Artificial intelligence is already part of the data workflow. It helps us explore datasets, detect patterns, generate visualizations, summarize findings and draft narratives in a matter of seconds. For any professional who works with data, communication or decision-making, this is no minor improvement. It is a structural shift.
But the more powerful these tools become, the more important it is to ask a key question:
AI can create the chart. But can your audience trust it?

This question lies at the heart of one of the major challenges facing data storytelling today. We are entering a stage in which producing charts, dashboards and explanations will become ever easier, faster and more automated. Yet generating a visualization is not the same as communicating meaning. And producing a narrative is not the same as building understanding.
For years, much of the conversation about data visualization has focused on choosing the right chart, reducing visual noise, using color more effectively, organizing information and making data more accessible. All of that still matters. But in the era of generative AI, the challenge shifts: it is no longer just about visualizing better, but about designing interpretation processes that people can trust.
Speed does not guarantee meaning
AI can speed up many tasks: suggesting visualizations, summarizing trends, turning data into text, generating headlines or helping us explore hypotheses. This opens up clear opportunities. It can democratize work with data, lower technical barriers and allow more people to take part in evidence-based conversations.
But it also introduces a risk: confusing speed with understanding.
A tool generating a chart in seconds does not mean that chart is relevant. A model producing a fluent explanation does not mean that explanation is correct. A narrative sounding convincing does not mean it is well founded.
AI can accelerate production, but meaning requires something more: context, interpretation, editorial judgment and an understanding of the audience.
This question is becoming a central part of my doctoral research and of the book I am preparing on Data Storytelling with AI: how we can use artificial intelligence to expand our ability to analyze, visualize and communicate data without losing the human judgment that makes stories meaningful, responsible and reliable.
The risk of synthetic confidence
One of the most significant risks of generative AI applied to data work is what we might call synthetic confidence.
Models can produce highly fluent answers, seemingly professional visualizations and explanations that sound authoritative. But fluency is not understanding. Formal clarity is not truth. And a confident explanation is not always backed by solid evidence.
A visualization can look flawless and be based on incomplete data. A recommendation can sound reasonable and fail to account for the organizational context. A narrative can emphasize a trend and hide the uncertainty that should accompany it.
That is why, in an AI-mediated environment, critical thinking is not an add-on. It is a core competency. We need to ask ourselves more rigorously: What are the data really showing? What are they not showing? What assumptions lie behind the analysis? How much uncertainty is there? What alternative explanation could this pattern have? What decision might this story influence?
AI can help us think, but it should not think for us.
From data visualization to decision confidence
I believe we need to broaden the conversation. For a long time we have talked about data visualization, dashboards and data storytelling. But in an increasingly automated environment, we need to move toward a more demanding idea: decision confidence.
Decision confidence is not just about someone believing a chart is well made. It is about being able to trust the reasoning that connects data to interpretation, interpretation to narrative, and narrative to action.
A good data story does not only answer the question “What do the data say?” It must also answer a more important one: “What does this mean for us?”
That is where the real value of data storytelling emerges.
Data do not speak for themselves. They need context, a question, an audience and a purpose. A dataset does not know an organization’s culture. A chart does not know the tensions behind a strategic decision. A model does not automatically understand the expectations of a leadership team, the resistance within a team or the human implications of a transformation.
That is why human context is no minor detail. It is what turns information into meaning.
The new role of the data storyteller
If tools can generate charts, dashboards and text more and more automatically, what will the role of the data storyteller be?
My answer is clear: it will be more important, not less. But the role will change.
The data storyteller will no longer be seen only as someone who “makes charts” or turns data into presentations. The role will be more strategic: acting as a translator between data, technology, context and decision.
Data storytellers will need to know how to work with AI tools, but also how to question them. They will need to speed up processes while protecting meaning. They will need to generate narratives while preserving editorial responsibility. They will need to help organizations not just produce more information, but better understand what that information implies.
In this sense, data storytelling is moving ever closer to communication for decision-making.
- Visualization helps us see.
- Narrative helps us understand.
- Trust helps us decide.

Designing for trust
Designing for trust means making the reasoning visible. Showing a conclusion is not enough; you have to help the audience understand how it was reached.
It means explaining where the data come from, what their limitations are, which analytical decisions were made and how much confidence we can place in the conclusions. It means avoiding overstated claims, not presenting as certain what is only a hypothesis, and communicating uncertainty when it matters.
AI can be a powerful ally in this process. It can help us explore alternatives, build prototypes, synthesize patterns and improve the clarity of an explanation. But the final decision about which story to tell, how to tell it and with what level of responsibility remains a human one.
The future of data storytelling will not be defined only by more powerful tools. It will be defined by better questions, better interpretations and better communication decisions.
Because the value of a data story is not measured by how impressive a visualization looks. It is measured by its ability to help people understand something relevant, discuss it intelligently and make better decisions.
AI can create the chart.
But trust still has to be designed. Artificial intelligence is already part of the data workflow. It helps us explore datasets, detect patterns, generate visualizations and draft narratives in a matter of seconds. It is a very powerful shift, but it also raises a fundamental question: when machines can produce charts and explanations almost instantly, what is the real value of data storytelling?
For me, the answer is becoming increasingly clear: trust.
This question is becoming a central part of my doctoral research and of the book I am preparing on Data Storytelling with AI: how we can use artificial intelligence to expand our ability to analyze, visualize and communicate data without losing the human judgment that makes stories meaningful, responsible and reliable.
AI can accelerate production, but meaning does not come from speed alone. Meaning requires context, interpretation, editorial judgment and a clear understanding of the audience. A chart can be technically correct and still fail to explain what really matters. A narrative can sound convincing and yet conceal weak assumptions, missing evidence or overconfidence.
That is why I believe we need to move from thinking only in terms of data visualization to thinking in terms of decision confidence as well. The value of a data story lies not only in people understanding the chart, but in their being able to trust the reasoning that connects data to interpretation, interpretation to narrative, and narrative to action.
In an AI-mediated environment, this demands more than technical skill. It demands data literacy, AI literacy, critical thinking and communicative responsibility. We need to ask better questions: What is the data actually showing? What context is missing? What assumptions are we making? What uncertainty should we communicate? Which decisions could this story influence?
AI can help us work faster. It can support exploration, synthesis, prototyping and communication. But responsibility for meaning remains human.
The future of data storytelling will not be defined by better tools alone. It will be defined by better judgment.
AI can create the chart. But trust still has to be designed.