
This shit is gonna get worse. The bots are brand new and they're already convincing. They'll get better and cheaper to deploy. The onslaught will increase.
The bots will also pull in increasing amounts of personal information about their targets. They'll use it to be even more effective at manipulation.
There's nothing realistic we can do to stop them. The best we can do is work to protect ourselves. Currently, that looks like this:
- Auto-forward every incoming call from an unknown number directly to voice mail.
- Start from the assumption that every voice mail form an unknown number is malicioius.
- Don't call the number back drirectly even if it's from a company/bank/service you use. Instead, find the companies phone number on their website and use that to contact them.
I helped Ron set up his phone. I suggest you do the same. Then, help friends and family who might not know how to do it on their own.
-a
Endnotes
Preying on the elderly is nothing new. These bots are just the latest iteration. They're going to be particularly nasty. But, senior citizens won't be the only ones targeted. We'll all have to deal with them.
The use of the "um" in the opening line in the video is particularly insidious. The computer isn't struggling to find the next word. It's an explicit deceit included to trick us into thinking we're talking to a person.
Here's the iPhone page showing how to set things up. It's done by following the instruction in the "Screen or silence unknown numbers" section and choosing "Silence: Calls from unsaved numbers will be silenced, sent to voicemail, and appear in the Recents list."
I haven't found an official support page for how to do it on Android devices. You'll need to look up your specific phone for how to do it.
AI is a giant topic. I have different thoughts about different parts of it. Some of those thoughts are complicated. When it comes to this type of shit, that's not the case. I can't think of a single improvement that'll be added to the world by deploying AI spam bots. Even if they aren't outright scams.
What I could get behind is AI being used on our side. Bots that provide protection from spam and scams.
Someone go build that.