AI agents can interpret and connect data, but they cannot guarantee trust. As automation grows in publishing and ad tech, new standards are needed to ensure accountability and authority.
As AI agents take on more autonomous roles in digital publishing, advertising, and content operations, the question of trust is moving to the center of industry debate. While large language models (LLMs) can now interpret context, map taxonomies, and bridge data gaps across systems, they cannot independently verify the legitimacy or authority behind the data and actions they process. For media and publishing professionals, this means that as AI-driven agents begin to manage workflows, budgets, and audience targeting, the need for robust standards and protocols is not going away-in fact, it is becoming more urgent.
Historically, standards like schemas, taxonomies, and APIs were created to help software handle ambiguity and enable communication between systems. Today, LLMs can infer meaning from incomplete or inconsistent data, translating between different definitions and structures. This technical progress has led some to argue that traditional standards are less important. However, the core function of standards has always been to establish trust and coordination between partners, not just to help machines understand each other.
AI can act as semantic middleware, interpreting and connecting disparate systems. But it cannot confirm whether data was authorized, whether a signal came from a legitimate source, or whether an action stayed within the boundaries of delegated authority. These are trust issues, not language problems. As agents shift from supporting human decisions to making decisions themselves, the inability of AI to vouch for the legitimacy of actions becomes a critical gap.
Current trust infrastructure-such as identity, consent, fraud detection, and measurement systems-was designed to answer narrow questions about access, permission, and validity. The rise of agentic automation introduces a tougher challenge: ensuring that every action remains traceable to a legitimate authority. This matters because AI can generate behavior that looks authentic, but appearance alone does not guarantee accountability or proper authorization.
As agents begin to negotiate deals, allocate budgets, and optimize campaigns, trust must be evaluated at the moment of action, not just in post-campaign audits. This shift requires standards that address provenance, permission, delegation, authority, continuity, auditability, and economic accountability. The standards of the next decade may focus less on data formats and more on mechanisms that establish and maintain trust in real time.
Media markets already rely on signals that claim to represent attention, intent, and value. AI will make it easier to generate signals that appear valid but may lack a clear connection to accountable origins. This is why standards must move into the decision-making layer, ensuring that every automated action can be traced back to a responsible principal. As highlighted in a recent analysis of how Stagwell is building transparency into its AI-powered ad marketplace, the industry is already seeking new ways to verify and validate digital transactions.
The assumption that better AI reduces the need for standards may prove false. As inference capabilities grow, so does the importance of trust infrastructure. The future of digital media will depend on systems where AI handles interpretation and adaptation, while standards govern authority, accountability, and trust. The next era will not just standardize information or transactions-it will standardize trust itself.
