Restoring trust in the age of artificial intelligence

We are far beyond the point of trusting casual scrutiny and human instinct to identify content generated by artificial intelligence (AI). Synthetic video and audio have become virtually indistinguishable from the real thing, meaning every piece of digital content we see has the potential to be fraudulent. The central issue is that when material seems authentic but can’t be easily and immediately verified, trust erodes. Addressing this challenge requires shifting from reactively determining whether something is “real” or “fake” to removing ambiguity entirely.

In this new information landscape, organisations are battling in an environment where scammers and malicious actors don’t need to be wholly convincing. So long as an image or recording looks credible enough, it’s easy to exploit victims. For organisations, even the most well-established, trusted brand identity can be damaged by false narratives, casting doubt on legitimate material.

AI ambiguity impacting bottom line

Advanced AI has lowered the barrier to entry for cybercriminals to create deepfakes and other artificially generated scams. With minimal technical expertise, scammers can now develop intentionally harmful or misleading information on an industrial scale, making it almost impossible for individuals to avoid these attempts. For modern businesses which rely heavily on digital communication, fraudulent content risks undermining relationships with both internal and external stakeholders.

Consumers are persistently targeted by false advertising, and fake promotions are rampant on social media. One recent example targeted a social media platform itself, with scammers using AI-generated videos of celebrities to promote fake TikTok services which used the platform’s official branding. Not only can falling victim to a scam damage trust in the business being impersonated, but the proliferation of AI content also makes it difficult to believe the legitimacy of genuine promotions.    

This also risks eroding investor confidence. Executives are now frequently deepfaked, promoting illegitimate advice or making requests for sensitive information. False corporate announcements can mimic actual business communications so closely that, even after being debunked as fake, it’s too late to undo the immediate damage.  

Awareness training and reporting AI content are no longer viable solutions. Instead, content must be immediately verifiable and show proof of a legitimate issuer and whether it’s been altered. This is why the emerging concept of content trust is becoming essential. Instead of trying to identify what’s fake after the fact, authenticity must be visible at the point of consumption.

From reactive verification to proof of provenance

To combat the uncertainty caused by AI abuse, digital content must be immediately verifiable and encrypted. Relying on retrospective checks after the content has already spread online and had its authenticity questioned is unsustainable. Trust must be embedded directly into content with cryptographic assurances. This ensures any viewer can immediately verify the source of an image, video or audio file, without waiting for subsequent validation from platforms, businesses or public opinion.

Organisations must therefore be able to cryptographically sign and verify digital content, creating a transparent, tamper-evident record of its provenance. This makes it possible to prove who produced the content, how it was generated and whether it’s been modified in any way. Unlike platform-based verification, these credentials are attached to the file itself. No matter how the content is then shared outside the originator’s control, it remains cryptographically signed and verified.

Beyond this, the strongest foundation of trust organisations can establish happens at the moment of capture. By embedding Coalition for Content Provenance and Authenticity (C2PA) standards into trusted devices, organisations can cryptographically sign and timestamp content the moment it is created. Cameras and other imaging equipment must all eventually have the capability for content to be signed and timestamped at the source.

Misinformation, brand impersonation and AI-generated fraud continue to impact businesses’ reputation and revenue. Confidence in digital media is suffering and without better assurances and stronger verification methods throughout the content lifecycle, the problem will only escalate.

The new standard for digital trust

Businesses are increasingly forced to respond to brand impersonation, misinformation and manipulated content after the damage has already been done. This reactive ‘firefighting’ approach is inefficient, ineffective and unsustainable. Instead, organisations must establish the authenticity of their digital content from the moment it is created, ensuring it remains verifiable for its entire lifetime.

As AI makes it possible to generate convincing images, video and audio at scale, authenticity must be something that can be proven rather than assumed. In this environment, trust becomes one of the most valuable attributes any piece of digital content can possess.

Effectively proving provenance requires a trusted foundation. The same proven cryptographic technologies that underpin resilient infrastructure, such as secure websites, must now extend to digital content that shapes public trust and underpins truth and security in a fragmented digital world.