anavat

"Capability stays inside the curriculum" implies that even much more advanced models are not able to go far beyond their pre-training data. Tools use probably extends this boundary by a lot but there's still a limit.

Yes, it was a criminal offense since 1961 (I didn't hear about a special law for gypsies but it could be). So there were homeless and unemployed, but only those who really wanted to be. It took effort.

It was far from utopia but it wasn't orwellian. It was different in a way that's hard to explain if you didn't experience it. The lack of public goods, lower living standards, and the oppressiveness of the state were somewhat compensated by the peace of mind coming from knowing that you couldn't become unemployed or homeless. This is what a lot of people miss when they say they miss USSR.

No, it actually continues running headless on the host, and you can reconnect from another laptop or mobile phone, or even ssh to the host and attach to the session. At least Codex desktop app works this way.

Sorry, it was more of a "shower thought" than a serious statement. The borders are blurry. Maybe it is alive. Maybe everything is alive. Maybe it is all relative. Humans classify as separate entities only on certain zoom level - zoom in and you get a symbiotic colony, zoom out and all you see is a population. Maybe AI is alive but only in the short time during inference. Maybe a continuously running claw is closer to a multiple generations of an organism going through evolution on each inference step. I am just musing.

It comes down to "what stays the same when all its atoms are replaced" and "how can we tell it is the same entity and not just an identical but different one". I think the answers are "shape" AND "continuity".

E.g. all "atoms" of a glider in Conway's Life get replaced every couple turns but an observer can tell it is still a glider because it keeps the shape, and it the same glider because it continues it's previous state.

This makes AI not quite alive because it's missing the continuity.

Because of compound efficiency and technological enablement.

Or, risking to beat the metaphor to death, because over a span of time I'll cross many more deserts than I would have on a camel, and because I'll cross deserts that I wouldn't even try crossing on a camel.

It is simple. Continuing your metaphor, I have a choice of getting exactly where I want on a camel in 3 days, or getting to a random location somewhere on the other side of the desert on a helicopter in few hours.

And being a reasonable person I, just like the author, choose the helicopter. That's it, that's the whole problem.

Okay, I'm gonna shoot myself, "ensloped" it is.

"I find your email deeply ensloping."

"This marketing campaign is going to enslope a lot of people."

"Feeling ensloped, I closed Instagram and looked out the window".

There needs to be a word for the feeling of sudden realization that you're reading an AI-generated text (or watching an AI-generated video) where you expected it to be human-authored.

Yeah, strong crypto bubble vibes. Everyone is building tools for tool builders to make it easier to build even more tools. Endless infrastructure all the way down, no real use cases.

Holy... That just unlocked a memory. When I was a 10 y.o. kid and a member of Young Pioneers organization in the late Soviet Union, we had a routine elderly care assignment. There was literally a task "to help N lonely elderly people next week"! We were split in pairs and provided with a list of addresses of nearby "targets" and a general guidance on what this help may constitute, like helping them with groceries, walking a dog, house cleaning, etc. And I remember there was a nice old lady who were treating us with tea and pies, and showing us some old photos, and who we then visited few times outside of the assignment just because she welcomed us and we liked it there. For me it was just some secondary quest that I barely registered and completely forgot about until now, but now I wonder if it was a big deal for her

Thanks for working on this! Language learning really needs a breakthrough.

Now, I tried the web app and chose to learn Greek as a beginner. And while I had better experience with your app than with ChatGPT or Gemini voice modes, I still got lost 5 minutes in because the AI tutor doesn't seem to have a plan for me, nor does it "see" my struggles. For example, after asking me about a hobby, it gives me a long sentence in Greek about how how it is nice to hike in mountains. Being absolute noob I cannot reply to it, nor even repeat it. And I don't even know what it is expected from me at the moment. A human tutor here would probably repeat a part of the sentence with a translation and ask me to repeat, or would explain something. The AI just sits there waiting for me to make a sound, and when I make it, it goes on on a tangental subject of beach vacations. :)

Again, this is still relatively not bad, and I'm going to give it another try.