What do the pope and most Australians have in common? Neither trust AI companies

It's not often that the pope, college students and Australians agree.

One topic that has brought these unusual bedfellows roughly onto the same page is artificial intelligence.

Last week, Pope Leo XIV published his first encyclical, a 43,000-word plea to address the threats posed by AI.

American college students have been expressing their concerns by booing esteemed commencement speakers who mention the new technology (the exception that proves the rule was Australian comedian Ronny Chieng, who was greeted with cheers for his message "F*** AI" at a Harvard ceremony).

It's not just a set of outlier students and pontiffs, either. Most Australians do not have particularly fond feelings towards the new technology, polling shows.

Australians dislike and distrust AI 

Australians have the equal most negative views towards AI in the world, according to an EY survey of people from 23 countries.

Almost half of Australians surveyed for the Minderoo Foundation say the risks outweigh the benefits, compared with about a third who believe the opposite.

Only 4 per cent of Australians say they trust their private information with AI companies, according to a new survey done by Australia's Office of the Information Commissioner.

Research from the Tech Policy Design Institute found that just 1 per cent of Australians say they have complete trust that AI will be used responsibly.

This feeling is manifesting in real-world politics. Just a few years ago, the construction of data centres serving AI and other tech services was mostly uncontroversial and rarely newsworthy.

Now, residents from places like Lane Cove in Sydney and West Footscray in Melbourne are opposing building "AI factories" in their areas. Last month, a developer pulled a proposal to build a data centre near Perth, citing community opposition.

It's a big enough problem that this week, Assistant Minister Andrew Charlton warned that Australia would face a US-style anti-AI backlash unless it kept Australians safe from the harms of AI.

So, why do Australians dislike and distrust this new technology so much?

After all, it's hard to imagine crowds cheering or a pope pontificating about another form of computing like this. (Although I would personally love to hear the bishop of Rome share his thoughts on the most promising approach in quantum computing.)

On paper, AI sounds like it should be popular. Its proponents predict that it will make us all richer, more capable and freer from drudgery thanks to the automation of boring tasks.

One school of thought is that the AI backlash is the predictable and typical response to any technology promoted as being transformative and disruptive. This growing antipathy towards AI is an issue that one Australian industry group representing tech companies called for the government to run a public campaign to "improve digital literacy and counter misinformation".

Fears about the workforce

While many Australians may not be as immersed in AI as its makers and promoters, there's a big factor other than what Australians aren't hearing about AI.

It's what they are hearing about it.

Last month, the chief of global bank Standard Chartered said that AI would replace "lower-value human capital". That's human beings.

Microsoft's AI chief said most tasks performed by white-collar workers will be automated within the next 12-18 months.

Anthropic CEO Dario Amodei predicted that 50 per cent of white-collar jobs would be eliminated in five years, pushing unemployment to 10-20 per cent.

Some of this is AI leaders warning about the consequences of AI replacing human work, but it's also part of the marketing pitch, too.

For all the predictions made by AI companies and the government about economy-wide productivity boosts or about helping businesses cut costs, most working people rely on their jobs.

What those people are hearing is that AI might make them more efficient, but it might also make their roles redundant. (Evidence of AI affecting Australian employment across the economy remains fairly weak, but it is early days.)

Another issue is the existential risks associated with the technology. Whether it's the fear that a rogue AI will go Skynet and destroy our systems or that it'll be used to create synthetic bioweapons, these potential downsides are another reason to dislike the technology.

There's also the immense amount of hype surrounding AI.

Bill Simpson-Young is a former CSIRO researcher who left to set up the Gradient Institute, an Australian not-for-profit dedicated to promoting the safe use of AI.

He reckons it's obvious why Australians don't trust AI: "It's because they've got bullshit detectors."

Simpson-Young says there's scepticism about the influence of American tech companies — some of which already have battered reputations because of their social media operations — and a fear about how the tech could further consolidate power in the hands of a few.

Then, there's the actual lived experience of many Australians who are already being exposed to AI and its impacts.

Australians are regularly being exposed to AI slop and fake news. They're hearing about school children being abused with non-consensual sexual deepfakes. Their favourite artists are talking about how companies have ripped off their content without their knowledge, consent, or compensation. They see the headlines about tech companies paying little tax.

'I don't trust it'

Crucially, Australians' views are grounded in using the technology. We're among the highest users of AI in the world. Instead of this uptake allaying our fears, our use is feeding it. Surveys done over time show that the more Australians use AI, the more they say they believe the risks outweigh the benefits.

As ACTU secretary Sally McManus told business and technology leaders at the Australian Financial Review's AI Summit earlier this week: "I don't trust it. On the other hand, I use it a lot."

This broad distrust added a rather gloomy note, coming up time and time again among Australian business and tech leaders who were otherwise enthusiastic about the technology.

Charlton's cautionary message was delivered at the same event. It was notably distinct from the standard stump politician speech about how the federal government wants to rev up AI investment and usage.

Other members of the government are pushing them to do more. Independent federal MP Kate Chaney released a discussion paper outlining 18 ideas for managing the benefits and risks of AI.

On the industry side, the group representing major data centre developers, Data Centres Australia, is now proactively engaging with local councils to try and make a positive case for how this infrastructure should be built in communities.

The growing tech backlash of the last decade has curdled into a febrile environment for AI makers who are trying to change Australian law, use Australian resources, and ask Australians to adopt their products.

The AI industry, business and government are now grappling with this issue, wondering why Australians don't trust AI and the people who make it.

Another way of looking at it is: what have they done to earn Australians' trust?