In this one he does the goodbye loop with the AI himself, then says I love you, and asks the AI to say it back. 🤡

It’s certainly not “real” interviews. But I wonder if they’re using the platforms/AI companies actually use.

Hilarious. Chaotic. Absurd. The more I watch of it, the more irritated and sad I become, though.

I wonder how much he prepared his AI, or set them up for failure with false information for the bit.

He has more interviewed by AI videos on his channel, which are quite absurd and funny.

All my friends suddenly love playing tennis and I have to talk to the AI about it! (/s)

Broken link. Looks like you pasted the URL twice?

I hate “Trump says” headlines because they spread his bullshit. Headlines show up without article content.

The interactive version of this is at https://teybannerman.com/strategy/2026/03/31/how-many-microsoft-copilot-are-there.html

It also allowed me to use text search because how else do you find the ones you are looking for? I looked in the wrong category.

Referencing current times in this way is crazy. Ignorant of ICE, surveillance, violence, and prejudice criticisms.

Billboards are not the only thing projecting an image.

deleted by creator

I have been playing assassin’s creed Valhalla, which I won. It’s 80% cutscenes, transitions, and traversal. I find it incredibly irritating, disconnected, and boring.

Sorry you had to go through that.

A single company doesn’t make the norm, though. Did you get insight into other companies too?

Just recently I watched a Japanese citizen saying its a common misconception, a prejudice. That just like there’s bad companies in other countries there are these in Japan. They just use a term, a label for them.

I use Nushell as my CLI shell, and I set up aliases/commands, so now I use dl, dl firefox, dl opus firefox etc.

It also helps in remembering or looking up what the syntax to include chapters, sponsorblock sections, or for downloading only a timeframe arguments and synax is.

haha, that’ll be a long linear list with a big library

https://www.sagexpo.org/#games

Closer to industry and physical production, there are more use cases than OCR. In data economy, statistics, and other data handling, machine learning is also a topic.

Machine learning is a wide field.

My company did a demonstration project on how AI could observe motor sensors to aid planned maintenance (recognizing and categorizing irregularities in movements).

I can’t speak to concrete numbers or prevalence in different fields or job market opportunities. How much maths is involved, as I wrote, depends on the field and application.

There’s security and possible support implications, and a general concept of reducing data use and storage, but I don’t see why they should take precedence for something like this.

Overall, yes, it can’t be significantly more than a few database records sitting around. The avatar image is probably more data than the rest of the account metadata.

Asking how the job will be seems like the wrong question to me, when you’re trying to answer what the studies will be. I would check with your uni instead.

Studies could be focused on maths or not. Jobs can be focused on maths if you’re developing models and ml in depth, but is likely unnecessary if you’re just training and assessing and embedding models, outside of maybe statistics, depending on use and needs.

LLM is a hot topic, but I suspect it did not overturn general ml.

I’m not in the field specifically, but we explore and discuss it at work, other team trained a model, and we use coding tools.

Well, shit. I only recently started using Trilium. If it were only some Claude commits, I would probably ignore it. But with your documented bad/issue experience, I guess I’ll seek alternatives.

I checked for price now. Not even PC yet. It costs 80 € or 100 €.

Great cgi movie :p