There’s a trust crisis afoot.

Earlier this week, Edelman’s annual Trust Barometer report suggested that trust is collapsing in America. In a survey of 33,000 people across more than 28 countries, only a third of Americans responded that they trust the government, a 14 percentage point decline from last year. Fewer than half of us trust the media. Even our confidence in business, which has for years remained strong, was shaken this year, dropping 10 points; just 48 percent of respondents trust business to “do what is right.” In the 18 years that Edelman has administered this survey, it has never recorded such dramatic drops across a single country.

But as Americans’ trust in nearly every institution dissipates, one industry seems insulated: Tech. That’s right, Americans still trust tech companies. Specifically, 75 percent of those surveyed said they trusted the tech industry to “do what is right,” a percentage that has remained nearly unchanged for five years. According to the poll, tech is the most trusted industry in America.

That’s surprising after a year marked by scandals and concern over tech companies accumulation of power. The question of whether tech companies could retain their privileged position reverberated through the World Economic Forum, in Davos Switzerland this week. Speaking on a panel entitled “In Tech We Trust,” Alphabet chief financial officer Ruth Porat advanced a theory. Trust in these companies is strong, she posited, “because technology continues to solve some of the world’s most pressing issues.”

She’s right, of course. It’s the justification the tech industry always uses for its existence and outsized influence. The idea that tech is insulated from negative popular opinion because its innovations are so astounding has been the prevailing corporate sentiment since before Google was a twinkle in its founders’ eyes.

In the interim, technology has altered the world in complicated ways. Tech companies, like Google, Facebook, and Twitter have lead a massive shift in the way we communicate, which has changed who gets to have a voice. By giving everyone a similar-sized microphones, tech platforms have flattened the playing field and allowed anyone to convey anything. While this has had many positive effects, this friction-free approach to publishing has had darker consequences too. Giving weight to any opinion that can game the system’s algorithms has eroded our faith in traditional institutions, helped aid in the rise of authoritarian regimes, and facilitated the proliferation of disinformation. All of this is reflected in the plummeting trust in other types of institutions in Edelman’s data.

Somehow, we come out of all of that trusting tech companies.

Trust, as a term, has a fuzzy definition. Edelman asks specifically about whether respondents think that institutions—businesses, NGOs, government and the media—are likely to do what is right. It’s a forward-looking metric that suggests that if people are confident in an institution’s future actions, then the current system will continue to function well.

It’s not a given that our trust in tech will continue; in fact, there’s reason to believe it won’t. Though Americans may trust companies like Google or Amazon, according to the Edelman survey, we’re growing less trustful that search engines and social media platforms will deliver us accurate news and information. (Roughly 70 percent of respondents said they “worry about fake news or false information being used as a weapon.”) If we lose faith in tech companies’ products and services, eventually we’ll question the businesses behind them. While the greater public may believe tech companies will do the right thing, our democracy’s watch dogs—both regulators, and the media—are beginning to exercise skepticism, and to demand information as best they can.

Nowhere is this crisis more visible than Facebook. The social network is being criticized widely, including by some influential early employees and advisers, for creating a product that may be harmful to its users. Speaking at a European tech conference last week, the company’s communications and public policy chief apologized for failing to do more, earlier, to fight hate speech and foreign influence on the platform. These types of concerns aren’t specific to Facebook: Any of the tech companies that optimize large amounts of data to deliver us products and services risk alienating the public as they use the power that comes with that information.

Speaking at Davos, Uber CEO Dara Khosrowshahi pointed out that consumers face a challenge in trying to understand tech’s influence in the age of big data. He called this an “information asymmetry.” In his previous job, as CEO of Expedia, Khosrowshahi said, customers were shown a tropical island while they waited for their purchase page to show up. As a test, engineers replaced the placid image with a stressful one that showed a person missing a train. Purchases shot up. The company subbed in an even more stressful image of a person looking at a non-working credit card, and purchases rose again. One enterprising engineer decided to use image of a cobra snake. Purchases went higher.

What’s good for a business isn’t always good for that businesses’ users. Yet Khosrowshahi stopped testing because he decided the experiment wasn’t in line with the Expedia’s values. “A company starts having so much data and information about the user that if you describe it as a fight, it’s just not a fair fight,” said Khosrowshahi.

But as Khosrowshahi illustrates, it’s difficult for outsiders to hold tech companies accountable for decisions that are so subtle they are imperceivable to users. And, as Uber’s questionable business practices under its former CEO, and Facebook’s reticence to deal with foreign agents meddling on its platform show, tech’s leaders haven’t typically set a strong moral compass without pressure.

The tech industry often responds to these concerns with a promise to be more transparent—to better show how its products and services are created and how they impact us. But transparency, explained Rachel Botsman in the same Davos conversation, is not synonymous with trust. A visiting professor at the University of Oxford’s Said School, Botsman authored a book on technology and trust entitled “Who Can You Trust?” “You’ve actually given up on trust if you need for things to be transparent,” she said. “We need to trust the intention of these companies.”

At Davos, some tech CEOs spoke in favor of regulation, arguing that proactive intervention might calm the public. Salesforce ceo Marc Benioff, also a panelist, compared some social networking companies to the tobacco industry: “Here’s a product – cigarettes – they’re addictive, they’re not good for you,” he said. But technology has addictive qualities, he explained, and product designers often work to make them even more addictive. “Maybe there’s all kinds of different forces trying to get you to do certain things. There’s a lot of parallels,” he said. (Keep in mind that it’s hardly self criticism; Salesforce is an enterprise company with different business incentives than the current crop of social networks.) Khosrowshahi also noted regulators have a role to play in holding tech companies accountable.

Even among tech’s leadership, no one voiced a clear idea of what form regulation might take. But from the conversation, on stage and beyond, it was clear that tech companies are worried about how, and when, the public will catch on to the power they now exert. For now, they’re operating on borrowed time. To keep the good graces of their users, they will need to find a way to maintain our trust—to continue to convince us that they deserve our faith as they pilot us into the unknown.

• Why techies still don't get that the rest of the world hates them
• Trust is a funny thing for Facebook's news feed to measure
• Google employees are fighting a dirty war over diversity