The Digital Transformation Playbook

AI Divided: How Trust & Adoption Vary Across the Globe

โ€ข Kieran Gilmurray

Share episode

The world isn't experiencing artificial intelligence uniformly, and our groundbreaking exploration of the University of Melbourne's massive global AI survey reveals surprising patterns that challenge conventional wisdom about technology adoption.

TLDR:

  • Emerging economies like India and Nigeria report 92% AI adoption rates, while advanced economies lag behind
  • About half of people in emerging economies have received AI training compared to less than a third in advanced economies
  • Healthcare AI is the most trusted application in 42 of 47 countries surveyed
  • 83% of people globally believe AI will bring benefits, but 79% express concerns about risks
  • Two in five people globally have experienced negative outcomes from AI use
  • 70% believe AI regulation is necessary, but 83% are unaware of existing AI laws
  • Nearly half of employees admit uploading sensitive company information to public AI tools
  • 83% of students regularly use AI, with widespread inappropriate use raising concerns about critical thinking skills

Across 47 countries and 48,000 respondents, my Google Notebook LM AI agents uncover a stark divide: emerging economies are embracing AI with significantly higher enthusiasm than advanced economies. Nations like India and Nigeria report staggering 92% AI adoption rates, while developed nations show more hesitation. 

This pattern extends beyond mere usage emerging economies demonstrate higher AI literacy, with about half receiving some form of AI training compared to less than a third in advanced economies.

Trust follows similar geographic contours. While global trust in AI sits at 58% overall, the confidence gap between regions is substantial. People generally trust AI's technical capabilities more than its safety or ethics, with healthcare applications earning the highest trust globally. When weighing benefits against risks, half of respondents in emerging economies believe the advantages outweigh concerns, while advanced economies remain deeply divided.

In workplaces worldwide, concerning patterns emerge despite high adoption rates. Nearly half of employees admit uploading sensitive company information into public AI tools, two-thirds rely on AI output without proper evaluation, and over half try to hide their AI use. Educational settings show similar trends, with 83% of students regularly using AI but often inappropriately, raising serious questions about impacts on critical thinking skills.

This research identifies trust as the central foundation for AI acceptance, with literacy, perceived benefits, and confidence in governance systems as key drivers. The striking regional contrasts suggest potentially divergent futures: emerging economies positioned to accelerate innovation through enthusiastic adoption versus advanced economies exercising what might be rational caution.

What responsibilities do we each hold in shaping an AI future that's not just powerful but truly trustworthy? How might we bridge the identified gaps in understanding and trust? We encourage you to reflect on what these findings mean for your world as AI continues its rapid evolution.

Support the show


๐—–๐—ผ๐—ป๐˜๐—ฎ๐—ฐ๐˜ my team and I to get business results, not excuses.

โ˜Ž๏ธ https://calendly.com/kierangilmurray/results-not-excuses
โœ‰๏ธ kieran@gilmurray.co.uk
๐ŸŒ www.KieranGilmurray.com
๐Ÿ“˜ Kieran Gilmurray | LinkedIn
๐Ÿฆ‰ X / Twitter: https://twitter.com/KieranGilmurray
๐Ÿ“ฝ YouTube: https://www.youtube.com/@KieranGilmurray

AI Host 1:

Welcome to the Deep Dive. Forget endless scrolling. We take the most insightful research, the best data and we really pull out the crucial knowledge you need.

AI Host 2:

Right Efficiently, without getting overwhelmed.

AI Host 1:

Exactly, it's for you, the learner, if you're keen to grasp important topics quickly, and today we're diving into something pretty significant.

AI Host 2:

Yeah, today's focus is this really far-reaching global report. It looks at how people across the world are viewing and, you know, interacting with artificial intelligence. This isn't just opinions, right, yeah, no, it's based on some serious research from the University of Melbourne.

AI Host 1:

Yeah, exactly, they surveyed what? Over 48,000 people. Huge numbers Across 47 countries. This was between late 2024 and early 2025.

AI Host 2:

Right, covering basically every major global region, and they used carefully designed representative sampling too.

AI Host 1:

So our mission here, for you, the learner, is to really pinpoint the most vital takeaways from all this research.

AI Host 2:

We're going to unpack how the world is actually using AI, thinking about it right now.

AI Host 1:

Expect some surprising data points. I think Key trends for sure.

AI Host 2:

And, hopefully, a clear understanding of what this all means, as AI just keeps evolving so fast.

AI Host 1:

Yeah, the report looks into trust real world use the good, the bad.

AI Host 2:

Perceived advantages, disadvantages yeah, and the call for regulation AI in work, ai in education yeah, it's all in there.

AI Host 1:

Okay, so let's dive in, let's get to those core findings.

AI Host 2:

Right. Where do we start? Usage?

AI Host 1:

Yeah, let's start with who's actually using AI. The report says, you know, intentionally using AI tools is pretty widespread globally.

AI Host 2:

It is, but there's a really noticeable difference, isn't there, depending on the country's economic status.

AI Host 1:

Definitely what stands out.

AI Host 2:

Well, it's this stark contrast, really, between emerging and advanced economies.

AI Host 1:

Okay.

AI Host 2:

So in emerging economies you see considerably higher rates of people using AI tools regularly or semi-regularly.

AI Host 1:

It's around 80%, 80%, wow.

AI Host 2:

Yeah, which is quite a gap when you compare it to advanced economies. The figure's closer to 58%.

AI Host 1:

That's a huge difference. Why do you think that is? Any ideas from the report.

AI Host 2:

Well, the report doesn't speculate too much on the why, but you could guess maybe AI tools or like meeting specific needs or solving problems in emerging economies that just resonate more strongly there, or maybe leapfrogging older tech.

AI Host 1:

That makes sense, a really significant divide. Are there specific countries that jump out?

AI Host 2:

Oh, absolutely. You look at India and Nigeria, both emerging economies. They report incredibly high usage, like 92%. 92%, it's massive. But then, interestingly, some other emerging economies, maybe in Eastern Europe, like Hungary or Romania, they show much lower adoption rates.

AI Host 1:

Huh, so it's not monolithic, even within emerging economies.

AI Host 2:

Not at all and in terms of what they're using, it seems general purpose generative AI tools are the most common.

AI Host 1:

So things like chat, gpt, the ones that create text code, that sort of thing.

AI Host 2:

Exactly, those are the most prevalent. Pretty much everywhere, people are jumping straight into the versatile stuff of publicly available AI Makes sense it's accessible. Yeah, and the report also found that folks in emerging economies seem to have well greater awareness and use of AI. That's already embedded in tech they use every day.

AI Host 1:

Okay, how much more?

AI Host 2:

So it's like 70% awareness compared to 61% in advanced economies.

AI Host 1:

Yeah.

AI Host 2:

And 88% usage versus 74%.

AI Host 1:

So maybe it feels less like a separate AI thing and more just part of the tech landscape for them.

AI Host 2:

Could be or perhaps it's just more visible more obviously part of their existing tech experiences. It's an interesting finding.

AI Host 1:

It really is. It makes you wonder about that perception difference. So they're using it more, but does that mean they understand it better? What about AI literacy?

AI Host 2:

Yeah, that's a crucial point and the report hits this hard. It turns out emerging economies also lead in what the researchers call AI literacy.

AI Host 1:

Which includes what exactly?

AI Host 2:

It covers training, general knowledge about AI and also feeling effective capable when using.

AI Host 1:

Right.

AI Host 2:

And the numbers are quite telling. About half the people in emerging economies say they've had some kind of AI training or education.

AI Host 1:

Okay, and in advanced economies?

AI Host 2:

It's less than a third. Big difference again.

AI Host 1:

Wow, that really does challenge some common assumptions, doesn't it?

AI Host 2:

It absolutely does. And it goes further. Two-thirds of people in emerging economies report having like moderate to high AI knowledge.

AI Host 1:

Compared to.

AI Host 2:

Less than half in advanced economies. Okay and feeling effective using AI. Three quarters in emerging economies feel that way, but only about half in advanced economies.

AI Host 1:

So it's not just more use, it's more training, more knowledge and more confidence.

AI Host 2:

Exactly, it seems to be a whole package.

AI Host 1:

Any particular country stand out for high literacy.

AI Host 2:

Yeah, definitely. Nigeria, egypt, the UAE, india, china, saudi Arabia they all show really high levels of both training and general knowledge.

AI Host 1:

And on the lower end.

AI Host 2:

Germany, the Czech Republic, japan. They report notably lower levels Interesting the Czech Republic, japan. They report notably lower levels. And here's something else that's quite neat the people who report higher AI knowledge. They also tend to want to learn more.

AI Host 1:

That virtuous cycle idea.

AI Host 2:

Precisely Globally, 83% are interested in learning more, but that number jumps to over 90% in most of those high literacy emerging economies.

AI Host 1:

Whereas some advanced economies show less appetite for more learning.

AI Host 2:

Yeah, Places like Australia and Finland show lower interest. It really looks like understanding breeds curiosity breeds more understanding.

AI Host 1:

Makes total sense. Okay, so higher use, higher literacy in emerging economies. What about trust? That's fundamental, isn't it? Do people trust this tech?

AI Host 2:

It's definitely fundamental and globally, trust is well mixed. About 58% view AI systems as trustworthy overall.

AI Host 1:

Okay, but trustworthy can mean different things.

AI Host 2:

That's a key point. The report makes People generally have more faith in AI's technical ability, you know, can it give helpful output? Does it work?

AI Host 1:

Right, the functional aspect.

AI Host 2:

Yeah, they trust that more than they trust its safety, its security or its ethical use.

AI Host 1:

That rings true. Like you might trust it to draft an email, but maybe not to diagnose a complex illness on its own.

AI Host 2:

yet Exactly that kind of distinction. And guess what? The economic divide shows up here too.

AI Host 1:

Let me guess, emerging economies trust more.

AI Host 2:

You got it. In advanced economies, about three out of five people are either unwilling to trust AI or they're just unsure.

AI Host 1:

Hmm, wary yeah.

AI Host 2:

But in emerging economies it flips. Roughly three out of five do express trust.

AI Host 1:

Consistent pattern? Is trust different for different types of AI?

AI Host 2:

Interestingly, not massively Levels are pretty similar for general AI, generative AI, even AI used in HR processes.

AI Host 1:

Okay.

AI Host 2:

But there's one big exception healthcare AI.

AI Host 1:

Ah, people trust that more.

AI Host 2:

Significantly more. It's the most trusted application in, I think, 42 out of the 47 countries surveyed.

AI Host 1:

Wow, why do you suppose that is?

AI Host 2:

Well likely because the potential benefits feel very direct, very tangible, very personal. Improving health outcomes that's powerful.

AI Host 1:

Yeah, that potential positive impact should definitely overcome some initial skepticism.

AI Host 2:

Seems likely, and the report does confirm that linked trust levels correlate pretty strongly with both acceptance and actual use of AI.

AI Host 1:

Makes sense. You use what you trust.

AI Host 2:

Right and the countries with the highest trust and acceptance Nigeria, india, egypt, china, again lowest Finland, japan, czech Republic, germany See the pattern.

AI Host 1:

Yeah, the same groups cropping up. Ok, so varying crust levels. What about expectations? What do people think AI will actually do for them or to them? Benefits versus risks?

AI Host 2:

Well, there's widespread optimism about benefits. A huge majority 83% believe AI will bring benefits.

AI Host 1:

Okay, belief is one thing. Are they seeing benefits yet?

AI Host 2:

Yeah, a good chunk are 73% report already experiencing positive outcomes. Things like you know, being more efficient, better accessibility, getting rid of boring, mundane tasks.

AI Host 1:

Standard AI promises really Pretty much.

AI Host 2:

However, this optimism is definitely balanced by widespread concerns about the risks. Well, 79% are concerned. Top worries include cybersecurity threats, losing human connection, the spread of misinformation that's a big one and, of course, job losses.

AI Host 1:

And have people actually run into problems.

AI Host 2:

Quite a few. The report says two in five people globally have personally experienced some kind of negative outcome from AI.

AI Host 1:

Two in five, that's actually quite high. That's not insignificant, no it's substantial.

AI Host 2:

And surprise, surprise, the economic divide appears again.

AI Host 1:

Let me guess emerging economies more optimistic about benefits.

AI Host 2:

Spot on 90% in emerging economies expect benefits, versus 79% in advanced, and 82% there say they've realized benefits, compared to only 65% in advanced economies.

AI Host 1:

So they're both expecting more and feeling like they're getting more.

AI Host 2:

Seems that way, and when you ask if the benefits outweigh the risks, Different answers again. Yep. Half the people in emerging economies think yes, benefits went out. But in advanced economies opinions are much more divided, pretty much split down the middle.

AI Host 1:

So that whole calculation, the trade-off, feels very different depending on where you live.

AI Host 2:

It really does. Nigeria, china, egypt. Highest belief that benefits outweigh risks, lowest Australia, new Zealand, a bunch of northern and western European countries like the Netherlands, sweden, finland, canada, ireland, France too.

AI Host 1:

Interesting list. More cautious perhaps.

AI Host 2:

Seems so. And one other point on this globally, that feeling that benefits outweigh risks, it's actually decreased a bit between 2022 and 2024.

AI Host 1:

Oh, so maybe people are becoming a bit more cautious overall as they see more of AI in action.

AI Host 2:

That could be it. Maybe the novelty wears off or the potential downsides become clearer over time. It suggests a shifting public perception.

AI Host 1:

That makes sense. That growing caution Okay. So, with all these benefits and risks floating around, what's the feeling about regulating AI? Do people want rules?

AI Host 2:

Oh yeah, there's a very strong public mandate for regulation. Seventy percent believe it's necessary.

AI Host 1:

But do they think the current rules are enough?

AI Host 2:

No, not really. Only 43% feel current laws are adequate.

AI Host 1:

So a desire for more or better regulation? What kind?

AI Host 2:

People seem to expect it at multiple levels 76% want international laws, global standards, 69% look to their national governments to regulate and 71% actually support co-regulation involving industry working with government.

AI Host 1:

Interesting mix, any specific area they really want regulated.

AI Host 2:

Misinformation Big time. 87% want laws specifically targeting AI-generated misinformation. That really stands out as a major public concern.

AI Host 1:

Yeah, you hear about that constantly. Deep fakes, fake news.

AI Host 2:

Exactly. It's clearly top of mind, but here's the catch.

AI Host 1:

What's that?

AI Host 2:

Despite wanting regulation, most people don't know much about what already exists. 83% admit they're unaware of any current AI laws or regulations.

AI Host 1:

Wow so strong demand, low awareness. That's a gap.

AI Host 2:

It is, and when you ask who they trust to actually develop and use AI responsibly.

AI Host 1:

Who comes out on top?

AI Host 2:

Universities, research institutions and health care organizations get the highest confidence ratings.

AI Host 1:

Makes sense and lower down.

AI Host 2:

Governments fare a bit lower. And then industry, big tech Confidence there varies a lot. It's generally lower in advanced economies compared to emerging economies.

AI Host 1:

Which ties back to the overall trust patterns we saw earlier right.

AI Host 2:

Exactly. It all seems interconnected.

AI Host 1:

Okay, let's pivot a bit. What about AI, where many of us spend a lot of time in the workplace? How's it being used there?

AI Host 2:

Well, the report shows pretty high intentional regular use at work 58% of employees globally.

AI Host 1:

And how often is regular.

AI Host 2:

About a third are using it weekly or even more often okay, and are these company provided tools or? Interestingly, mostly not the most common tools are often the free, public generative AI platforms like chat, gpt.

AI Host 1:

Again, it suggests employees are driving this adoption themselves, sort of bottom-up so workers are just bringing it in, finding uses for it pretty Pretty much.

AI Host 2:

And following the pattern, employee adoption is higher in emerging economies 72% using it regularly there, versus 49% in advanced economies.

AI Host 1:

Okay. Are they finding it helpful? What are the impacts?

AI Host 2:

They definitely report performance benefits, like getting tasks done faster, improving quality, but there are also mixed impacts.

AI Host 1:

That so.

AI Host 2:

Mixed views on whether it increases or decreases workload. Overall Mixed impacts on human interaction at work. Compliance is an issue and job security worries are real. Two in five employees think AI could potentially replace jobs in their field.

AI Host 1:

That job, displacement, fear again. It seems pervasive.

AI Host 2:

It really is and the report also flags some well concerning usage patterns. They call it complacent and inappropriate use.

AI Host 1:

Uh-oh, like what.

AI Host 2:

Get this. Nearly half of employees admit they've gone against company policy by like uploading sensitive company info into public AI tools.

AI Host 1:

Wow, that's a huge security risk.

AI Host 2:

Massive and two-thirds say they rely on AI output without really evaluating it properly. Over half admit they've made mistakes because they rely too much on AI.

AI Host 1:

Yikes. So uncritical acceptance leading to errors.

AI Host 2:

Exactly and, adding to the concern, over half also admit they try to hide when they use AI for work tasks.

AI Host 1:

Lack of transparency too. Why hide it? Fear of judgment Policy violations?

AI Host 2:

Could be all of the above. It's just a culture. Transparency too. Why hide it? Fear of judgment, Policy violations Could be all of the above. It suggests a culture issue, maybe.

AI Host 1:

Definitely worrying trends? Are companies keeping up with policies and training?

AI Host 2:

It seems not really Governance and training are lagging behind this rapid adoption. Only about half of employees in advanced economies say their company even has training or policies on responsible AI use.

AI Host 1:

And for generative AI specifically.

AI Host 2:

Even lower. Only two in five report specific policies for tools like ChatGPT.

AI Host 1:

So employees are using these tools widely, sometimes inappropriately, and the organizational guardrails are often missing.

AI Host 2:

That's the picture and what's more? Trust in AI at work and the feeling that their organization supports responsible use. That's actually declined in many countries since 2022.

AI Host 1:

Declined. That's not good.

AI Host 2:

No, it suggests things might be moving in the wrong direction in some workplaces regarding trust and support.

AI Host 1:

Okay, so a complex picture in the workplace. What about education? Students must be all over this tech right.

AI Host 2:

Oh, even more so than employees, it seems. Regular AI used by students, especially university level, is incredibly high 83% 83%, and are they finding it useful? Yes, they report similar benefits efficiency in studying, getting personalized learning, help reducing workload, maybe even stress.

AI Host 1:

But I sense another but coming Are the same issues of misuse popping up.

AI Host 2:

Unfortunately yes, and maybe even more intensely. The report states that inappropriate, complacent and non-transparent use is widespread among students.

AI Host 1:

Raising concerns about.

AI Host 2:

Big concerns about over-reliance potentially undermining critical thinking skills, collaboration abilities and just the basic fairness of assessments. How do you grade fairly if everyone's using AI differently or hiding it?

AI Host 1:

Yeah, huge implications for learning Are schools and universities providing guidance?

AI Host 2:

Again, it's lagging. Similar to the workplace. Only about half of students say their institution has policies, resources or training for responsible AI use.

AI Host 1:

So the students are adopting faster than the institutions can adapt.

AI Host 2:

Seems that way, and the report notes, students are actually more likely than employees to report inappropriate use and over-reliance.

AI Host 1:

That could seriously impact their development, couldn't it? Long term?

AI Host 2:

Absolutely. You worry about skills they might not develop if they lean too heavily on AI too early. The report mentions mixed impacts being reported by students on their critical thinking, communication skills, trust in the academic setting and fairness.

AI Host 1:

Wow, skills, trust in the academic setting and fairness. Wow, okay, we've covered a ton of ground usage literacy, trust, benefits, risks, regulation, work, education. Phew, let's try to pull it together. What are the main things driving whether people trust and accept AI?

AI Host 2:

Right, the key drivers. Well, the report is really clear Trust itself is the central pillar for acceptance. If people don't trust it, they won't accept it.

AI Host 1:

Makes sense and what builds that trust?

AI Host 2:

AI literacy is a big one. That combination we talked about knowledge, feeling capable, having training that consistently boosts both trust and acceptance.

AI Host 1:

So understanding it helps you trust it.

AI Host 2:

Generally, yes. Also, expecting benefits is a strong motivator. If you think it's going to help you, you're more likely to trust and accept it.

AI Host 1:

Okay, literacy expected benefits anything else.

AI Host 2:

Yes, and this is maybe the most significant factor influencing trust, specifically those institutional factors, meaning how adequate people think the regulations are and how much confidence they have in entities like governments, tech companies, researchers to use AI responsibly. Those factors have the strongest pull on trust levels.

AI Host 1:

So it's not just the tech, it's the whole ecosystem around it, the governance.

AI Host 2:

Exactly People need to feel the systems are in place to manage it well and demographically, ai, training and income level consistently pop up as having the strongest effects on trust and acceptance.

AI Host 1:

Interesting. What about gender?

AI Host 2:

The report found no significant gender differences across the main indicators.

AI Host 1:

Okay, so summarizing that big divide we kept seeing emerging versus advanced economies, what's the final take?

AI Host 2:

Well, the contrasts are just really striking, aren't they? Across the board, ai use trust, acceptance, literacy, perceived benefits.

AI Host 1:

Emerging economies are generally higher on all those.

AI Host 2:

Yes, generally higher adoption, greater trust, better literacy, more positive attitudes overall, which you know could position them to really accelerate innovation, maybe gain a technological edge.

AI Host 1:

And the lower trust in advanced economies.

AI Host 2:

Is that just Luddism, or the report suggests it might be seen as well. Rational caution. Perhaps it reflects a clearer view of the current state of AI, Maybe more skepticism about existing governance. Or maybe it's linked to their lower self-reported AI literacy they feel less equipped, Perhaps more wary.

AI Host 1:

That makes sense. Are emerging economies not seeing the downsides, then?

AI Host 2:

Oh, they are experiencing negative outcomes too. Similar rates in some cases, but it seems the benefits feel much more amplified to them, outweighing the negatives more strongly in their perception.

AI Host 1:

Got it. So, putting this all together, what are the big implications? Where do we go from here?

AI Host 2:

Well, the clear message is the need for a really balanced approach. We need strategies to integrate AI that actively try to maximize all those potential benefits.

AI Host 1:

While seriously mitigating the risks we've talked about.

AI Host 2:

Exactly Proactively Strengthening responsible AI use and building that trusted adoption. It has to be based on evidence, on understanding what works.

AI Host 1:

And this research helps provide that evidence base.

AI Host 2:

Absolutely. It gives critical insights for businesses, governments, educators, anyone shaping strategy, policy or practice around AI.

AI Host 1:

And that call for international cooperation seems vital too.

AI Host 2:

Definitely On guidelines policy regulation. Given how global AI is, this survey really provides that crucial empirical data for developing AI in a human centered way.

AI Host 1:

Because the public concerns are real, right. Malfunctions, misuse, overuse.

AI Host 2:

Yes, and addressing those isn't just a technical problem. It needs technical fixes, sure, but also social changes, organizational shifts, better regulation and individuals taking responsibility too. It's multifaceted.

AI Host 1:

Okay. So for you, the learner listening to this, the key takeaways are pretty stark AI adoptions happening fast globally, but there are these deep divides, usage attitudes, especially between advanced and emerging economies.

AI Host 2:

Yeah, and this real tension between the potential benefits everyone sees and these persistent, maybe even growing, concerns about the risks.

AI Host 1:

Plus a strong public demand for regulation coupled with low awareness, and underpinning it all is the critical role of AI, literacy and trust If we don't get those right, it's hard to see how we navigate this well, which leads us to a final thought for you to chew on. Go ahead.

AI Host 2:

Given everything, we've discussed these diverse global views. What responsibilities do we really have Individuals, companies, governments. What's our role in shaping an AI future that's not just powerful, but actually trustworthy?

AI Host 1:

And maybe more specifically, what concrete actions could actually bridge these gaps. The research highlights gaps in understanding, gaps in trust. What works? Good questions they are. We really encourage you to dig deeper, if this sparked your interest. This Deep Diver member is based on that extensive research by the team at the University of Melbourne. Think about what these findings mean for you in your world.

AI Host 2:

Absolutely. Thanks for joining us for this deep dive.

People on this episode

Head Shot

Mr Kieran Gilmurray

Host