When AI Came for My Clients, Not My Job

In 2023, Edward Frank Morris made a decision most people would call reckless: he fired himself.

At 33, based in Southampton in the UK, Morris walked away from a stable career in copywriting not because he hated it, but because he could see where it was heading. The future wasn’t looming anymore. It was already drafting emails, writing landing pages, and pitching clients.

By the time most people were still debating whether AI was a “tool” or a “threat,” Morris had reached a blunt conclusion: my clients won’t need me for much longer.

He wasn’t wrong.

A 2025 Generative AI Susceptibility Index later confirmed what early adopters were already feeling. By 2023–24, nearly every UK job showed some exposure to generative AI, with major cities like London seeing 35–36% of roles affected. Across the Atlantic, a Stanford study found that entry-level workers aged 22–25 in AI-exposed roles—customer service, accounting, software development had seen employment drop by 13% since 2022.

This wasn’t theoretical. It was personal.

Today, Morris is the CEO and Lead Prompt Engineer at Enigmatica, a leading UK prompt-engineering and AI consultancy. Speaking to Newsweek, he explained why he chose to “self-fire” and what that decision says about the future of work.

Seeing the signs early

Like many copywriters before the AI boom, Morris had played with early automation tools basic text spinners that rewrote content in seconds.

“Back then it was Jarvis now Jasper,” he says. “You’d give it a brief and it would spit something out almost instantly.”

What set him apart wasn’t just curiosity, but timing. Morris had long been interested in futurism and AI, and he’d been quietly tracking a small research lab most people hadn’t heard of yet: OpenAI.

During the GPT-2 era, he signed up for beta access. Then came November 2022 and everything changed.

When Morris got early access to ChatGPT, he immediately sensed the shift. At first, he used it like everyone else: silly emails, quirky blog posts, messages written just for fun. But curiosity turned into obsession.

“I clicked the ChatGPT tag on Twitter and it was chaos,” he says. “People were building crypto bots, coding full programs from one sentence, making it roleplay historical figures some were doing stuff that was… very questionable.”

The rabbit hole went deep.

When writing became prompting

Morris started spending hours refining prompts tweaking verbs, stripping fluff, adjusting tone. Without realising it, he was translating years of human communication skills into something new: AI communication.

And that’s when it hit him.

“With the right prompts, ChatGPT was already as good as the average copywriter,” he says. “The writing was on the wall. It was only a matter of time before clients skipped me and went straight to AI.”

Since then, the technology has only accelerated. “The leap from GPT-3 to GPT-5.1 is massive,” Morris says. “Especially if you know how to prompt it.”

Today, he scrolls Reddit and sees story after