The battle for brand trust in a world of Agentic AI
By Ash Diffey (pictured), Vice President Australia and New Zealand, Ping Identity
As Artificial Intelligence (AI) accelerates into the corporate mainstream, a profound shift is reshaping the foundations of business identity and brand trust.
Once the specialty of marketing campaigns and customer relationships, brand identity is increasingly being filtered – and in many cases determined – by autonomous AI agents.
Recent reports note that half of business executives plan to deploy AI agents this year, with adoption expected to surge to 82% within three years. These agents are tasked with recommending purchases, authorising transactions, and determining which companies and products are even seen, often without human oversight.
For companies accustomed to shaping perceptions through advertising, brand campaigns, and customer engagement, this represents a seismic change.
AI agents are indifferent to traditional marketing muscle, loyalty programs or carefully crafted narratives. Instead, they respond to structured data, metadata, and clean signal-to-noise ratios.
Algorithmic probability
Historically, customer interactions have been rooted in human-led interfaces such as login pages, loyalty cards, or personal consultations. Identity has always had a face, however, thanks to the rise of agentic AI, this is changing.
Consider a CTO seeking a new CRM platform. Rather than comparing a dozen websites and vendor proposals, the CTO’s AI assistant can scan structured reviews, analyse language patterns, check for trusted metadata, and instantly produce a recommendation.
However, in this scenario, the software vendor’s brand has been reduced to a weighted probability in an algorithm’s decision matrix. There is no narrative or no emotional pitch – only data.
This raises difficult questions for boards and executives. What happens when AI agents are manipulated or gamed by bad data? Who bears liability if an AI makes a decision based on false or biased information? What are the ethical implications when an AI assistant’s only loyalty is to efficiency?
The future of brand identity
Beneath these questions lies a larger challenge: the erosion of control over digital identity.
In fact, a recent survey found that only 11% of Australians, among the lowest of any country surveyed, have “full trust” in the organisations that manage their identity. Trust is eroding quickly, with 82% of Australians reporting more concern about personal data security than they felt five years ago, alongside growing demand for stronger authentication and tighter government regulation. The dilemma now is that in the era of agentic AI, identity is no longer what a company says it is but rather what machines interpret it to be.
And machines are notoriously indifferent to intent. They do not care if a document was supposed to be updated or if a brand’s positioning has changed. Worse, they are vulnerable to adversarial prompts, spoofed data sources and biased training inputs, making misrepresentations and hallucinations alarmingly common.
In this environment, the winners will not be the loudest marketers. They will be the ones who are most machine-readable, structured, and verifiable.
The role of trust
Identity has always been rooted in trust and, by extension, ethics. Humans are inclined to trust what they can see, hear, and verify. AI agents, however, do not share this instinct.
As intent becomes synthetic, action automated, and consequences ambiguous, the ethics of digital identity can no longer be left to chance. This is an area that requires urgent and significant attention.
Existing governance systems, built for human-led decision-making, are ill-equipped for this new paradigm. What is required is a new layer of ethical mediation embedded within digital identity frameworks.
What business leaders can do
While the coming transformation may feel daunting, there are practical steps executives can take to prepare for the agentic era.
First, begin provisioning AI agents as if they were users. This means giving them their own identities, credentials and policies. Treating agents as users will make it easier to track their behaviour and enforce ethical guardrails.
It is also vital to continuously monitor behaviour. Risk and trust are dynamic, so organisations need to keep watch on patterns from both human and machine actors in order to detect anomalies early.
Leaders should ensure that systems have kill switches built in. If an AI agent malfunctions or behaves inappropriately, the organisation must be able to revoke its access immediately.
At the same time, company identity must be structured for machines as human-friendly branding alone is no longer enough. Identity should be underpinned by structured data, metadata and machine-readable documentation so that agents can interpret it accurately.
Finally, every decision made by an agent should be auditable. Each action must leave a visible, explainable and reversible trail, allowing organisations to treat the decisions of AI agents with the same scrutiny and accountability as those of human staff.
The new era of identity
Identity is no longer simply about who logs in. It is about who acts, on whose behalf, and with what authority. The business world is witnessing the emergence of a new era of digital identity where trust must be designed into systems from the start.
Companies that embrace this shift will be better positioned to thrive in an AI-driven economy. Those that fail to adapt risk disappearing into the noise of machine decision-making.
