Prompting: A New Job Opportunity in the Age of Generative AI

In the age of generative AI, opportunities are emerging for workers in the form of prompter jobs. With the launch of OpenAI’s ChatGPT last November, the field of generative AI has exploded, leading to rampant speculation that AI will take over the majority of jobs.

However, this negative outlook has given rise to new employment possibilities as businesses and organizations seek to harness the power of language models to automate tasks and increase efficiency. As a result, prompter jobs are becoming an attractive option for those seeking employment in the field of AI.

“Thanks to ChatGPT, prompt writing is now a highly valued skill. A tech company is offering $300,000/year to hire a prompt writer,” stated Dr. Mushtaq Bilal, a researcher at the University of Southern Denmark.

The ongoing layoffs at tech companies and the integration of AI into their products also indicates the potential of prompt writing as a profession. However, Bilal argues that “most academics don’t know much about prompt writing.” In a detailed Twitter thread, Dr. Bilal explained prompting as “writing a prompt that gets you your desired result.”

“Writing a really great prompt for a chatbot persona is an amazingly high-leverage skill and an early example of programming in a little bit of natural language,” said Sam Altman, co-founder of OpenAI, who built the ubiquitous ChatGPT and DALL-E.

Perfect prompting

Providing an appropriate prompt to an AI is crucial in obtaining the desired results, as a brief or random input may not result in a convincing output.

“In the article ‘The Case For The AI Prompt Engineer,” we explored the basis of why some well-constructed prompts will elicit responses from the AI Hidden Layers,” wrote Brian Roemmele, Editor-in-Chief at research firm Multiplex.

Roemmele stated that sharing a video example of results from the prompt he invented will “help understand the way GPT is weighting some sides to a debate.” While using such a “powerful prompt,” he found “many novel ideas over the years” as he showcased a comprehensive debate script with his prompt.

“Just as powerful is the way GPT has established the “easy” path to responses and the knowledge it had but “forgot” that usually does not show up in one short question and answer typical prompts,” stated Roemmele.

Ethan Mollick, a professor at The Wharton School, argues that he is suspicious of the claim that “prompt engineering” is a big deal.

“I have a strong suspicion that “prompt engineering” is not going to be a big deal in the long-term & prompt engineer is not the job of the future. AI gets easier. You can already see in Midjourney how basic prompts went from complex in v3 to easy in v4. Same with ChatGPT to Bing,” stated Mollick.

Prompting engineering is “about clearly communicating exactly what you want,” said Simon Willson in response to Mollick.

Starting with a simple prompt and gradually increasing the level of complexity can lead to better results. On the other hand, if a complex task is presented to ChatGPT right from the beginning, the results may lack precision and nuance, explained Dr. Bilal.

This article is originally from MetaNews.