Welcome to Thunderdome

TheCatt wrote: ↑Wed Nov 26, 2025 12:53 pm But for a few certain things (editing movies/photos/music), it's amazing.

I relent that they have always been strong there, but amazing? I wouldn't go that far. Pros don't use Apple. They use professional equipment. At least in the TV and movie industries.

Tell me this isn't the exact story at companies around the globe right now.

Lol

Absolute truth. We rolled in out to 35 users this year. Less than 10 are using it.

M365 CoPilot is pretty damn solid (and just Chat-GPT under the hood). I cannot believe they're going to lose market position to Gemini. Gemini sucks. It's constantly wrong.

Gemini pro is really good. Regular Gemini is improving.

We use GitHub copilot, and about half our company uses it now.

We are trialing the ms365 copilot.

The people that should be most concerned about this all live in India...

Yeah, you can't tell because NEITHER is helpful.

Honestly, this is one area where maybe AI can improve things. customer service is god awful.

At least she speaks English.

There really is no RAM: https://www.cnbc.com/2026/01/10/micron- ... msung.html

Jensen Huang thinks we're the problem with AI.

I get what he's saying, but he should shut the fuck up. People SHOULD be skeptical of AI.

It'll be interesting when they have to start charging more for their services (All AI providers). See if they can finally deliver. It keeps feelings like bizarro world to me. I see people who claim all sorts of great things from AI, but no one I know sees it, and I don't see it when I use it. (skill issue?)

TheCatt wrote: ↑Mon Jan 19, 2026 9:35 am I don't see it when I use it. (skill issue?)

I have thought the same thing.

I was listening to an account from a medical researcher who discussed their heavy use of AI, and they said it caused them to be more hands on than normal. Sure, it could do some things faster, but it had no concept of context, so it had to constantly be prompted about what to look for and do.

Reminds me of a line from A Fish Called Wanda.
"Apes can't read Nietzsche."
"Sure they do Otto, they just don't understand it."

This all makes sense. Not all of these AI companies are going to survive, so someone has to die...

Let's go Claude and Gemini!

Gemini is rapidly becoming the boss.

Grok is kicking everyone's asses in tests, but the user base isn't there.

Co-Pilot is just ChatGPT.

Anthropic and Claude are other big names, but I don't know a ton about them.

Say OpenAI fails, it's obvious one of these other companies will pick up their infrastructure, but what about the code for the AI? I wonder how valuable that would be to one of the other AIs?

Leisher wrote: ↑Wed Jan 21, 2026 3:44 pm Anthropic and Claude are other big names, but I don't know a ton about them.

Claude Code appears to be the current HOT THING in development.

My boss and I were just fucking around with AI and discussing all of this, and it occurred to me that I overlooked Microsoft's role with OpenAI. If OpenAi were to fail, I think MS would be there to catch the pieces since CoPilot is ChatGPT.

Also, CoPilot just told me that my D&D alignment based on everything it knows about me would be Lawful Good. Essentially, I'm Superman or Captain America, except sexier in a thicker way.

Leisher wrote: ↑Wed Jan 21, 2026 4:55 pm and it occurred to me that I overlooked Microsoft's role with OpenAI. If OpenAi were to fail, I think MS would be there to catch the pieces since CoPilot is ChatGPT.

Depends on who has the claims in bankruptcy court. But yeah, MS would try. Maybe there'd be an asset auction.