UltimaParadox

quote:

---

Anthropic begs to differ

---

Even the article you posted refutes what you just said

quote:

---

before accounting for revenue shared with distribution partners, including Amazon (AMZN), and the cost of training its model

---

Ah yes if we ignore our largest cost we made money!!

So no they are not making a profit

VOO and VTI

Pretty much all you need

quote:

---

The point I’m making is it’s not data center noise, it’s construction noise. That will end once center is built.

---

Space X already disclosed the noise is coming from 60 gas powered turbines at the site.

quote:

---

A private entity could do their job more efficiently and probably for 1/4 of the cost…

---

The first thing a private entity would do is cut unprofitable routes.

The USPS can't do that, and no profit company would deliver to everyone.

That's why it's a government service

I've tried chunks as well. However since every model has no concept of pages on PDFs it becomes a guessing game in what is contained in a chunk.

You end up searching the file yourself for the information, rather than getting false information.

The worst part though is even in small files they still hallucinate horribly.

Gemini Enterprise is the worst. Even giving it small documents and telling it to prove what it states with documentation. It confidently states wrong locations with fake information.

You can't point it to page numbers, since it has no concept of pages.

Codex at least say it can't process the file these days with the latest model, as it seems like they have given up on it lately. It will churn through almost a million tokens to give you no answer lol.

At this point we are trying different file formats. But every manufacturer on the planet puts their documentation in PDF. So unless it's something like a markdown file, I have little hope it can read anything remotely sophisticated.

Claude actually did better if you asked if for information instead of providing the source. But also once you try to tell it to point you to the reference to double check it generally has no way to consistently cite it's sources.

Some models definitely gotten worse over time.

LLMs are pattern matching to a fault, they just suck at folding human made documentation

This is just another Monaco... Unfortunate

quote:

---

How do all the companies develop at essentially the same pace? It doesn’t make sense

---

They don't? Open AI and anthropic LLMs are far ahead of the competition. Google is probably next. Then the open models are behind them.

However the open models are catching up quickly. I am of the form believe that a completely different approach from the current approach to LLMs will need to occur to reach any sort of AGI.

I think the public's definition of AGI will not match what these companies will state is AGI as they push for more investment.

In the second half of 26 right now we are in the efficiency and monetization transition phase.. That is why there is so much fear mongering as anthropic nears IPO. Fear drives investment

OpenAI financials are an absolute disaster everyone knows it. They have no real path to profitability right now as models are switching to efficiency gains.

Open models are catching up fast and do well enough to avoid the massive costs of frontier models which are not improving enough to justify their massive cost increase.

Users aren't paying for AI, and businesses that do pay prefer anthropic.

Anthropic isn't making money yet either. As they approach IPO they will ratchet up the stories of AI dangers as they have shown it drives fear and investments. Even though it's clear Dario just wings it for the last 5 years.

2 TE and still can't run for the first

Need to give KR an easy throw on first...