The AI ‘boom’ has seen an ever-growing number of businesses and organisations using AI to assist with their communications. Artificial intelligence is no longer just an easy way to create some comical photos or suggest how you should redesign your living room. It’s now a bona fide tool increasingly trusted to help create content or generate strategies, ideas and plans.

Whilst it is, without question, an incredibly valuable addition to any team’s arsenal, this article argues that it can’t – and shouldn’t – be relied upon for strategic planning or content creation.

As quick and efficient as AI tools can be, there are limitations in the material they can produce when compared to the knowledge and skillset of a human. It would be better to frame AI systems as a useful tool for experts to utilise, rather than an engine that can replace us (at least, not yet).

AI can be used to enhance or tweak an idea – but it takes an expert to understand when human judgement must take over. The ‘strategy’ in communication still comes from human insight, expertise and experience.

Here then, are five things AI can’t do within communications – proof that, in an increasingly digital world, communications professionals are still as essential as ever.

 

1. AI can’t understand or replicate genuine emotion

Whilst AI is excellent at quickly pulling things together in a structured fashion from various sources, it lacks the underlying emotion of a human writer. A computer cannot ‘feel’ the way we can or apply the sort of emotional intelligence within a piece of writing that’s most likely to appeal to a certain audience or demographic.

Whilst AI can prove useful in terms of applying a ‘learned’ emotional perspective, that is only because it has been programmed to provide that response. It doesn’t feel, it doesn’t have its own experience to draw upon and can’t provide its own emotional outlook. You can ask it to provide an empathetic note based on criteria and context you give it, but it doesn’t understand how to be empathetic itself – which means it won’t necessarily hit the mark and there’s a good chance it will miss important nuances or get things wrong.

There’s a very real difference between AI ‘slop’ and what comes from experience, understanding and the heart.

 

2. AI struggles to produce experience-based context

When a human writes, we do so with a clear idea of what we believe would be most impactful based on our lived experiences. We each put pen to paper (or fingers to keyboard) in a way that is unique to us, based on how we have been shaped to that point as content creators and, as communications professionals, what we know our audience needs.

Whilst AI learns from what people put into the machine, it doesn’t live those experiences – and therefore communications professionals will always be a step ahead in being able to produce ‘rich’ and engaging content.

Communications professionals – and humans – have an insight and intuition for situations and scenarios based on their experiences. We learn from what we’ve tried and done previously to get better and to understand what would work in a similar situation again. It’s these nuances and our ability to predict and recognise behaviour and develop understanding that sets us apart from AI systems.

 

3. Humans are creative in a way AI just can’t be

Similar to the way in which humans have lived experiences, people are able to think outside the box and really let their imaginations run wild when it comes to writing or being creative. An AI chatbot can only be as creative as the prompt it is fed and the information it is able to search. It can’t come up with a unique or thoughtful idea of its own volition. AI itself said it can create ideas that seem unique by combining multiple complex ideas but it doesn’t have a ‘volition’ and everything it knows is learned or are patterns. It doesn’t have ‘independent, self-driven’ invention.

AI looks at what has been done previously and what is readily available to provide answers and suggestions that meet the ask being presented. It is only humans who can think outside the box and therefore inspire new trends and ideas that simply wouldn’t and couldn’t have been borne of AI.

An experienced team of comms experts can use AI to provide suggestions, proofread or provide feedback on an idea – but human judgement and natural creativity cannot be replicated.

 

4. AI struggles with producing the right intent – you must get your prompts exactly right to illicit the correct response

If you’ve ever tried cracking a joke to AI – or tried to introduce it to sarcasm – you might have found the concept went straight over your device’s head (no pun intended!) It needs to be programmed to recognise it because AI struggles to understand the intent of writing or ideas unless you give very specific context. Only with complete prompt clarity will it understand the right way to act or behave.

There could be a slight nuance in something you ask or need and, if AI fails to pick that up, it may give you an answer that is entirely different to what you need or expect. If you’re looking for a unique way to describe a new car you’ve invented and ask it for a description, it might just tell you that it’s a piece of metal with four wheels. Unless you are clear and precise in what you ask – ‘give me a playful product description for an innovative car that is the first of its kind to utilise laser beams’ in this instance for example – you might find it falls way below the mark.

It can also take things incredibly literally, so using language devices like metaphors without any context can confuse the system. Without relevant grounding and context, the intent of an answer might be way off the mark.

 

5. AI may have plenty of sources, but that doesn’t make it reliable

The basis of everything produced by AI is data and information, not all of which is necessarily reliable. AI engines use crawlers to find information from various websites to compile what they are led to believe is a correct answer, or the right content for you based on information assumed to be accurate.

That means that whilst it can present a strong argument for any point, the knowledge that it provides could come from untrustworthy sources. If enough websites shout untruthfully that Keir Starmer is the current King of England for example, AI could soon accept it as the likely truth and present people with that answer when prompted.

It’s therefore key that you double check the sources used by AI and whatever output it gives you. Even when turning to AI to provide a basic comms message, for example, it’s important to check for nuances. It can present something that at surface level seems reasonable but will still need communications expertise to ensure the content is relevant, accurate and presented in a way that matches your standards, as well as the ethos and values.

 

A summary on AI communications

Although AI is an incredibly helpful technology as a guide and ideas shaper, among other things, we should never fall into the trap of believing it can do it all. It lacks human authenticity, empathy and understanding, so should be approached with the appropriate level of caution.

Some high-profile cases already exist where organisations have fallen foul of AI’s charms. One example is Klarna, which decided to replace 700 members of staff with an AI-based system. Expecting efficiency and a quick-fire solution to customer service struggles, the payment giant instead had to reverse its recruitment decision. And why? Because AI couldn’t grasp the subtle nuances and emotional clarity needed with each unique case.

When personal understanding and human judgement was really needed, AI failed to deliver. Complaints surged, customer satisfaction plummeted and frustration grew. Ultimately, Klarna pivoted back to real people in an effort to restore customer trust and brand reputation. The use of AI in their communications didn’t work.

As communications professionals, we have expertise that has been developed over years of hard work and knowhow gained from different situations. Whilst AI can assist us and improve efficiency,it shouldn’t be a replacement for our ideas, sentiment or intuition. The judgement, creativity and emotional intelligence of professional communications still needs communications expertise.

These are elements that a machine cannot replicate and remains proof, if it were ever needed, that although the robots might be coming, they’re not ready to replace us yet.

To read more of our blogs, you can click here.

Share