Home For Fiction – Blog

for thinking people


art

Walrus Insulter: a Fun Little JavaScript Program

September 9, 2024

Does the word “insulter” exist in a standard dictionary? I don’t know and I’m too bored to search. But the name Walrus Insulter definitely does, in my dictionary. Obviously for those who know him, Walrus Insulter refers to none other than the supreme overlard [sic] of puns, my cartoon character Punning Walrus. If you haven’t met him yet, just click the “Free Walrus” link on the top menu. Or here’s the direct link for the lazy ones among you…

The whole thing, as they usually do, began on a boring cloudy afternoon. I was playing with my home network, trying to find new ways to annoy my neighbors (more of this in a while), when I realized that the best… walrus for the job is of course Punning Walrus.

Walrus Insulter. Screenshot from the program
Walrus Insulter: Don’t mess with the Walrus 😛
(more…)

ChatGPT vs Gemini Turing Test: Hilarious but Insightful

August 12, 2024

I recently read The Emperor’s New Mind, by Roger Penrose, and a small part of it describes the (in)famous Turing test. In simple terms, it’s a process where a machine (typically a computer) can demonstrate intelligence (and, some would argue, consciousness) enough to be indistinguishable from a human’s. With such intelligent and mature – cough, cough – artificial intelligence models like the ones we have today, the hilarious idea materialized: I should stage a ChatGPT vs Gemini Turing test!

Both Alan Turing himself and Penrose in his book expected future computers to be able to pass the test. Turing referred to a 30% success rate by the year 2000, whereas The Emperor’s New Mind, published in 1989, mentions the year 2010. In any case, to me it seems Turing would certainly assume a large language model like ChatGPT or Google’s Gemini to pass the test.

I mean, it almost feels like magic: You ask it questions and it answers, seemingly “like a human”. Artificial intelligence can be an amazing tool – also for writers. But the whole concept behind a Turing test is to “unmask” artificiality. In other words, you only need one type of question to blow the computer’s cover.

And that’s what happened here. Hilariously, with the help of another computer!

ChatGPT vs Gemini Turing test: AI render of two cartoon computers facing each other
Since my topic is a ChatGPT vs Gemini Turing test, I might as well use an AI render to visualize it…
(more…)

“Am I the Asshole?”: The Art of Self-Assessment

July 15, 2024

There is an often quoted claim suggesting that if you run into an asshole in the morning, you ran into an asshole, but if you run into assholes all day, you’re the asshole. This isn’t always true – there are never black-and-white answers – for reasons we will examine, but it implicitly focuses on an important issue: How do we determine whether we’re right or wrong? How do we determine, “am I the asshole”?

There’s even a Reddit thread where people share incidents with strangers and expect them to answer, “Am I the asshole”? Of course Reddit, like the internet at large, relies on consensus. If 10,000 people insists you’re wrong, they must be right… Right? At least that’s what the bandwagon fallacy would like us to think. Obviously enough, this takes us back to the “assholes all day” problem.

But again, there are never easy answers.

So in this post, let’s try to unpack all this. Let’s see why we can’t rely on public consensus to figure out whether we’re right or not, and what we can do about it.

Am I the Asshole? blurry image of people
Humans are social animals. We want others’ approval. But what if others are wrong?
(more…)