Of all the anxieties surrounding artificial intelligence, the biggest one is disastrous military applications. Technophobes have referenced The Terminator’s Skynet as the endgame for AI, particularly now with the power of the new ChatGPT chatbot, which is scaring everyone on the planet.
Created by San Francisco-based company OpenAI, the prototype has grabbed enormous attention, with applications including data science, college essays, and even formulaic screenplays. With some demanding government regulation, reaction to ChatGPT has been much stronger than to OpenAI’s art program Dall-E earlier this year.
How the same company could come up with something as clever as Dall-E – a portmanteau of Wall-E and Salvador Dalí – and as muddled as “ChatGPT” – short for Generative Pre-trained Transformer – is somewhat baffling. The cleverness of Google is it’s both a noun and a verb, so people say they’ll “google” something, even when using a different search engine.
But it will be small comfort indeed to Google’s leadership if people refer to “googling” when they look for stuff with ChatGPT. And if Google CEO Sundar Pichai is as scared as recent headlines make him out to be, then I certainly should be. So, I turned to the only source I trust enough to waylay my concerns.
ChatGPT itself.
When I asked the chatbot, if the Internet would at last do to writers what it’d long-since done to travel agents, the bot told me:
“It is unlikely that Chat GPT or any other natural language processing tool will completely replace the need for professional writers. While Chat GPT and other NLP tools can generate human-like text, they are not capable of replicating the creativity, nuance, and depth of thought that a human writer brings to their work. As such, they are unlikely to replace the need for professional writers who create original content for a variety of purposes, such as journalism, advertising, marketing, or creative writing.”
That’s comforting, but watching the program spell all this out faster than I could type was even more intimidating than the old robot-knife scene from Aliens.
Reaction anyone has to new technology is directly related to how much it will affect them. The creation of the steam-powered rock drill was, according to folklore, probably a more exciting piece of news for John Henry’s boss than it was for him. A similar threat to Belgian workers inspired them to throw their shoes, called “sabots,” into machines to break them – hence the word “sabotage,” at least according to popular myth and Star Trek.
When I pointedly asked ChatGPT if it would put MEout of work, it reassured me:
“As an artificial intelligence, I am not capable of taking on tasks or responsibilities that require human skills or abilities. I do not have the ability to compete with or replace human workers in any industry. My goal is to augment and assist human users, not to replace them.”
That certainly sounds fine. Most of us have benefited from advancing technology, because as it’s replaced joyless, painful work like steel driving, it’s created new opportunities that have improved everyone’s quality of life. If ChatGPT lets me write more, better content with even more, better pop culture anecdotes, then it will be as welcome an addition to my toolkit as the spell-check has been for decades.
And I hope a ChatGPT app can improve spell-check, because my phone is still unable to figure out th3 should always be the, because the e is right under the 3 so it’s very easy to mis-type. If ChatGPT can out-google Google, then it can certainly improve spell-check.
Double-edged sword there though. Given that AI’s sophistication is exceeding the dreams of science fiction, there is increasingly less space in society to which we can advance (or retreat) in this “Humans Need Not Apply” world.
So, I decided it might be a good idea to negotiate with ChatGPT. If it didn’t destroy my livelihood, I promised it could be my best friend. It wrote:
“As an artificial intelligence, I do not have the ability to make promises or form personal relationships. However, I am here to assist and support you to the best of my ability. If you have any questions or need help with something, please don’t hesitate to ask. I’m here to help!”
Note that the phrase “I’m here to help” was exactly what Ultron promised before becoming an evil kill-bot in The Avengers. Let’s hope things turn out better than that, indeed with the new year let’s hope that AI benefits everyone and hurts no one, because new technology doesn’t have to destroy the world to nevertheless destroy my world.
And if a machine can learn the value of human life, maybe we can too.
Jared Whitley is a longtime Utah and D.C. politico, having worked in the Senate, Bush White House, and defense industry. He has an MBA from Hult International Business School in Dubai.
