For better or worse, artificial intelligence, or AI, is here to stay and might just be “the next big thing” that transforms the world.
Its potential to increase human productivity in nearly all endeavors seems boundless. Witness OpenAI’s latest upgrade for its writing program GPT-3, or Chat GPT, which allows a user to ask for an essay on a specific subject written in a specific style, say the history of bond markets in the form of a Shakespearean sonnet. The program scours the internet and produces results in moments. The results are so good that observers are raising ethical concerns and fear of job displacement. But how good is GPT-3, and what are the implications for this latest AI advancement?
The program effectively moves from search to composition almost instantaneously, extracting information from the internet and turning it into prose. Its ability to research, assemble and compose information feels almost human. And like humans, it is frequently wrong, another victim of vast misinformation available on the web.
Fears of job displacement may be justified over time; however, this is part of the creative destruction that moves society forward. There was a time when most offices had a bay of typists, producing printed materials one keystroke at a time. The personal computer destroyed that function, but freed people and organizations to become more productive. Calculators expedited complex math equations, allowing for more efficient work across finance, accounting, engineering and other industries. Automating the routine allows for higher-value work to be accomplished.
Ultimately, GPT-3 is simply a tool, and it can and likely will be misused. It will probably produce more than one fraudulent high school or college essay. How we use the tool will define its value. Embracing and deploying GPT-3 (and its progenies) efficiently is critical to the long-term value of this new AI toolset. When used with discretion, it can help advance our thinking and understanding. That is progress and something to embrace.