AI is changing how quickly enterprises can make decisions across the customer journey, but faster decision-making is creating new questions about CX quality and accountability.
In fact, recent TheyDo research found that 68% of enterprise decision-makers believe AI has made decisions faster and more confident, yet 38% say AI-driven decisions have also created unintended CX problems.
To ensure they remain customer centric, CX leaders must assess AI by the quality and customer impact of the decisions it supports over speed or automation alone.
Jochem van der Veer, CEO at TheyDo, told CX Today that, whilst effective, AI does not remove the need for human accountability over its decisions.
“AI has compressed the time between the question and the answer," he said.
"But it hasn't removed the responsibility for what happens next.”
Speed Without the Full Picture
The growing use of AI in decision-making is creating a growing gap between confidence and quality, revealing that faster, more assured decisions do not always translate into better customer outcomes.
In fact, the Decision-Confidence Gap report revealed that 51% of businesses have seen complaints or support queries increase from AI-driven-decisions, with 42% experiencing a decline in customer trust.
These stats are reportedly driven by the fact that 77% of leaders say they now feel expected to make decisions more quickly because AI enables them to do so, creating a tendency to treat speed as evidence of progress.
Whilst AI can accelerate a decision without providing broader context on its potential impact across the customer journey, that context becomes particularly important when decisions are made within fragmented organizations.
“Faster and more confident doesn't mean better,” Veer highlighted.
“Most enterprises don't necessarily have an AI quality problem. The models are pretty good. But they have a decision-making problem.”
Teams using AI to optimize individual touchpoints may likely create friction in another area of the journey, meaning AI may identify a genuine change in a metric, but the response can still be wrong if teams lack visibility into the factors driving that change.
This reflects a wider concern that AI can produce an accurate answer based on the data available while still delivering a poor outcome if it lacks the broader customer context
Fragmentation can make the consequences of errors harder to contain as the issue increases across an organization.
“The risk isn't necessarily scaling AI," he explained.