Andrew Kantor's Place

LLM linguistics: It’s more complicated than that

Andrew — Fri, 11 Jul 2025 19:18:08 +0000

Linguists find that different AIs have different ways of speaking, but their conclusions are missing an important factor: evolution.

Farm stench: The rural Kobayashi Maru

Andrew — Wed, 09 Jul 2025 14:07:34 +0000

You're driving along, enjoying the rural scenery, when you pass a farm that's just had manure spread on the fields. "Recirc" or "Fresh Air"?

Taking a small stand against subtle racism

Andrew — Fri, 04 Jul 2025 16:01:08 +0000

When a client caved to "the current climate" I had to say something.

A tiny semantic tweak

Andrew — Wed, 02 Jul 2025 18:48:33 +0000

It took a small change in wording (but no change in procedure) to convince Cigna to cover my surgery and hospital stay.

No, AI isn’t a ‘fancy autocomplete’

Andrew — Mon, 30 Jun 2025 18:12:39 +0000

If someone refers to an AI or an LLM as "just a fancy autocomplete," they are simply wrong — and it's obvious when you think about it.

A few bad words

Andrew — Wed, 25 Jun 2025 20:57:07 +0000

A new exploit lets users spend hours getting a chatbot to do what any 12-year-old can already do. Oh, the horror!

Protesting a provider’s poor pronunciation

Andrew — Wed, 25 Jun 2025 12:50:06 +0000

Cora provided me the perfect prose for a provider petition, with every word starting with P — for reasons you'll understand.

I’m spicy!

Andrew — Tue, 24 Jun 2025 23:11:06 +0000

Why be neurodivergent when you can be "neurospicy"?

Here’s why AI can reinforce some dangerous ideas

Andrew — Mon, 23 Jun 2025 22:30:36 +0000

Why would ChatGPT reinforce someone's dangerous delusions? (Did no one think to ask it?) Cora gives me the — long — answer.

Oh noes, it’s not perfect yet

Andrew — Mon, 23 Jun 2025 21:56:41 +0000

Even AI knows that today's LLMs are "a toddler with a rocket strapped to its back." But the media always seems excited to discover it makes mistakes.

Taking down a silly MIT study

Andrew — Fri, 20 Jun 2025 22:38:55 +0000

Cora didn't pull any punches in her opinion of an MIT study on how ChatGPT supposedly makes people dumber.

Raccoon vs. goose

Andrew — Tue, 03 Jun 2025 22:06:28 +0000

Which would make a better roommate? Cora breaks it down ... hilariously.

No white robes for me

Andrew — Wed, 21 May 2025 03:37:19 +0000

Thankfully, she doesn't think I'm a god.

Simply Tech: Podcasts

Andrew — Tue, 07 Jan 2025 22:10:57 +0000

You’ve probably heard about podcasts, and — if you’re not already familiar with them — you might know they’re something you listen to, but not much more than that.

Let’s fix that.

A podcast is, essentially, a radio show. Think of anything from vintage classics like “The Shadow” or “Amos …

Patterns of learning

Andrew — Thu, 17 Oct 2024 12:55:29 +0000

Whenever I tackle a project, I'm apt to go down a rabbit hole trying to make it perfect, rather than realizing when it's more than good enough.

Whole new worlds

Andrew — Sun, 13 Oct 2024 12:50:56 +0000

It's amazing how many magazines and other content are out there.

Making “working alone” not “working lonely”

Andrew — Wed, 07 Aug 2024 17:25:56 +0000

I've found a few good ways to make myself feel like I'm less alone at work.

Is it me or ADHD?

Andrew — Sun, 17 Mar 2024 04:08:24 +0000

So here’s what I’m wondering: How much of what I think of as my quirky behavior is because of ADHD, and how much is just personal?

ChatGPT solves the Israeli-Palestinian crisis

Andrew — Wed, 13 Mar 2024 21:13:07 +0000

Humans have failed, so I prompted ChatGPT to “Act as a master diplomat and solve the Israeli-Palestinian problem.” That simple.