re: I asked Google if TigerDroppings is racist. Here’s their answer

quote:

---

Can a human mind be programmed?

---

Ooh bait and switch!

Just remember this moment.

Because you have no idea how an LLM works yet told yourself that you do.

quote:

---

Google seems to be embellishing a bit, in my view.

---

Google's is filled with leftist BS and bias. It is always trying to manipulate people into being leftists or being their useful idiots.

quote:

---

You'd do far better assessing what you do, or do not, understand, than assuming what I know. Don't assume, address what you yourself know.

---

Assuming?

It's obvious how little you know on this topic.

Let's break down where you lost the plot.

Wikipedia has a good summary

quote:

---

Computer programming or coding is the composition of sequences of instructions, called programs, that computers can follow to perform tasks. It involves designing and implementing algorithms, step-by-step specifications of procedures, by writing code in one or more programming languages. Programmers typically use high-level programming languages that are more easily intelligible to humans than machine code, which is directly executed by the central processing unit.

---

We the developers write the code (programming) that creates the attention layer, that is something we control.

Different people then take that set of instructions and feed whatever data they want into it. No programming involved, just point the developed model at a set of data and go. it's like you telling YouTube which song to play next, or your computer which video game to load.

The model then creates a massive set of weights that are the resultant matrices.

That is training a model.
https://www.nvidia.com/en-us/glossary/ai-training/

They are not the same, they have never been the same, they are functionally different.

quote:

---

AI response is severely limited by the sources it is programmed to access, and those it is not.

---

You keep using the word programmed, but models are just given data, you don't write lines of code that say go explicitly here, maybe in a college class, but not at a Google/OpenAI/Anthropic level.

Again, the same model can be trained for different purposes by different people without "programming" anything.

quote:

---

In so far as prompting an LLM, I've described exactly that in other posts .. and this one. The prompts issued in this instance led to an AI "apology" for the previous and illogical response.

---

This is because you don't know what you are doing or describing.

Depending on which model you are using, if you are working directly from google search it has no idea what you said previously.

quote:

---

When enabled, memory helps ChatGPT automatically remember useful context from your chats, files, and connected apps to personalize your experience, so you don’t have to repeat yourself as often.

Memory controls are available in Settings > Personalization > Memory. You can choose to enable or disable memory at any time.

---

In google Gemini it's
https://gemini.google.com/saved-info

As I mentioned before there are specific ways (assuming you have the settings on) to get it to remember.

It's as easy as saying "Remember that..."

quote:

---

Training, independent of computer programming, implies programming has reached independent agentic or even sentient levels. If not, training is a programming component

---

It's painful to see you struggle.

quote:

---

AI training is the computational process of teaching an AI model to recognize patterns and logic within data. It transforms raw information like text, code, or images into intelligence, enabling the AI to perform tasks it wasn't explicitly programmed for-such as reasoning, problem-solving, and content generation.

---

I'm fine with you struggling with this for another dozen pages searching for a gotcha.

But you sound really really ignorant and out of your depth.

quote:

---

You are sooo stupid lol.

It's cute, you deeply out of your depth saying things you don't understand.

Keep going.

---

Yes, fundamentally you can't even figure out how to get a general tool to save a memory.

You keep fundamentally misusing and misunderstanding terms in a field you just don't comprehend.

But yes, you can prompt for a picture.

Good job, so proud of you.

You've obviously given up on your nonsense about

quote:

---

"Google is lying, of course, because it is in the programming."

---

I hope you've learned something.

quote:

---

But yes, you can prompt for a picture.

---

You are an effing moron.

Who did I "prompt" for a picture, nitwit?

quote:

---

You've obviously given up on your nonsense about "Google is lying, of course," because it is in the programming."

---

You appear to be Illiterate, drunk or high. You either don't understand what you're reading or are too altered to assimilate it.

quote:

---

What do they do? What is ths .... ?? code ??

---

So this is where you are clueless, the attention layer is what is used during training to create the weights, the same attention layer trained on different sets of data will produce completely different answers.

The vast majority of all AI models use the same basic software in different configurations.

Training is what produces the answer that you complain about.

It's like you clicking on a song on youtube, not liking it, and then claiming the "programmers" are out to get you lol

I may be mocking your ego, but I am willing to teach you lol

quote:

---

I may be mocking your ego, but I am willing to teach you lol

---

Ask the same for DU