Brand Trust in the Age of AI: Advice for the CMO

One of a chief marketing officer’s most important jobs is to build and nurture customer trust. Trust is fundamental to the customer-brand relationship, particularly in categories like financial services and healthcare where customers view the brand as a trusted provider of valuable advice. Every brand interaction has an impact on customer trust.

So what will happen if and when advice is provided by artificial intelligence instead of a human financial advisor or a nurse? While financial services customers may be familiar with canned responses from robo-advisors, future AI-powered consultations may be far more unpredictable, due to the nature of AI itself. ChatGPT and similar large language model AIs are a black box: Exactly how they choose what responses to provide isn’t necessarily understood, even by their creators. And their answers may vary over time for reasons similarly unknown.

Let that sink in for a moment.

Marketers carefully tend to customer engagement. That’s why call center representatives have scripts, brand websites offer carefully crafted content, and email and social media experiences are engineered to foster successful customer journeys. All of this maximizes brand trust. What if unlocking the potential of AI allows it to make decisions and take actions that don’t align with your brand?

For high-stakes categories like financial services and healthcare—in which the information dispensed can have profound real-world implications—I think ceding control to AI will be, frankly, terrifying. And not just when it comes to information accuracy. Suppose an AI’s answer is correct but not aligned with the organization’s approach to, say, portfolio management or cancer treatment. You can’t exactly manage AI the way you do a human physician or financial advisor.

In the end, the CMO is responsible for ensuring brand trust across all customer touchpoints that may come to have an AI component.

Additionally, legal and compliance teams will undoubtedly have long lists of concerns about liability, and there will likely be licensing and certifications for use of AI in regulated industries. Near-term implementation of AI will probably stick safely to lower-stakes customer engagements. But in the longer term, it seems inevitable that AI will be used for more nuanced, complex and subjective customer interactions.

A qualified professional represents your organization, and the advice provided is a direct reflection of your brand. If that advice comes from AI, will the brand similarly receive credit (or blame)? Will automation via AI diminish your brand, the way having to navigate an automated phone system may annoy customers?

Or might unimaginably capable AI—efficiently handling inquiries and serving customer after customer with no hint of weariness—actually enhance brand reputation? In a 2022 survey conducted on behalf of Amazon Web Services, 74% of respondents said they “are more likely to spend more with brands that allow them to avoid contacting a live agent.” For demographic segments already predisposed in this direction, AI might be viewed as automation on steroids.

This brings us to the question of business models. Financial services firms, healthcare providers, lawyers and the like charge for advice. Can they continue to do that with AI-driven customer interactions?

Even today, unrelated to AI, there is debate in healthcare about whether providers should bill patients for services delivered via the MyChart app. In other words, is MyChart ancillary to care? Or is it actually care? And what about medical information that could be provided by AI in the future?

Or, alternatively, could we be in an AI “arms race” where customers will make decisions based on the quality of advice provided by AI tools?

Having identified—but not resolved—so many critical AI-related issues, I reached out for help. I asked ChatGPT, “Should CMOs be worried about ChatGPT?”

Its answer: “CMOs may want to consider the potential impact of language models like ChatGPT on their field, as they can generate large amounts of written content quickly and accurately. However, it is important to note that ChatGPT is just a tool, and its effectiveness will depend on how it is used. CMOs can also use it for creating and managing effective marketing campaigns. It could also be used for customer service, creating personalized content and much more. It can also be integrated with other technologies to make marketing more efficient and effective.”

Note how ChatGPT’s response neatly sidestepped the question of whether CMOs should be worried and instead focused on the advantages that ChatGPT offers CMOs—just as an agile human interviewee might have done.

Rather than worry too much about AI, I think CMOs should carefully manage its potential impact on brand trust by incrementally incorporating it into the customer experience. Customer experience teams should begin with one defined area—for example, transaction support—and collect data on customer response to the AI-enhanced experience before rolling out the capability more widely.

In the end, the CMO is responsible for ensuring brand trust across all customer touchpoints that may come to have an AI component. Given the pace of technological change—the next release of ChatGPT will have exponentially more language to draw on than the current model—we can be certain that AI will continue to confront us with challenges our poor analog brains will struggle to comprehend.

A version of this article was originally published in Forbes.