AI vs Social Media. Which do we Trust More?

50 percent. That's how much of the global population now trusts generative AI when searching for information about companies and brands. 

More than Instagram. More than Facebook. More than social media in general. 

And that number climbed 7 points in a single year. 

The data comes from the Page Society's annual Harris Poll study. 14 countries, more than 15,000 people surveyed in December 2025. Consumers aren't just using AI more. They're believing it more. And most of them aren't clicking through to the source. They're reading the AI answer and moving on. 

For communicators, this changes the job. It's no longer enough to optimize for search engines, you now need to optimize for the AI that sits on top of them. What does generative AI say when someone types in your brand name? Is it accurate? Is it current? Or is it surfacing something you said or did 20 years ago that no longer represents who you are? 

Listen For

3:26 How is declining consumer trust reshaping the role of communication leaders?
4:52 Why do consumers now trust generative AI more than social media for brand information?
12:28 What are the biggest risks of disinformation and AI hallucinations for brands?
15:45 Why can’t AI replace human experience and critical thinking in communications?
20:22 How should brands optimize for AI search while maintaining trust and accuracy?

Guest: Rochelle Ford, Ph.D., APR, CEO The Page Society

LinkedIn | Website 

Page-Harris Poll: Confidence in Business Index 2026

Doug

Substack | Website | LinkedIn

Farzana

Substack | Website | LinkedIn

Are you a brand with a podcast that needs support? Book a meeting with Doug Downs to talk about it.

Apply to be a guest on the podcast

Connect with us

LinkedIn | X | Instagram | You Tube | Facebook | Threads | Bluesky | Pinterest

Stories and Strategies is the Official Podcast Sponsor of IABC World Conference in Toronto June 14-16, 2026

Click here to check it out https://wc.iabc.com 

Transcript

Lady Emily (00:01):
Every generation swears it will never trust the machine. And every time they're wrong.

Farzana Baduel (00:10):
In the 1880s, newspapers, they started to print photographs, but readers, they didn't trust them. They trusted illustrations drawn by a professional interpreted by a human hand. A photograph, it felt off. It felt cold, mechanical, unmediated. And within a generation, nobody trusted the illustration anymore. Now, for most of the 20th century, families trusted the stranger in a suit who knocked on their door selling encyclopedias. Then they trusted a CD ROM. Then Wikipedia. Each time the experts said, "People are fools." Maybe they were, maybe they weren't, but they were efficient. They moved their trust to whatever made knowledge easier to reach. Now, in 1989, a company called Fair Isaac introduced the credit score. Before that, a banker looked you in the eye and decided if you were good for the money. People were outraged, a number instead of a handshake. Nobody understood how it worked. Now it determines whether you get to buy that house or not.

(01:23):
 In 1998, Google put a search bar on a blank white page. Librarians were horrified. You would trust a machine to tell you what's true. Most people never scrolled past the first result. Many never left page one. Google even had a button for it. I'm Feeling Lucky. Now, that is a pattern. It has never once reversed. You can argue every one of these shifts was a mistake. Maybe the librarian was better than the algorithm. Maybe the banker’s eye was sharper than the credit score, but that's not really the point. This is not a story about what people should do. It's a story about what people actually do and why we do it. Today on Stories and Strategies, AI just became the most trusted source your audience uses to learn about your brand. The question is, what is it telling them? My name is Farzana Baduel.

Doug Downs (02:38):
My name is Doug Downs. Our guest this week is Rochelle Ford joining today from Elon, North Carolina. Hi, Rochelle.

Rochelle Ford (02:45):
Hello, Doug. How are you?

Doug Downs (02:47):
I'm good. I'm good. I got to ask you where Elon is at, and I'm sure you get asked all the time. Did you name the city after them? What are you doing

Rochelle Ford (02:57):
there? Absolutely not. Elon is here in North Carolina. We are in between Greensboro and the Raleigh Durham area. And it's actually the university’s home here is one of America's top 100 known for undergraduate teaching and learning. And I used to be the dean here. So a lovely home, nice bedroom community. Great place to eat. Should be

Doug Downs (03:20):
coming up at the top of the search, the top of the search, not secondary at all.

Rochelle Ford (03:25):
That should be the top.

Doug Downs (03:26):
Exactly. Rochelle, you are the CEO of the Page Society, former university president, as you were saying, and a PRWeek Hall of Fame inductee who has spent a career training the profession’s next generation of leaders. You now lead the organization that tells the world’s biggest companies what trust looks like. And this year’s data says it looks very different than it did 12 months ago.

Rochelle Ford (03:51):
Well, yeah, it has. The confidence that consumers have in business continues to decline. They want to have confidence that we can make a difference on issues that are important to us. And unfortunately, we are continuing to disappoint consumers. And the question is, how can professional communication leaders help businesses to restore trust and confidence that they can make a big difference in society?

Farzana Baduel (04:23):
Now, one thing that we discussed earlier is the incredible insight that comes in from the Harris Poll. So Page Society, very well known for its insight. And one of the findings I thought was really surprising, and I wanted to ask you what you thought about how the trust in generative AI now surpasses trust in social media. When people are searching for information about brands, tell us about that because that was completely new to me.

Rochelle Ford (04:52):
Yeah, that was a surprise to so many in our survey this year that when you look at where consumers go for information, you didn't used to think about generative AI, but when you look at the fastest growing source of trusted information, people are going to generative AI, things where you actually have to go search for the information. And now generative AI search engines, about 50 percent of consumers across markets, talking 14 major markets, 50 percent trust AI versus only 49 percent in general trust social media. Facebook, 44 percent, Instagram, 45 percent. So there is a shift going on. So something, if it's not in people's media mix, then you need to change it up.

Farzana Baduel (11:28):
Yeah. I think probably these LLMs aren't going to have that sort of free terrain that the social media companies had for quite a long time because people are much more aware of regulation and the impact of unregulated communication channels. So I think it's going to be really interesting to see what happens with these LLMs. The other thing I wanted to ask you is you mentioned that it's important for PRs to be thinking about the presence of their organizational brands on these LLMs and to then make sure that they're accurate. But what about disinformation? What about bad actors who are thinking that actually a lot of searchers moved on to these LLMs and actually we're going to figure out what sources they actually draw from, and then we're going to fuel our disinformation through those channels. Do you see concern around the disinformation aspect targeting these LLMs as well?

Rochelle Ford (12:28):
Absolutely. Disinformation is one of the biggest concerns that's on chief communication officers’ minds and the agencies that support them because first, these LLMs also hallucinate and they make up information and sources that don't exist at times. And so bots talking to bots is a concern. So we cannot take the human element out of it. What are the detection tools that we can find out that if something is misinformation and where is that source? Where is it coming from? Is it a hallucination? Is it a malicious actor? And there are a lot of agencies and technology firms that are helping to detect that sort of thing, but you also have to have discernment to know if something is a bot storm that is not even human driven versus something that is a malicious actor that's spreading information and being able to handle crisis situations in a very proactive way.

(13:30):
 And that's a new tool in people's crisis communications belt that you have to be able to handle. I'll

Doug Downs (13:37):
give you an example of hallucination and trust in AI. Somebody sent me an RFP last week and I, sorry, I'm not crazy about RFPs. I never like

Farzana Baduel (13:46):
I don't think anyone

Doug Downs (13:47):
is. No, but I was a sub to it, but I did give it a human scan. Then I did what I think a lot of us would do is I dragged it into Claude, and I had a conversation with Claude and what was in there. And a very key piece of feedback, which I was building to go into the RFP, was a complete and utter hallucination by Claude. And if I had not humanly spent time looking at that, it would have gone through. And when I asked Claude, "What are you doing?" it said, "Oh, you're right. I invented that." But that's still trust in AI, Rochelle. And we're still going to do it because heuristics tell us, not that we're lazy, but we'll cut corners.

Rochelle Ford (14:36):
Well, you cannot take the human out of it. One of the things that Page has been doing is a partnership with Scriptorium. It's a think tank out of Europe and founded by actually two of our Page members. And we're looking at what are those risks that are involved with having agentic teams and how do you also communicate with agentic stakeholder groups. And you cannot take the human element out of that. We did an experiment at our spring seminar that just happened in New York City back in March and we created an agentic team and we're like, "How would you handle this case?" And it was interesting. Sometimes our agentic agents worked with us, sometimes they provided really bad information, but if we were not experienced comms leaders with information from that sector, they could take us off in a really bad direction. So experience remains critical for a chief communications officer or in understanding that sector.

(15:45):
 So you cannot substitute AI for experience because your experience, Doug, knew, "You're doing something." But someone who is inexperienced and has no background knowledge would not know to correct or to question. And so there is an obligation for us as comms leaders to make sure that we reinforce the critical thinking skills, that we reinforce the experience that brings to the table so that we can detect, we can find information that might be misleading and wrong and knowing how to fact check. So I'm coming from higher ed before I came into this role. My God, what is totally important now for these future generations is teaching those skills, those fact checking skills, making sure that there's some critical reasoning skills that go on. And we've got to push that all the way down through K through 12 to get the youngest students involved and having those critical thinking skills.

Farzana Baduel (16:44):
And Rochelle, you're in this incredible vantage point with the Page Society because your members are the CCOs, the chief communication officers of some of the largest, most influential companies in the world. What I wanted to ask you is what traits and skill sets do you think that PRs need to really start thinking about developing if they want to one day be like these CCOs who are members of the Page Society, bearing in mind how things are moving so fast?

Rochelle Ford (17:20):
Well, you have to continue to have curiosity, one. Folks who are not curious or are not willing to experiment and to be able to lean into change, that curiosity has to remain. But the soft skills, those critical thinking skills are the utmost important that you are taking a skeptical look at information in order to solve problems. Problem solving skills still become critical. Some of the basic things can be taught, grammar and all of that. AI can help with your spelling, et cetera, but is it going to really reach an audience of things that are those conversations that are held offline? Because what happens in the meeting and the conversations that are happening, that nuanced thing and connecting things that were not recorded, things that were said in the hallway, what have you, bringing that information, things that you observe in the real world have to be brought to the table.

(18:17):
 So you have to be able to take all those senses that you are experiencing and bring it to problem solving and that you're not relying upon it. And so it sounds like very basic, critical thinking, problem solving, curiosity, but that person who's able to connect dots will be so much better. The person who's able to make those probes and to have that careful analysis, even though AI will help expedite and make things more efficient, it's the human that actually makes the decision. It's the human that is going to really understand what that gut check is and has to make the choice.

Farzana Baduel (19:00):
Now that's a positive way to end this conversation.

Doug Downs (19:03):
Love it. Yeah.

Farzana Baduel (19:04):
Humans are still needed, still relevant and

Doug Downs (19:07):
for now.

Farzana Baduel (19:11):
And judgment as well is so incredibly important because what you find with all these AI tools is not to outsource your thinking, your critical thinking. You need to have judgment in terms of what you put into the AI tool and you need to have judgment in terms of how you edit what comes out of the AI tool. And so I think that is still safely with us.

Rochelle Ford (19:33):
Well, I think it's important. And just think it was an AI that brought us together when we met in the UAE not too long ago. It was human, real interaction, and it builds relationships. Real human interaction builds lasting relationships. Bots can talk to bots, but humans build relationships. Relationships is what makes up the public relations industry. And so we have to keep the humans in the loop to make the decisions, to build the relationships, because that's where trust is going to happen. Got to prove it with action. And Doug, we're going to continue to tell the truth and tell stories with context, and that'll restore confidence.

Doug Downs (20:15):
Awesome. Thanks so much for this, Rochelle.

Rochelle Ford (20:17):
Thank you. Thank you.

Doug Downs (20:22):
Here are the top three things we got today from Rochelle Ford, who's excellent, by the way. My goodness. Number one, optimize for AI search, GEO, or I use the term AEO. Doesn't matter. Just like SEO, brands now need their content discoverable and accurate across generative AI engines. Number two, trust varies by market. That was interesting. AI trust levels differ dramatically by country, so global brands have to tailor their media mix accordingly. And number three, never outsource your judgment. AI speeds things up, but hallucinations are real. Human experience and critical thinking remain irreplaceable, and hopefully the humans don't hallucinate.

Farzana Baduel (21:07):
Absolutely. Now, if you would like to send a message to our guest, Rochelle Ford, we've got her contact information in the show notes. Now, Stories and Strategies is a co production of Curzon Public Relations and Stories and Strategies podcasts. If you liked the episode, please do leave a rating and possibly a review, and a big, big thank you to our producers, Emily Page and David Olajide. Lastly, do us a favor, forward this episode to one friend, and thank you so much for listening.