The question “How are you using Artificial Intelligence (AI) in your work?” has been a part of almost every professional conversation I’ve had in recent months. Like a magic wand that instantly makes everything we touch better. But I’ve concluded that we’re asking the wrong question after seeing how various organisations handle the use of this technology.
We should ask why we actually need AI rather than how to use it. This is because the difference between these two questions shows the difference between carefully adopting new technology and just going with the flow.
We have a duty to use this technology wisely as communication professionals, particularly those of us who collaborate with organisations that aim to make a genuine social impact, such as through cultural diplomacy. Not because we should embrace it blindly or reject it on principle, but because we should see how it can enhance our abilities without taking the place of the most important component of our work: human judgment.
A common belief about AI is that it will help us get answers faster and more accurately. Chatbots that respond to queries, algorithms that segment audiences, systems that generate optimised content for each platform. It all sounds efficient and promising.
But this method is predicated on a flawed assumption: that our primary problem is to obtain answers, when, in fact, we need to ask better questions.
The questions we ask in communication strategies determine how well we do our work. Are we communicating to inform or to generate action? Does our audience need more data or more context to be able to interpret it? Does the message we consider “clear” really resonate with the experiences of those we want to reach?
These questions require an understanding of cultural, social and political contexts that none of today’s large language models – such as ChatGPT, Claude or Grok – can fully grasp. For example, they need to know how to deal with complicated human relationships when working with diverse groups or handling touchy issues like countering false information or building trust.
AI can help us process information more quickly, identify patterns in large volumes of data, or translate content into multiple languages. But asking strategic questions, interpreting those patterns, and deciding what to do with that translated information remains irreplaceably human.
Working in communications means constantly translating between different languages, contexts and frames of reference. When an international organisation wants to communicate about sustainable development, for example, the same concept must resonate with European donors, local implementers in Africa and beneficiaries in Latin America. Each audience brings its own cultural references, priorities and ways of understanding social change.
AI can figure out which keywords work best for each audience, which formats get the most interaction, and which times get the most views. It can even suggest adaptations in tone or structure. But the decision about which aspects of the message to emphasise for each context, how to manage tensions between different cultural frameworks or when it is necessary to adjust the entire strategy requires a type of analysis that only comes from human experience.
I have seen projects where the data showed high social media engagement, but qualitative conversations revealed deep misunderstandings about the project’s objectives. Or campaigns that generated excellent metrics but failed to build the trust necessary for long-term work.
Not only does human judgment explain data, it also finds gaps in that data. It recognises when a high metric may be masking a deeper problem, or when an unexpected response may signal an unanticipated opportunity.
In multicultural contexts, this judgement becomes even more important. AI can detect and analyse why a message is not working in a certain region, but only human experience can interpret it with rigour. The problem may lie in the language, cultural references, the political moment or more subtle social dynamics.
How to preserve openness and, consequently, audience trust is one of the things that worries me the most about the use of AI for strategic communication.
When organisations use algorithms to segment audiences or personalise messages, are they being transparent about those processes? Do communities understand how decisions are made about what information they receive and when? Is there clarity about what data is collected and how it is used?
When we work with groups that have been harmed by manipulated information or where trust in institutions is weak, algorithmic opacity can destroy relationships that have been built over years. It is not enough for algorithms to work well; people need to understand how they work and have some degree of control over that interaction.
This does not mean revealing all technical details but rather developing accessible ways to explain how technology influences the communication they receive. It also means creating mechanisms for audiences to provide feedback on these processes and, where possible, opt for alternatives.
Algorithmic transparency also means honesty about limitations. If an AI system helps translate content, are we clear about what cultural or contextual aspects may be lost in translation? If we use algorithms to analyse sentiment on social media, do we acknowledge that they may not capture irony, specific cultural references or local power dynamics?
Maintaining this transparency requires human judgement to decide what information is relevant to share, how to communicate it in an understandable way, and how to integrate feedback to improve processes.
The ability to make messages more relevant to each person is one of the most appealing promises of AI in communication. Imagine being able to automatically adapt the same concept for different audiences: more technical language for specialists, more emotional approaches for beneficiaries, and more pragmatic frameworks for donors and implementing agencies.
This capability is genuinely valuable, but it comes with significant risks if not handled carefully. The main danger is fragmentation of the core message to the point where different audiences receive conflicting information or develop incompatible understandings of the same project.
I’ve witnessed instances where automatic adaptation caused beneficiaries to perceive a project in one way, implementers in another, and donors in still another. Technically, each audience received information “optimised” for their interests, but the result was a loss of coherence that complicated coordination and created mistrust between groups.
It’s important to use AI to enhance human adaptation rather than to replace it. Algorithms can suggest adjustments to tone, format, or emphasis, but the decision about what to adapt and what to keep constant must remain strategic and human.
This means clearly defining which elements of the message are non-negotiable—the core values, objectives or commitments that must remain consistent—and which aspects can be made more flexible for different contexts. It also involves creating mechanisms to verify that adaptations are not creating confusion or unintended contradictions.
In my experience, the best adaptations arise when technology informs but does not determine strategic decisions. AI can identify which words resonate best with different audiences, but the strategic communicator decides whether using those words is consistent with the broader objectives of the project.
The orchestra conductor metaphor helps me think about how AI will affect our profession in the future. It’s kind of like having better instruments or musicians, but someone still has to choose which symphony to play, when to speed up or slow down, and how to make the different parts sound good together.
The value of the strategic communicator is not in executing tasks that can be automated, but in exercising the judgement that allows those automated tasks to generate coherent and strategic impact. This includes deciding when to use technology and when not to, how to interpret the results it produces, and how to integrate those results into broader strategies for building trust and engagement.
It also means developing new skills. We need to understand enough about how these systems work to use them responsibly, identify their biases and limitations, and communicate transparently about their use. But we don’t need to become programmers; we need to remain strategists.
The most sophisticated AI in the world cannot replace the human ability to sit down with a community, listen to their real concerns, understand the power dynamics that influence how they interpret information, and design communication processes that build genuine trust and sustainable engagement.
What it can do is free us from routine tasks so we can spend more time on those necessary conversations, help us process feedback more systematically, and suggest connections or patterns we might not have seen.
In this future, professional success will not be measured by our ability to compete with machines, but by our ability to lead processes where technology amplifies our most valuable human capabilities: strategic judgement, cultural sensitivity, building authentic relationships, and the ability to ask the questions that matter.
Ultimately, AI will only be as good as the questions we ask it, and as useful as the judgment we apply to its answers. And those remain profoundly human capabilities.
How are you managing the balance between leveraging AI and maintaining human judgment in your work? I would love to hear about your experience. Connect with me on LinkedIn to continue this conversation about the future of strategic communication.
The EU AI Act for communications teams is about more than learning new tools. This guide explains the 2026 requirements and how to turn them into clear rules for use, review and transparency.
A decision may be well-founded yet still come across as poorly explained. This happens in a call for proposals where many organisations are left out.
A campaign can carry a responsible message while being distributed through an experience that pressures people. Trust is also designed.