In the past few years, we’ve entered a veritable revolution of artificial intelligence (AI) technology. With the emergence of generative AI, which is capable of generating text, photos, and even videos based on user prompts, and specific AI tools designed for revolutionizing certain professions, millions of people have become concerned that their jobs are on the chopping block.
This is a perfectly legitimate concern. If a sufficiently powerful AI can do everything you can do, but better, faster, and cheaper, why would your employer need you?
However, it’s incredibly likely that AI is going to displace jobs, rather than replace them, at least generally. How can we be confident that this is the case?
A Case Study: Legal AI
The easiest way to start answering this question is to use a case study: legal AI.
According to Law.co, legal AI has developed significantly in the past few years, with current capabilities that can “take over many of the most menial responsibilities associated with legal professions, such as document review, document drafting, legal research, and more.” In a matter of minutes, a properly engineered prompt can save a lawyer hours of research and effort.
However, most lawyers who have used this type of tool have no fear that they’re going to be replaced anytime soon. After all, these are the people using the tool in the first place. They’re the ones guiding the generative AI with engineered prompts. They’re the ones fact checking the work to make sure it’s accurate. They’re the ones taking the work and presenting it in court, or to clients, or to partners.
It’s true that there are many responsibilities and tasks within the legal profession that may never be touched by human hands again. But that doesn’t mean that lawyers are going away or that we no longer need human lawyers to help us navigate the legal world.
Why Technology Typically Displaces Without Replacing
So what is it, exactly, that makes technology typically displace workers, without really replacing them?
AI, like many technologies, is somewhat limited by the following concepts.
· Technology needs development, implementation, and guidance. Technology will always need development, implementation, and guidance from human beings. There’s no such thing as technology that requires no external human touch. Even if we have incredibly powerful machines and tools, we’ll still need humans to use them.
· Changes are slow, even with explosive tech growth. Fax machines became obsolete in the 1990s, and yet they’re still used as a primary medium for sending and receiving information for countless businesses. That’s because even the most impressive, universally beneficial technology advancements are usually slow to roll out to the general public. There may be a few startups using AI almost exclusively, but most businesses and organizations are years to decades behind the leading edge of the technology curve.
· Technology is never perfect. Even very impressive technologies are never perfect. Taking inspiration from our example above, legal AI is capable of making mistakes, submitting false information, and even plagiarizing materials; amazing as it is, it still requires human intervention and review for reliability.
· Human connections will always matter. It’s also important to recognize that human connections will always matter, at least to some people. It may be possible to use an AI engine to draft a legal document for yourself, but you’d probably feel much better having a face-to-face conversation with a human lawyer who can explain the intricacies of that contract to you. It’s going to be very hard to ever replace this need for human connection.
· New, surprising roles emerge. As we’ve seen for hundreds of years, every technology that eliminates some jobs has replaced it with new, potentially surprising roles. Today, there’s an incredible demand for software developers, despite the fact that this career was practically nonexistent 50 years ago.
· AGI is a totally different animal. So far, most of the AI development breakthroughs we’ve seen in the past few years have been relegated to narrow intelligence applications. These types of AI are very specialized at accomplishing one specific kind of task. They don’t have “intelligence” as we typically think of it; they just happen to be disproportionately good at one, narrowly defined thing. Artificial general intelligence (AGI), in contrast, does have more conventional intelligence – and could be much more competent and flexible than its narrow-minded predecessors. AGI may have truly disruptive power in our economies and businesses, but comparatively, narrow AI just doesn’t have the same disruptive potential.
An Uncertain Future
There’s one major caveat that we need to address here: we don’t know exactly what the future holds for AI. It is possible that AI scientists and engineers are capable of developing an absurdly intelligent AI, or that narrow AI becomes so capable and competent that it does replace certain people in certain types of jobs. But even in these scenarios, new roles are likely to emerge and the need for human connection is never going to go away.
Whether you’re an optimist or pessimist about the future of technology, and whether you’re embracing the AI revolution or are somewhat afraid of it, it’s important to come to terms with the fact that the future is unpredictable. All we can do is remain adaptable and do the best we can with the knowledge and resources we have.
Leave a Reply