Tantrums, defiance, growing pains, and confusion. These characterize the infamous toddler era we know as the Terrible Twos. This key developmental phase happens when young children act out and acquire new skills but lack the ability to understand exactly what it is they’re doing. And if ChatGPT were human, it would be in its Terrible Twos. This year, Gemini and Claude will also turn two.

For those living under a rock – or a large boulder – everyone is talking about generative artificial intelligence (gen AI). Gen AI tools, such as ChatGPT, are helping people write anything from parking ticket appeals and sales pitches to freelance articles and cover letters for job applications.

Despite the job insecurities that arrived with gen AI writing and research tools, a recent Adweek survey has revealed that two in three marketers claimed they aren’t worried about losing their jobs to AI.

Just like toddlers are often misunderstood during their Terrible Twos, I’m coming round to realize the extent to which these AI writing tools are also misunderstood. Although companies around the globe are investing hundreds of billions of dollars in gen AI, the hype doesn’t yet seem to be delivering significant positive ROI, brand affinity, or consumer sentiment. 

Here’s why writers like me shouldn’t worried about being replaced by AI – at least for now.

Gen AI imitates but can’t innovate

A toddler will try to imitate what it sees and replicate the words and the tone of voice it hears. That’s why, if you utter f*** a few too many times around your two-year-old, you only have yourself to blame if they then start dropping the f-bomb. 

Young children in their developmental phase also struggle to use the right words and tone at the right time, in the right context – or they overuse certain words within their vocabulary. Gen AI tools, such as ChatGPT, Grammarly, Writer, Copy AI, and Ryter, similarly struggle with tone of voice and diversifying vocabulary – there are in fact now lists of words and phrases circulating on the internet that help you detect AI-generated content. 

These include:

• a rich tapestry
• indelible mark
• valuable insights
• unwavering commitment
• lets us delve
• navigate the challenges

I watched language specialist Nick Parker speak last November at CopyCon 2024, an event hosted by ProCopywriters. He urged during his speech that companies approach AI with caution: “It’s not whether AI can do it. It’s whether, upon learning AI created something, it changes how we feel about it.”

Of course, AI tools can write grammatically sound text. But consumers (especially, native speakers) are already able to identify AI-generated text, and research has identified that the use of AI decreases emotional trust. If companies rely on AI for their creative output – advertising, copywriting, client correspondence – this will likely snowball into decreased trust and lower purchase intention, which may end up hurting a company’s bottom line.

Companies might not care too much about the longevity of their marketers’ jobs, but they do care about the longevity of their brand reputation and revenue. Putting aside budget for human writers and creatives in general to produce the client-facing materials matters when it comes to consumer sentiment and engagement.

Gen AI’s emotions are still developing

A young child undergoes tricky emotional development in its Terrible Twos and isn’t yet able to regulate its emotions effectively. I’m not the first person and I won’t be the last to point out that ChatGPT and other gen AI writing tools also lack sophisticated emotional intelligence.

When I first started my editing and writing career at an agency in 2017, the first lesson I learned was that typos and inconsistency ruin a brand’s reputation. It’s hard to be taken seriously as a brand if you commit these copy crimes. Of course, my colleagues also used to acknowledge that everyone is human, and mistakes slip through QA.

 Now, eight years later, I’m almost relieved to spot typos when I read online copy. It means that a human (a real mind) has crafted the content. Mistakes show our quirks and personalities as humans with emotions. I’m now more likely to take that brand seriously because it values its products enough to understand

a) if AI can write its USPs, these probably aren’t really USPs – AI just scrapes and steals information from other websites
b) only creative human minds can find unique angles that tap into a consumer’s emotions and struggles

A talented writer doesn’t just create grammatically correct sentences; they create worlds. And consumers only buy into worlds if the copy is effective, personal, and researched enough.

There’s also a discrepancy between how quickly it takes to train an AI model and how quickly a human can change their mind. Humans are fickle. We change our emotions and beliefs quickly and irrationally. That’s why companies must also develop and update their products quickly to avoid becoming irrelevant. I’ve worked at companies in the healthcare and tech spaces, where products change almost every week and it’s been critical as a copywriter and marketer to adapt my messaging and copy to respect these changes.

Emotionally immature AI tools like ChatGPT in their current state are not equipped to effectively keep up with many real-time human and product developments. These tools can scrape the consumer sentiment that exists online, through Reddit posts and reviews, and but they can’t scrape the conversations good copywriters have as they work closely with stakeholders, product teams, and customers to ensure the copy always aligns with the latest consumer behavior.

My reassurance to writers

Yuval Noah Harari wrote in his book 21 Lessons for the 21st Century: “It is much harder to struggle against irrelevance than against exploitation.” He was referring to the threat of technology taking over human jobs. 

I empathize with writers and marketers who are scared about losing their jobs – and with anyone in any sector who is nervous about the implications of AI on their job. But my reassurance to writers is that they will always be more effective in writing customer-facing copy because:

• Your emotional intelligence helps create strong brand connections with consumers
• You write based on data, numbers, and on real conversations that can’t always be scraped from the internet
• What you write might be based on inspiration, but it’s not imitation
• You make mistakes, but you don’t hallucinate
• Your carbon footprint is considerably lower

It’s important for writers to remind themselves, clients, and employers that you can and absolutely should leverage gen AI for lower-risk non-client-facing use cases, such as early-stage ideation and brainstorming. Any use case that could sabotage consumer trust or cause customer outrage – as we have witnessed – is where copywriters must keep writing.

Leave the tantrums, defiance, growing pains, and confusion to ChatGPT and friends, and let creatives to do the work while we wait to see if these tools ever come of age.