AI Deepfakes Are Reshaping Trust: What It Means for Truth

Overview

Artificial intelligence has quickly become one of the most prominent technological shifts of the 21st century. Ordinary users can now access not only to the world’s collective knowledge, but also to tools that can build, in seconds, tasks that once required entire teams.

But with every major technological leap there are tradeoffs. Some of the major concerns include environmental impact and intellectual ownership, but one of the most destabilizing risks is something else entirely: AI-generated deepfakes.

Synthetic media that appears nearly indistinguishable from reality. As the technology improves, it is increasingly blurring people’s ability to distinguish what is true.

While there is no shortage of concern around this issue, perspectives on its impact differ. Some view deepfakes as a challenge society will adapt to over time. Others see them as a more fundamental threat—one that could undermine trust in information itself.

Resilience argument

Optimists claims that this isn't a new trend, our social media and news outlets have long been plagued by misinformation and deepfakes are an evolution of existing threats like photoshop and fraudulent news that can be managed by detection tools and human innovation.

AI misinformation is a growing risk, but also being actively addressed:

The World Economic Forum identified AI-driven misinformation as a top global risk, and emphasizes solutions like detection and governance frameworks.¹
AI itself could help reduce misinformation:
- Some research suggests AI systems can align with fact-checkers and reduce misinformation exposure.²
A study showed training and media literacy tools can improve detection accuracy by up to 13 percentage points.³

While deepfakes introduce new challenges, they are believed to be part of a broader pattern of technological disruption. We’ve already adapted to misinformation before — this is the next phase

Structural risk argument

Critics argue that deepfakes represent a more fundamental shift. More than simply spreading false information, they introduce the “liar’s dividend”—a phenomenon where real evidence can be dismissed as fake, allowing politicians and public figures to deny reality itself.⁴

The concern is not just deception, but the erosion of trust in information altogether.

Public concern around the issue is already widespread:

74% of people say they worry about the societal impact of deepfakes.⁵
72% of consumers report concerns about being deceived by AI-generated content.⁶

At the same time, the technology is becoming increasingly common:

67% of people report seeing deepfakes regularly online.⁷
In one study, only 15% of participants correctly identified deepfakes after the fact.⁸

To critics, deepfakes are not merely an evolution of misinformation, but the emergence of a new information environment—one where the traditional rules surrounding evidence, trust, and reality itself no longer clearly apply.

Proposed Laws & Responses

In 2025, the bipartisan TAKE IT DOWN Act became one of the first major U.S. laws targeting harmful AI-generated deepfakes, particularly non-consensual explicit imagery.⁹

The European Union’s AI Act has introduced transparency and watermarking requirements for some AI-generated content.¹⁰

Tech companies are also developing detection software in realtime to weed out manipulated media.

Microsoft has experimented with AI-powered authentication tools capable of detecting signs of manipulation.
Adobe's Content Credentials aims to create a digital record showing how images and videos were created or altered.

Nibbles Take

Will deepfakes bring present a new wave of misinformation, never before seen? Maybe?

Yes technology is becoming more realistic, but the root issue may be our own misguidance.

Before deepfakes existed, many of us were already living in echo chambers, where beliefs are often reinforced. Social media algorithms and partisan media have made it easier for people to accept information that aligns with their views and dismiss information that does not regardless of validity.

Studies have shown that misinformation was spreading rapidly online long before generative AI emerged. In a 2018 study, researchers found that false news spread significantly faster and farther on social media than truthful reporting.¹¹

One of the most laughable examples came years before the rise of AI: persistent conspiracy theories on former President Barack Obama’s birthplace. Despite public records confirming he was born in the United States, speculation continued for years that he was born in Kenya.

In short the truth has never been our strongest talent, not to say the dangers are not real.

An example comes from The Boys TV show, where the character Homelander allows a passenger plane to crash after failing to save it. In later seasons, when footage of the incident surfaces it is dismissed as a deepfake created by his opponents.

While fictional, the scenario reflects real-world concern surrounding AI-generated media: not just the ability to create fake evidence, but the ability to discredit real evidence.

This may ultimately become one of the most dangerous aspects of deepfakes. Corrupt governments and public figures could dismiss authentic footage as fabricated, and at the same time push manipulated media to justify political crackdowns and military actions.

What is state of trust right now?

Currently about one-third of Americans reporting even a fair amount of trust in the news, deepfakes are entering a society already skeptical of traditional media.¹²

Yet that skepticism may create an opportunity. Along with the rise of deepfakes we may also see an increase in the value of credible journalism.

Organizations like The New York Times and Bellingcat have already expanded efforts to authenticate images and videos circulating online, a step toward rebuilding public trust.¹³ ¹⁴

Final verdict

Deepfakes may amplify misinformation and manipulation, but they are entering a society that was already struggling with trust. In that sense with journalism, public policy, and technological innovation all moving to address the problem, there is hope that society will adapt as it has before.

Deepfakes have already changed how we consume information. Now we just have to rebuild trust quickly enough to keep pace.