When OpenAI launched ChatGPT, many felt that AI and robots were taking over the world, an overused sci-fi trope. 

Experts say generative AI could be one of the most revolutionary AI technologies in the history of humanity. But every new technology comes with its fair share of ominous doom in public discourse — a threat of replacing humans. Generative AI, however, served us with a hot new job — prompt engineer. 

“​​Prompt engineering is a relatively new profession created due to the widespread adoption of generative AI,” says Kais Zribi, general manager for the Middle East and Africa at Coursera. Prompt engineering is one of this year’s three new and developing vocations, according to the World Economic Forum.

Experts say they are crucial in generating high-quality outputs from generative AI models. According to data from LinkedIn, the number of job listings mentioning GPT has increased by 51% from 2021 to 2022, with recruiters worldwide seeking AI-related skills and workers responding in kind.

WHAT’S THE JOB?

“They are responsible for ensuring that the content generated by an AI model is accurate and relevant. The job of prompt engineers will be critical as companies navigate the impact of this new technology,” Zribi adds.

Abu Dhabi-based 28-year-old Hamza Ali’s job entails creating text-based instructions that he inputs into the back end of AI tools so they can perform tasks like starting a blog post or pitch email with the appropriate tone and factual information, among other things. 

Interestingly, he doesn’t need to create any technical code to accomplish this. Instead, he types a few instructions to get the desired results. 

This is how prompt engineering works. For conversational AI systems like ChatGPT, it is about creating and refining prompts. By using these prompts, AI models perform better, and consumers have more productive conversational experiences. Prompt engineering resembles quality engineering for AI chatbots in certain ways.

HOW TO BECOME A PROMPT ENGINEER?

And people want to learn this new skill as it has a first-mover advantage. According to a Bloomberg report, the prospective salary range for this position is $335,000. 

So, how can one become a prompt engineer? 

“To become a prompt engineer, one should start with a degree in computer science and have a solid foundation in Natural Language Processing. Deep understanding of language and linguistics, programming skills like Python, Java, and familiarity with machine learning frameworks like Keras,” says Ankit Satsangi, founder and chief technology officer at AHAD.

One must be creative and analytical when creating the most effective prompts. Analyzing and spotting response patterns is frequently required, but thinking outside the box can also help develop fresh ideas. Many online courses today are designed expressly for well-liked huge language models like Microsoft’s Bing and OpenAI’s ChatGPT.

FUTURE OF PROMPT ENGINEERING

Since AI programs are becoming more adept at anticipating user needs and creating prompts, it is unclear whether prompt engineering will be important in the long run. 

In a LinkedIn article, Amy Daali, Founder & CEO at Lucea AI, says, “As AI continues to become more integrated into our daily lives, the demand for skilled prompt engineers will only continue to grow. And as generative AI evolves and advances, experts will always need to optimize and improve chatbots’ performance.”

“By adding prompt engineers to the generative AI design cycle, we are keeping the Human-In-The-Loop (HITL) and ensuring that AI systems are developed ethically and responsibly,” she adds.

However, as base models advance and gain the capacity to transform subpar prompts into superior ones, AI can automate the engineering of prompts. “But other jobs related to generative AI integration in business operations, such as Chief AI Ethics Officer, will become increasingly important since the platforms can easily be used for misinformation and propaganda,” says Zribi.

“Generative AI will change jobs as we know them, and jobs that once required content creation, for example, will focus more on guiding, editing, and fact-checking outputs from a generative AI partner,” he adds.