The Rise of the Cyborg Writer
AI won't take your job. It's somebody using AI that will take your job.
"AI won't take your job. It's somebody using AI that will take your job."
— Richard Baldwin
Garry Kasparov's hand hovered, then committed to his move: Knight to A5.
In chess, it's often said that "a knight on the rim is grim," and as the grandmaster leaned back in his chair, the grim reality of his position against IBM's "Deep Blue" computer started to crystallize. After alternating wins in the first two games of the six-game match, the next three games had all ended in draws. It was down to the decisive final game. Commentators would later assess this ill-fated move as the beginning of the end for Kasparov — and a harbinger of the dawn of AI supremacy. The machine's cold, calculating moves had squeezed Kasparov's options — never making a flashy play but always increasing its advantage. Under mounting pressure, Kasparov saw that he had been outmaneuvered. Rather than allowing himself to be check-mated, he resigned the match.
This 1997 match between Kasparov and Deep Blue marked a historic defeat of the world's best human chess player by the industry's most powerful machine. Although it had been close until this final game, Deep Blue's victory was clear-cut. From that point on, the odds would only keep tilting in the computer's favor. Today, Magnus Carlsen, who is recognized as the best player in the world, enjoys an "Elo rating" of 2882, while computers rank well above 3400.
But a funny thing happened on the way to the AI apocalypse.
First of all, humans kept on playing chess — robot superiority be damned. Second, the machine did not remain invincible. It turned out that the combination of a human being and a computer was more powerful than the computer alone. Hence, "Cyborg Chess," where humans and computers work together, is a sport unto itself with its own reigning champions. In this game, creativity and cooperation with computers are rewarded over sheer speed or brainpower alone. And here's the thing: an average chess player with above-average skill at wielding the computer's help can defeat a grandmaster with or without the computer.
In 2005, Steven Cramton and Zackary Stephen, club players with ratings of 1685 and 1398, respectively, outwitted three grandmasters in the computer-assisted PAL/CSS Freestyle Chess Tournament. Imagine a pair of junior varsity players outmaneuvering Shaquille O'Neal, Michael Jordan, and Kobe Bryant in a 2-on-3 basketball game. Though less technically skilled, perhaps, their cyborg edge delivered the win.
Fast forward to November 30, 2022. A superior level of artificial intelligence exploded onto the scene in the form of a chatbot called ChatGPT. Within weeks, millions of mere mortals created free ChatGPT accounts. Its easy-to-use interface (a simple chat window) returned human-sounding answers to questions. Tech geeks and non-techies alike began tapping into AI's deep knowledge, learning how to prompt or train AI for even better results. The growth of this field has no end in sight, with recent upgrades of "multi-modal" capabilities that can turn speech to text, text to speech, text to image, and even text to video.
This new frontier has many clamoring for regulation — fearing widespread economic displacement — and more attention is being given to the risks of artificial intelligence. Even if you haven't jumped on the bandwagon, you've likely heard AI lingo like large language models, generative pre-trained transformers (GPT), and Silicon Valley's hottest new job title: prompt engineer.
Winners and Losers in the AI Age
During the industrialization of England in the 1800s, the arrival of the mechanical loom gave rise to a group of anti-industrialization zealots. The "Luddites," as they were known, made a sport out of smashing the new warehouse-sized weaving machines. They feared that the new textile factories would eliminate good old-fashioned high-paying weaving jobs. And they weren't wrong. How many professional weavers do you know today?
Most economists argue that the long-term result of the Industrial Revolution was a net improvement in welfare for all. After all, fewer people doing the same work with machines means higher overall productivity. And higher productivity translates into higher wages. Sure, many workers are displaced in the short run, but in time, the unemployed get absorbed into other sectors. In theory, machines liberate people to focus on more meaningful work.
But economists also acknowledge that new technology breeds winners and losers. The concept of skill-biased technical change holds that technology tends to empower those with complementary knowledge and capabilities the most. It's the same lesson we can draw from cyborg chess: the machine is powerful, but the real power comes from the combination of humans and machines.
The entire "knowledge economy" has been built around the symbiosis between educated workers and computers that augment the ability to put their knowledge to use. AI accelerates this phenomenon.
Unless you work in customer support, your job may still be safe — for now. However, you are at risk of being outcompeted long-term by those who have a better grasp of these tools than you. AI might not take your job, but someone using AI will.
In the years ahead, the gap will widen between the skilled and the unskilled — the AI-assisted, and the unassisted. You stand to win or lose to the extent that you learn to work with AI. Even the plumbing contractor who learns to use AI to communicate better with his clients will gain an advantage over his neo-Luddite competitors.
AskAI → Improve Writing
My "aha" moment with AI came with the release of a new suite of AI tools embedded right inside Notion — the main software platform I use in my work as a writer and radio/podcast producer.
I highlighted a block of my rough, unedited transcript within Notion as if I were going to reformat the text in bold or italics. But in addition to the usual formatting options in the menu bar, Notion presented me with a shiny new button labeled ✨AskAI.
Unlike OpenAI's world-famous ChatGPT, Notion's new AI function didn't present me with an empty chat box but with a set of commands — the ability to ask AI to do something with my own content. The first preset option under AskAI was Improve writing, next to an icon of a magic wand.
So, I scrolled down, hit Improve writing, and waited for the response.
In seconds, NotionAI cleaned up and improved the writing of an entire block of rough transcript. Any writer worth their salt can recognize good writing, whether crafted by humans or machines. And NotionAI's "improved writing" was good. The output was concise and logical, yet the speakers' personalities were preserved.
My previous editing process involving hundreds of decisions had been simplified to just three clicks:
Highlight text → ✨AskAI → Improve writing.
After seeing NotionAI do my job better than me, I felt a budding sympathy with concerns about AI replacing creative workers. What if I wanted to keep doing my job the old-fashioned way? My attitude towards AI shifted from skepticism into a mix of curiosity and apprehension.
Was I about to become a victim of the "great replacement" of workers with machines?
Or, perhaps, would enlisting AI free me from lower-level editing tasks? Would it allow me to focus on higher-order tasks, such as repurposing transcripts into valuable content like innovative articles and books?
Instead of resisting technological progress like the Luddites and smashing my computer, I could choose to master the art of AI-assisted writing. I could learn everything about the emerging craft called "prompt engineering" or "AI whispering" to coax my desired outputs from AI.
I had a new goal: I was going to surf the AI wave rather than drown in its churn.
Level Up or Go Extinct
The solution didn't lie in any single "killer app" or "superprompt" but in developing an intuitive feel for how to command AI, coupled with a few basic principles and a framework for getting consistent results.
AI is not a machine for generating answers or polished content out of thin air. Instead, it's a tool for transforming and refining human-generated ideas and inputs. This distinction between tools and machines is subtle. Whereas a machine is designed to be "foolproof" — such that any fool can operate it — a tool requires skilled hands to operate. There is a world of difference between AI-generated content, churned out by a machine, and AI-assisted content, in which you leverage AI as a tool in the writing process.
Feeding AI the right prompts and source material, AI can organize, distill, and build upon your purest ideas — speeding up the writing process while maintaining the quality and specificity of ideas. It can clarify your thinking and purify the signal from your noisy brainstorms. Beyond improving sloppy writing or a rough AI-generated transcript, it can help structure a logical argument, fill in the gaps, and smooth awkward transitions.
But wait... there's more. How else can AI help you?
Rewrite text in the voice and style of another person
Extract follow-up "action items" from a meeting transcript
Simplify complex scientific papers for easier understanding
Translate complex concepts into accessible analogies or metaphors
Offer feedback on tone, readability, and alternative phrasing
Provide instant translations for foreign language text
Identify repetitive words or phrases and suggest synonyms
Going forward, you have two options.
Will you recoil at the rapid pace of change and resign yourself to potential obsolescence?
Or will you build complementary skills to work with AI that make you irreplaceable?
Within five years, if you level up, telling people, "I'm a writer," could carry more prestige than saying, "I'm a doctor." AI-assisted writing salaries could rival those of software engineers. With the emergence of AI language models like GPT-4, English is transforming into the hottest new programming language. You can now "code" through conversational prompts. No technical degrees are required — just a keyboard, an internet connection, and a willingness to experiment.
This post is adapted from "Commanding the Page: The AI-Assisted Way to Improve Your Writing, Publish Your Ideas Faster, and Future-Proof Your Creative Career" (2023).