Explained: What Is Prompt Engineering, AI Role Sam Altman Is Betting On for Future?

OpenAI CEO Sam Altman in a recent interview discussed the future of the global job market in an AI-dominated world. Responding to a question from Varun Mayya, founder of the Aeos Group, about the future of jobs, Altman stated that entirely new categories of jobs, previously unseen, will emerge.

Altman described prompt engineering as a primary example of an entirely new job created due to AI.

“In terms of entirely new jobs, the first example we saw of that was Prompt Engineer. Before AI came along, there was no such thing as a prompt engineer; not only that, it would have been hard to conceptualise that there would be such a job,” he said.

What Is Prompt Engineering?

Prompt engineering is the practice of structuring or crafting natural language instructions, known as prompts to guide generative AI models toward producing the best possible outputs across text, image and audio modalities.

It is the practice of guiding an AI model through commands, context and constraints to a customised output. It includes the selection of words, phrases, examples and context to align model responses with user requirements, and can range from simple query phrasing to complex multi-step prompting strategies.

For instance, in order to make an AI model write a poem, the individual shall command the model to “write a poem,” assign the word limit and rhyme scheme (constraint), and instruct it about the “topic, genre and metre” of the poem (context).

History of the Practice

Prompt engineering as a practice originated in the early days of generative AI, when researchers proposed crafting diverse NLP tasks as question-answering problems using unified prompts. The practice gained some attention with the launch of GPT-3 in 2020. However, due to the model’s limitations, it remained largely confined to developer circles.

The viral release of ChatGPT (GPT-3.5) in late 2022 made prompt writing an essential skill for generating coherent and context-aware responses from LLMs. This marked the first time prompt engineering was recognised as an actual job.

As newer, more advanced LLMs entered the market, the initial excitement for prompt engineering jobs dried up. While there was demand in the market, consumer enthusiasm for the role reduced.

This sentiment changed recently when OpenAI released its latest image-generation model, which reignited debates about AI displacing human jobs. The argument that a well-crafted prompt can generate outputs comparable to those of a highly skilled designer resurfaced, boosting consumer interest in prompt engineering as a viable profession.

As per data from Grand View Research, the global prompt-engineering market was valued at $222.1 million in 2023 and is projected to expand at a compound annual growth rate (CAGR) of 32.8% from 2024–30.

Is It a Potential Job?

The growing adoption of generative AI has made prompt engineering an important role for optimising model outputs, leading businesses in sectors like marketing, software development and education to hire dedicated prompt engineers.

Major job platforms reflect significant demand, with ZipRecruiter listing nearly 9,808 openings with hourly rates up to $86, and Indeed reporting an average US salary of $118,007 per year, highlighting the role’s financial appeal.

On the other hand, The Wall Street Journal report suggests that advancements in upcoming models may introduce self-prompting capabilities, potentially automating routine prompting tasks. The report argues that prompt engineering is likely to evolve into a supplementary skill integrated into roles like finance, HR or legal, rather than remaining a standalone job.