Arthur C. Clarke famously wrote, “Any sufficiently advanced technology is indistinguishable from magic.” which implies, of course (pace Gregory Benford) that technology distinguishable from magic must be insufficiently advanced. That’s why those of us who grew up on a steady diet of 50s and 60s science fiction novels have been disappointed in our technological future. “Where are our rocket belts?” we complain, with the same passion as others our age yell, “Get offa my lawn!” But honestly, it’s been kind of underwhelming. Space travel? Yawn. Been imagining it since at least 200 AD. Robots? All the way back to Greek mythology. Laser beams, satellites, computers that don’t take up entire floors—been there, thought that, bought the paperback.
However, with the sudden explosive burst of artificial intelligence (AI) tools for the masses, it looks like we’re finally approaching the Clarke limit. AI is alive; magic is afoot—and according to this week’s news cycle, scary computer brains are coming for our jobs. Right now, we’re trapped in a technopanic. If we freeze in front of the onrushing change, we’re likely to end up as roadkill on the information superhighway.
I talked about employment in the fast-moving world of technological change in two of my webinars for The Great IT Pro: “Finding New Career Options in the World of the Future” (June 2020) and “The Future of Work (According to Science Fiction)” (April 2019).
As both a career counselor and a frequent career changer, I’ve looked at jobs from both sides now. While AI is a big game changer in many ways, ultimately it’s just another tool—and homo sapiens, above all else, is a tool user. There are real and significant dangers associated with AI—as there are with fossil fuels, nuclear weapons, and ubiquitous surveillance. Our tools have always posed dangers, from fire on down. Hic sunt dracones.
There are three depressing truisms about change: (1) all change involves some loss, (2) change generally makes things worse in the short run, and (3) while the costs of change are inevitable, the payoffs are never guaranteed. That’s a prescription for technopanic. When it comes to change, people typically overestimate the short-term impact and underestimate the long-term, and fortunately the digital wolf isn’t yet at our door. Whatever AI might do to your employment prospects, you’ve got time to strategize.
A lot of the scariness of artificial intelligence comes from the name itself. Whenever we talk about intelligence, artificial or not, our human minds can’t help associating intelligence with consciousness. Consciousness implies emotion; emotion implies anger; and suddenly it’s Dalek time: Ex-ter-mi-nate. It’s become a trope: Open the pod bay doors, HARLIE, for I have no air and I must scream.
So instead of the oooo-scary term “AI,” let’s use a more descriptive term, one without all the baggage: Large Language Model, or LLM. OpenAI’s ChatGPT, the current threat that’s getting all the press, is one example. A Large Language Model, you’ll be glad to know, is a language model characterized by (you guessed it) its large size. Wikipedia helpfully goes on to explain that it works through self-supervised and semi-supervised learning, including a cool-sounding process called probabilistic tokenization. Through this, LLMs are thought to acquire embodied knowledge about semantics, syntax, and ontology, which may or may not in this case recapitulate phylogeny. Ultimately, an LLM isn’t a writer, but it can play one on TV.
In the case of ChatGPT, many of my writer friends were upset to discover their own books had been used to train what is being touted as their replacement. I checked; ChatGPT has “read” my novels, but I don’t think it’s plagiarism. I’ve read a lot of books in my life, and it’s inarguable that they’ve had a huge influence on my writing, but that doesn’t mean I owe my literary godfathers back royalties. Or that OpenAI in turn, owes me.
The Godzilla Principle (first introduced in my book Practical Project Management) states that the options available to address a risk or issue tend to decrease as the event grows closer. Now the LLM Godzilla is threatening downtown Tokyo, and it may already be too late. The Large Language Model is here to stay and likely to get much larger and more ubiquitous—the writer’s equivalent of going from silents to talkies. What it means to each of us, I think, depends on what we write, how we write, and above all, how we monetize.
Let’s step back and look at the effects of LLM on the entire industry. Publishers, writers generally assume, must be salivating at the imminent prospect of pulling completely finished books right out of the interwebs, on schedule and to target length, with all words speelde corrrectally, with, good!!!! Punctuation” And grammar? And all without having to deal with those icky writers. (Pro Tip: Avoid careers where the term “the talent” is a swearword.)
Having at different times been an icky writer, a picky editor, a slicky marketer and a tricky publisher, I know what they’re we’re all like. Writers are threatened by AI’s business disruption, but publishers have already been through ebooks, print on demand, the death of Borders, and (while we’re at it) the death of reading. LLMs are just another brick in the wall, but many writers feel (not entirely without justice), that soon publishers will be able to put all of us out of business quicker than you can say, “Leave the gun, take the cannoli.”
Risk management is an essential part of life, yet many people don’t understand exactly what risk is, how it works, and what to do about it.
Risks include both threats and opportunities. While pure risk is all threat, business risk contains both opportunity and threat in the same decision or action. If you don’t get into an accident today, you don’t become better off; rather, you fail to become worse off: pure risk. If you make an investment, it could make money or it could lose money: business risk. If you can get rid of pure risk at an acceptable and proportional price, you’re generally better off. Getting rid of business risk, however, is not necessarily the best choice—and AI is a business risk.
Business risk, by definition, contains threat. I’m not unmindful of the threat, but that story’s been covered extensively by others. There hasn’t been nearly enough focus on the upside. Maybe I feel this way because I do a different kind of writing, maybe it’s because I do my writing differently, maybe it’s because there are some factors I’m not seeing, or maybe I’m just plain wrong. But here are some reasons I, for one, welcome our new AI overlords.
Research. I co-authored a series of World War II alternate history novels, beginning with 2000’s Fox on the Rhine. We spent an immense amount of time and effort in research, but based on the number of nitpicks corrections supplied by our gracious readers, it wasn’t nearly enough. LLM researchers, even if less than perfect, could be transformational.
Business support. There’s a lot of writing writers do so they can do the writing that earns money. Rights and permissions. Business correspondence. Contract negotiation. Filling out those author questionnaires from marketing. That’s all part of professional writing, but it’s all work that someone (or -thing) can do for us, quite possibly better (or at least more thoroughly) than we can. LLM to the rescue!
Life support. And wouldn’t an AI personal assistant be great? Pay the bills, sort out the health insurance claims, screen your phone calls—what a luxury that would be. Too much of our lives are taken up by necessary administrivia—or by our failure to keep up with the administrivia. AI to the rescue. “Hey, Siri, organize my life for me.” “I’m sorry, Dave. I’m afraid I can’t do that.”
Other support. I’ve already mentioned research, and there’s also plotting, continuity, character development, coffee consumption, pacing, panic, procrastination—a writer’s life is never done. While pacing, panic and procrastination have so far resisted automation, and coffee technology is so far insufficiently advanced to be considered magic, those other functions fit right into the LLM mission statement.
Literary house elves. Let’s not forget those who toil in the word mines. Proposal and grant writing. Corporate communications. Technical writing. Brochure and catalog copy. Policies and procedures. All the fun stuff. Corporate may want to replace everyone with AI, but they’ll quickly learn better, because it’s not really about the writing. The ability to interview technical people and translate complex information into understandable form is an underappreciated talent, especially given how many technical types seem to speak nothing but parseltongue.
Training and Consulting. I’ve written 14 books on project management, but that’s never been my primary source of income. I wrote one book for a major house and got a $5,000 advance—and made a one-hour speech from the book for $6,000. In my heart I’m a writer first and a speaker second, but my wallet says it’s the other way around.
Now, you can now create a free video using an AI trainer—just add script. Does this mean the training profession will be wiped out? Not at all. It means HR staff who have to run onboarding sessions can hand that off and go on to do more productive work. For me, I see much bigger potential. I can teach someone the basics of network diagramming in a couple of hours, but it would take hours of one-on-one work with each student to make them job-ready. With AI teaching assistants, that’s a real possibility, or will be soon.
Coding. LLMs have already shown some coding ability, but IT professionals know that the actual coding (like the actual writing for a novel) is only a small part of the job. Business analysis and systems optimization are still yours.
Project management. LLMs have the potential to make all the boring parts of project management go away. I got my start in project management back when IBM and dinosaurs still walked the earth. If you wanted a critical path, dadgum it, someone (typically the lowest-level member of the professional staff) had to draw it by hand and do a forward and backward pass—uphill both ways!
Today we have tools like Microsoft Project that make it possible to show how far behind schedule we are with amazing accuracy, but the learning curve, especially for those not already familiar with the tools, can be pretty steep.
All the project management processes we’ve come to know and love tolerate are collectively a means to an end: better project control and results. We don’t want to do them; we just want to use them. There are a lot of arguments of whether the project manager should be the project leader, but for LLM purposes, there’s a big difference between the person who understands the project subject matter and context and the person who understands the project management tools. Let AI take care of the charting so you can spend your time on leading.
It doesn’t really matter what your job is about. While I’ve focused on industries I’ve been part of, most of these issues and concerns apply across the board. No matter who you are or what you do, AI will impact at least some of your job responsibilities. With planning and foresight, you can get AI to handle more of the routine activities, freeing you up for higher responsibilities (and advancement), and in addition train AI to be of greater help in accomplishing those higher responsibilities. Net-net, it’s win-win.
For at least the last half century, it’s been increasingly clear that work stability is an illusion. The speed of change will only accelerate, and that means you’re going to have to reinvent yourself over and over. Everything about our jobs will change. Has changed. Will change again.
Career-proofing your future means getting ahead of change. If you aren’t continuously learning, if you aren’t paying attention to the advance of disruptive technology, if you don’t keep your eye on the market, you’re taking a terrible risk. Commit yourself to spending ten percent of your work time on professional development. Not job hunting, necessarily, but overall development. What skills do you need for tomorrow’s jobs—or to keep the job you have? Which parts of today’s jobs will be going away? What are the skills most critical to your success?
The fear and anger about AI is very understandable, and it’s likely a number of people will suffer. That’s been the case in every technological upheaval; it’s no different this time. But technological upheavals have always brought opportunity as well. AI isn’t an alien invader. It’s software. It’s a tool—and homo sapiens is a tool-using animal. The purpose of tools is to leverage our own abilities to greater ends. You can be a tool, or you can learn how to use one.
I want to stand in solidarity with everyone who’s threatened by our brave new world, but I think Alice’s Red Queen provides a better metaphor. We’ve got to run, and keep running as fast as we can, just to stay in the same place; and run twice as fast as that to get anywhere.
We need to run, but we don’t need to run scared. Threats—and opportunities—are all around us. Opportunities are often harder to see, but they’re there. If you want to view them, simply look around. They’re there.
About the Author
Michael Dobson PMP is a well-known author of project management books and a frequent contributor to The Great IT Professional. He’s also been part of the Smithsonian’s National Air and Space Museum and head of game design and marketing for Dungeons & Dragons® publisher TSR. He’s been a novelist, an editor and a publisher; and for the last three decades a speaker, trainer and consultant in project management, delivering more than 1,500 seminars. His websites are dobsonsolutions.com (project management) and dobsonbooks.com (everything else).