Follow us on
social media!
-
Our Hours :
Tuesday 9:00-6:00
Wednesday 9:00-6:00
Friday 9:00-6:00
More Artificial Intelligence
(Here are some broad categories of Artificial Intelligence)
One of the funny behaviors of some AI Models
is to be a sycophant. Always flattering you.
"That was a really good question" is a very common AI statement.
Early AI developers—especially those working on modern Deep Learning and Large Language Models (LLMs)—were deeply surprised by what their models could do. In fact, "surprise" is practically a core theme in the history of modern artificial intelligence.
The root of this surprise comes down to a fundamental shift in how software is made: instead of writing explicit, line-by-line instructions, engineers built an architecture that learned from data.
Developers found themselves caught off guard by their own creations.
AI generates complete answers to Prompts. It doesn't just tell you where to find the answer like Google Search (what everyone has been using for years)
But rather:
* It Generates a paragraph or more answering
your prompt.
* Plus it may reference your previous Prompts
on the topic, even from weeks ago.
* It engages in a Chat with you.
* And has follow-up questions for you.
Some AI Models will look through your device's
camera lens and discuss what its seeing.
e.g. prompt="You see this? what can you tell me about it?
Early developers were often surprised because they couldn't easily peek inside to see why a model made a specific decision. It was an eerie experience: the math worked beautifully, the output was highly accurate, but the exact internal logic was—and in many ways still is—deeply obscured.
Cool links to check and stuff to try
Note:
For links to web pages with Apps/Programs that require a download to your PC or Smartphone, there is a fair chance your PC won't have the computing power (especially graphics power) to be able to run some of AI Models. You will find yourself waiting and waiting and wondering if things are even working. AI Models simply require more computing power.
All the Links below will open a new browser tab/page and may send you to trial versions of software or require that you have a Facebook or Instagram or X (Twitter) account to view. Some stuff is free and some is not.
Hallucinations is an AI term that means the results from a prompt is just wrong. Sometimes comically wrong.
These are Instagram links
Unrequited deference - there a natural tendency while communicating with AI (via Prompts) to say "Please and Thank You". I know, I do it myself.
AI Prompt: "Please a train me on how to have a conversation entirely in beginner's Spanish. Assume we are at a crowded restaurant in Madrid and I am trying to order, but you are a slightly impatient waiter. Correct my grammar inline using brackets."
There is no need to say "Please". If anything it wastes computing resources. AI is a tool. You aren't polite to your garden shovel.
If you just ask for a quiz, it will often give you the questions and answers all at once. If you tell it, "Act as a coach. Give me one problem at a time, wait for my input, and score me," it transforms into a highly effective, interactive tutor.
Google Gemini AI
