Will copywriters become prompt engineers?
ChatGPT, Jasper, Copy.AI, and dozens of other AI writing tools are vying for everyone’s attention, from marketing managers to university students. But copywriters have been watching with an air of anxiety — spurred on by articles such as The Guardian’s “I’m a copywriter. I’m pretty sure artificial intelligence is going to take my job” and the Washington Post’s “ChatGPT took their jobs. Now they walk dogs and fix air conditioners.”
Some copywriters, content marketers, and freelance writers have redefined their role as “prompt engineers,” a term coined in light of the increasing prominence of advanced language models such as ChatGPT. However, the question remains: is this a valid and sustainable career path, or merely a passing fad capitalizing on the AI frenzy?
The birth of ChatGPT whisperers
Copywriting is the art and science of creating persuasive and engaging content for various media platforms. Over the years, the role has changed significantly due to the evolution of technology, consumer behavior, and marketing trends. Now, the copywriting world is about to undergo another transformation with the explosion of generative AI.
But some forward-thinking writers have been looking for other roles they can branch into. One such career path is prompt engineering, or the more intriguingly named, ChatGPT whisperers. AI ‘prompt engineer’ jobs have been popping up online, paying up to $375,000 and implying you don’t need a tech background. But what is prompt engineering, anyway?
What is prompt engineering?
Prompt engineering is creating and optimizing prompts for natural language processing systems. And the ‘prompt’ part is a text input that guides the system to produce a desired output. Prompt engineering involves choosing the right words, format, and context to elicit the best performance from the system.
A prompt engineer designs and develops prompts for conversational agents, such as chatbots, voice assistants, and virtual assistants.
Some of the skills needed to be a prompt engineer are:
- Natural language understanding: the ability to analyze and interpret natural language inputs from users and extract relevant information and intents.
- Prompt design: the ability to craft prompts that are clear, concise, informative, and conversational, and that elicit the desired responses from users.
- Prompt evaluation: the ability to test and measure the effectiveness of prompts using various metrics, such as user satisfaction, task completion, error rate, and conversational quality.
- Prompt optimization: improving and refining prompts based on feedback, data analysis, and best practices.
A prompt pathway
So, how do you become a prompt engineer in 2023? Although some articles claim that little to no technical knowledge is needed to become a prompt engineer, most positions require some specific technical skills. Thankfully, there are plenty of resources to get you started with prompt engineering, such as this Beginner’s Guide to ChatGPT Prompt Engineering.
Additionally, a number of prompt engineering courses have emerged, such as Prompt Engineering for ChatGPT, but there are also other free prompt engineering resources available.
Will a prompt engineer’s career be… prompt?
Some experts argue that prompt engineering will be important in the near future, while others claim that it will be obsolete as AI models become more intuitive and self-generating.
In truth, prompt engineers are not likely to disappear in the future, but rather evolve and adapt to new challenges and opportunities. As prompts become more complex and diverse, “ChatGPT whisperers” must develop new skills and knowledge to keep up with the changing demands. They will also need to collaborate with other professionals, such as content writers, data analysts, and user experience designers, to create prompts that meet the needs and expectations of different audiences and contexts.
I haven’t read about many copywriters who became “prompt engineers” outside of a few job sites.
Bots will only get better at understanding natural language. I don’t see this as a long-term career for people without real computer engineering skills.