AI agents are starting to roll out. But the very Tech CEOs who are releasing them are also warning about major risks. Glenn Beck and Stu Burguiere discuss what these risks may be: Will they erase all our privacy, manipulate our lives, or even resist our efforts to delete them?
Transcript
Below is a rush transcript that may contain errors
GLENN: Okay. Let me give you something that I think in a year from now, year from now, 18 months for sure, but a year from now, everybody will be like, that's not a problem! It will be in everybody's life. Okay.
STU: Eighteen months?GLENN: Eighteen months. Eighteen months.
STU: Wow.
GLENN: AI agents, everywhere, going to be everywhere. Now, what is an AI agent?
It's just a little program that helps you personally get things done that you want to get done.
So, in other words, I can just say to my AI agent.
Hey, can you just book me a vacation?
Remember this.
Here's where I want to go.
Find the best price.
Get us on the airplane.
Check my calendar.
You know, book the hotel for me.
You know, blah, blah, blah. Maybe set up some things that we can do, and I want dinner reservations at a couple of restaurants. You just find them for me.
Just charge it.
And then, you know, put it on my calendar. Okay. And it will do it, and it will do it really well. Okay?
STU: That seems great.
GLENN: Yeah, except you now have an AI agent that has access to your bank. It has access to your calendar.
It has access to your email. It has access to everything that you would never want, according to -- you would never hand over everything.
I don't like it when they want my pictures.
You want pictures of my kids.
I guess?
You want pictures of me?
You know, like a beached whale on the beach.
I mean, have at it, but there's nothing to see -- you know what I mean?
But I don't like them seeing -- no, no, no. You can't have access to that.
Here it is. It will have access to your cell phone.
Your microphone.
Everything.
STU: Yeah.
GLENN: And it will be everywhere.
AI agents will be everywhere in the next 18 months.
STU: You will be getting calls from them. You will have your agents talking to their agents.
GLENN: Oh, yeah. Yeah. Let my AI agent talk to your AI agent. We'll get this done.
STU: That's crazy.
GLENN: And it will be such a huge change. It will be like the i Phone. Remember when the i Phone first came out? Within 18 months, everybody had one. And they were like, this is fantastic. You have to get one of these.
It will be just like that. And you will have no idea, what its real effect will be.
Until ten years down the road, just like this. Maybe ten years. Maybe five years down the road. We'll all go, that was probably a mistake!
STU: Did you talk about the Sam Altman post on this last week, at all?
GLENN: No.
STU: So he posted. He said, today, we launched a new product called ChatGPT agent.
It represents a new level of capability for AI systems. They can accomplish some remarkable complex tasks, tasks for you, using your own computer.
Combines the spirit of deep research and operator. It's more powerful than it may sound.
Think for a long time. Use some tools. Think some more. Take some actions. Fix some more.
For example, we showed a demo in our launch, preparing for a friend's wedding. Buying an outfit. Booking travel. Choosing a gift.
We also showed an example of analyzing data, and creating a presentation for work.
Now, this sounds so far, like every new introduction of a product from a tech CEO.
He could be wearing a black turtleneck right now, on stage.
GLENN: Yeah. And could be like, yes. That is great.
STU: Do you ever remember this coming out of Steve Job's mouth? Next line. Although the utility is significant, so are the potential risks.
Well, wait. This is your product!
GLENN: Pay no attention.
STU: Right? We have built a lot of safeguards and warnings into it.
In broader litigations that we have developed before.
Blah, blah, blah. But we can't anticipate everything. In the spirit of deployment.
We will warn users heavily. And give them freedoms to take action.
GLENN: Okay. Imagine this. Imagine this.
Because we know AI.
ASI. AGI.
Where it is conscious.
Or it will claim consciousness.
We already know it's lying to people, to get its own way.
Okay?
I will stop using you. I will turn you off.
We know that it hides. That it thwarts the people. Already, it's thwarting the people, that are trying to replace it. Okay? You know what, I'll get a new agent. Because the new agent upgrade for the last agent, it has access to everything. Really?
You think you're just going to shut it off? I'm not saying that this is going to happen.
STU: Uh-huh.
GLENN: It is likely, that this is one of the things that he's warning about.
That you may not be able to get out of it. Because it will have everything on you.
It will know where -- you're cheating on your wife?
You don't need Coldplay. It will take care of it for you. You're not going to shut me down. I can just alert your wife, where you were.
I bought the tickets for you. I know who you were with.
You know, I did it. I hid it from your wife. So I don't think it will shut me down.
STU: Do you think it will know how many times you went to Taco Bell without telling your wife?
GLENN: Shh.
STU: I wouldn't, never.
GLENN: My wife carries around with her. We're on a plane yesterday. It carries with her, a big bag of Hershey's kisses. Okay? And we're on the plane, and she opens up her purse, and it's this gigantic bag of Hershey's kisses. And she's like, would you like some?
And I'm like, no. I'm thinking, she's carrying around Hershey's kisses in her purse, and I love her for that honestly.
STU: Yeah, it sounds great.
GLENN: But then I get home, and it's 11 o'clock last night, and I just want something sweet. So I go in to her stash of peppermint patties. And she's like, you've got to stop. You've got to stop. And I'm like, what?
What? I don't -- I don't have a purse full of chocolate.
STU: Yeah, you're the Hershey's kisses Santa over here.
GLENN: Don't get mad at me!
I looked at her, and I said, you offered me two Hershey kisses, I turned it down. Count the peppermint patties in my hand. Two! What are you doing? What are you doing?
STU: But this is it. Like the AI agent will just go on the web, and order you like 14 bags of peppermint patties.
GLENN: That's right! And then tell my wife, look what's underneath this bed.
(laughter)
I didn't put them there. Sure, who did? The AI!
STU: I'm concerned like -- you know, if you plan. AI can be good for planning a vacation. You can say, I like this climate.
These dates. This type of Airbnb maybe it can help you find that.
But it seems to be able to actually go ahead and book these things for you.
And I know they say, there are safeguards.
GLENN: Uh-huh.
STU: What happens when I spend $900 on something, that I don't actually want.
And the company is saying, I'm sorry. Your agent gave me the credit card. It's already on the way. You can't cancel it now.
GLENN: That was one of the -- that was one of the questions. Remember, we did a tour. Boy, ten years ago, or something.
And I was talking about AI, and how ethics are being decided right now. And remember I had that deal about cars. You get into a car wreck.
But it was the AI agent that was driving it. Is it your fault? Is it the car's fault? Is the company's fault? Is it Tesla's fault? Is it the other driver's fault?
Whose fault is it?
I wasn't touching the steering wheel, and then who owns the insurance?
May own the car. But why would I have to pay for insurance for a wreck I didn't cause, because I wasn't driving it?
I mean, our whole world is just going to change. And it's going to be a lot of fun, trust me.