Ofer Familier, Co-Founder and CEO of Dig.

getty

​AI can fake anything now. A post, a glowing review or a quote from your CEO can all be produced in seconds, and audiences have stopped trusting any single piece of content on sight.

Platforms have responded with tools like “AI-generated” labels that either they or content creators can deploy. That’s a reasonable start, but once anything can be faked at scale, verifying real content one post at a time hardly moves the needle.

Transparency has outgrown defensive post-by-post disclosures, and brands should stop thinking about it that way. They should instead prove credibility across the whole operation, consistently, over time, with posts owned by real people who stand behind the content, and a record that holds up when someone goes back to check it.

That is a shift in what transparency means, which will require brands to develop new strategies for building trust going forward.​

The Current Status Of Trust Infrastructure

Over the years, a set of tools and standards has emerged to signal when content is AI-generated.

The EU AI Act now requires disclosure for AI-generated content. The C2PA standard, backed by Adobe, Google, Microsoft and others, attaches provenance data to media. Platform-level labels from Meta, TikTok and YouTube mark posts as AI-generated or made with AI.

That is meaningful infrastructure, but it has not closed the trust gap. While Zendesk found that 65% of leaders see AI as essential, 75% believe a lack of transparency around it will increase customer turnover.

The scale of the problem is already visible in the wild. Researchers recently documented a growing ecosystem of AI-manipulated ads on TikTok using deep-faked celebrities to promote fraudulent products, all of it running despite the platform’s own AI-content labeling policies.​

Why Content Labels Can’t Fix It

The trouble with content-level labels is structural, not a question of tighter enforcement.

A label on one post implies that the unflagged post next to it is the authentic baseline, and that implication is often incorrect. A lot of unflagged content already has AI threaded through it, because brands use AI for copywriting, image generation, scheduling optimization, comment responses and much more. Among the 78% of brands already using AI in external marketing creative, 87% use it for product images, 80% for marketing copy and 77% for background visuals, according to WFA research.

Tagging one specific post as “AI-made” while leaving everything else unlabeled tells consumers, by omission, that the rest is fully human-made. That is misleading, and people have caught on. Gartner found that 61% of consumers say they frequently question whether the information they rely on is reliable, and 68% frequently wonder whether the content they see is even real.

Since AI is now a foundational feature of how content gets made, flagging isolated instances does not reflect that reality. If AI is embedded in production, verification has to move to that same level.

The binary underneath all of this is breaking down as well. Because AI is now so pervasive, brands and consumers can no longer treat “AI-made” as a synonym for fake, or “no AI” as a synonym for authentic. That distinction needs to be rethought entirely.​

What Structural Transparency Looks Like

In my view, two assumptions have to go. Transparency can no longer be a one-time disclosure, and it can no longer rest on simply pointing to where AI was used as proof of good faith. Both treat a moving system as if it were a fixed snapshot.

From there, a few practices start to matter more than any label.

The first is disclosing an AI policy at the operational level. Rather than tagging individual outputs, a company can maintain clearly stated policies that tell consumers exactly which workflows use AI, how it is being used and what guardrails keep that use aligned with company values. (Adobe’s Content Credentials is a good example of this.) Harvard Business School research on government found that transparency at the operational level builds trust that lasts, rather than trust that fades after a single disclosure. I think the same lesson can apply here.​

The second is attaching real, named people to statements and decisions, especially where AI was involved, instead of communicating through an anonymous AI voice. As AI becomes ubiquitous, the value of a genuine human voice rises accordingly. People trust people, and they want to see the individuals standing behind business decisions. A comparative study of human-produced and AI-produced news stories found that, despite advances in generative AI, human authorship still holds a clear trust advantage with audiences.

The third is gaining real-time visibility into how the brand is represented and discussed online. As AI-generated content, deepfakes and coordinated bot activity increasingly target and distort brand narratives from the outside, companies need to see how they are being portrayed as it happens. Proactively watching for AI-driven narrative swells, coordinated inauthentic activity and misinformation gives a brand the ability to catch and respond to a reputational threat before it compounds.​

Humans In Control

AI is embedded throughout the content pipeline now, and audiences know it. That’s exactly why genuine transparency can no longer mean flagging individual posts or pieces of content.

It must become a published, standing document (the AI-era equivalent of a privacy policy) rather than a one-off disclaimer.

Brands need to build primary proof of trustworthiness by disclosing their AI policy so consumers understand where, how and why AI is used. They also need to put the real people behind their decisions out front and stay vigilant to how AI is shaping their narrative from the outside.

Narratives can now be hijacked or manufactured in real time, so reputation defense must catch deepfakes and manufactured stories as they happen, not after they’ve spread.

Every one of these moves keeps a human in control while AI does more of the work, and that is what earns consumer trust from here on out.​​

Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?