The Wit's Guide to Artificial Intelligence
Or, half wit at scale
If the development of artificial intelligence is like the early days of the automobile, then I’m the guy most excited about the invention of the hood ornament. Yes yes, AI will revolutionize all communication, but let’s get to the clever quips already!
“A witty saying proves nothing.”
“The secret of being a wit is to know the proper value of everything, and the amusing value of nothing.”
— OpenAI in the style of Voltaire
The second issue of this newsletter, titled “Oscar Wilde Will Not Be Automated” and written waaaay back in 2019, was innocently chuffed about the possibility of AI wit. Oscar Wilde, that famous polisher of witty paradoxes, often seemed to be working off a formula. He prized both artificiality and intelligence and his many detractors claimed a robot could do what he did. As one wrote:
“Indeed, I am not sure that Mr. Edison could not, if he gave his mind to it, design an apparatus for turning out ‘Oscarisms’ automatically. We might put our pennies in the slot, press a button, and draw out ‘Wilde’ paradoxes on tape by the yard. It would not require nearly such elaborate mechanism as the late Mr. Babbage’s once famous Calculating Machine.”
By 2021, it seemed like we had the technology. Oscar Wilde was on the verge of being automated with GPT-2. Now, finally, we could turn out epigrams by the score! I fiddled around with what was available and generated a few, the best of which was:
“The greatest advantage of the typewriter is that it stops a man from thinking about himself.”
— OpenAI (2021) in the style of Oscar Wilde
Now, in 2023, it looks like artificial intelligence is here for real. ChatGPT has dazzled the masses at a new level. The most ballyhooed immediate use for this technology is the mass production of, to use a technical term, bullshit. Instead of spending laborious person-hours writing reams of text that no thinking person would ever want to read, we can have a computer do it! High-school essays! SEO-optimized marketing copy! Towering heaps of nutrient-rich fertilizer! All now available instantly!
“The height of cleverness is to be able to conceal it.”
— Duc de la Rochefoucauld
"The greatest wisdom is to know how to appear foolish, and to be able to dissemble one's wisdom."
— OpenAI in the style of Duc de la Rochefoucauld
Wit, per Alexander Pope, is what “was oft thought but ne’er so well expressed.” Current AI can instantly retrieve what’s oft thought, but the expression is by definition thoughtless. And so artificial intelligence, in its current form, is half wit at remarkable scale.
“Why, I've heard it said that an infinite number of monkeys with an infinite number of typewriters and an infinite amount of time could eventually write the works of Shakespeare. But I say, why wait?”
— OpenAI in the style of Groucho Marx
It remains fascinating, and I’ll continue to look for the AI equivalent of a hood ornament. Here’s hoping the first time I notice it won’t be on the hood of an oncoming car.
“I can resist everything except the temptation to use artificial intelligence.”
— OpenAI in the style of Oscar Wilde
“Artificial intelligence is like a Fran Lebowitz joke: it’s clever, but it’s not funny.”
— OpenAI in the style of Fran Lebowitz
“I heard they're developing artificial intelligence that can walk into a party and not only not talk to anyone, but also not even realize it's a party.”
— OpenAI in the style of Mitch Hedberg
“The ideal love affair is one conducted by post.”
— George Bernard Shaw
Valentine’s Day is just around the corner, which makes one wonder: Do cupids grow up to be angels? Or do they all die young of tragic archery accidents? Every time you vote for next week’s subject, I’ll confiscate the weaponry of a clumsy cherub.
Generating nonsense at scale
Get Wit Quick No. 187 maintains a robotic arm’s length relationship with all technology. If we can design an internet entirely written and read by machines, perhaps the newspaper industry can stage a comeback. My book Elements of Wit: Mastering The Art of Being Interesting discussed Twitter wit with naive optimism, but then it was 2014. A machine may never truly be able to love, but it could infinitely tap the ❤️ below.