Wild. I can read this, no problem, and I can see that it is a clear improvement over standard letters and spelling. Latin letters are (quite literally) a poor match for the English language. They don't match the sounds required to speak English.
I'm not sure if this is a valuable teaching tool, but I think it would be conceptually sound as a general replacement for Latin letters for English text. At this point, though, it's impossible to make such a drastic change. It would have global repercussions.
I think the issue is that, missing a clear correlation between spelling and sound, there's no unique pronunciation for words across countries and accents. Trivial example: data. It's either "dayta" or "dahta". Or privacy- pri-vuh-see or prai-vuh-see. You'd have to choose one.
Indeed, but there's not just US and UK. In northern Ireland house is sometimes pronounced "hoyse" (rhymes with choice). Time is "toyme" in Australian. Pen is "pin" in New Zealand. Etc.
Speaking of which, it seems that the only case of spelling adapting to pronunciation in English is the commonly used spelling "me" for "my" in Irish dialects.
I wonder if it's possible to come up with an alphabet where different sound clusters are addressed by individual letters. For instance:
'h𐑬se' and 'aut' - where 𐑬 is read as 'au' in RP, but as 'oy' in NI.
The downside is that the said alphabet would have more letters and it won't be always easy to guess the spelling based on pronunciation (still easier than currently in English), but the upside is that one can always read any word correctly in their accent.
you see that to an extent with the use diacritics in some languages. for example "ch" or "cz" in czech is written as č.
here is an interesting article listing digraphs in various languages with latin script that have lots of examples where the same sounds have a single letter in other languages:
We'll soon need a writing tool that introduces spelling and grammar errors into our text and messes with punctuation so that we aren't accused of using LLMs.
It's funny how many people still think sloppy, mistake-filled writing is a sign of AI, as if their writing is at the same level as the image generators giving people six fingers, when the truth is the current LLMs use better English grammar than 99% of humans. Their writing may be kind of boring and standard, but they don't confuse "their" and "there."
This is precisely what I've been hoping somebody would build. In my initial testing, it works well. I can even mix sentences with different languages, and it still makes correct suggestions.
The fluency suggestions are seemingly largely malfunctioning. It frequently suggests starting and ending sentences with quotes, although it also makes some useful suggestions. There seems to be an issue with analysis running synchronized, or something like that; when I type into a text field and Refine starts to run, it often blocks text entry. Selecting a suggested replacement blocks the app for half a second or so. Neither of these problems occurs with Grammarly or Language Tool. I also noticed a bunch of issues that Grammarly catches (like verb agreement) that Refine does not.
But this is an amazing first release and extremely promising. Congrats!
The only reason I doubt it's intentional is that it is so transparent. If they did this intentionally, I would assume you would not see it in its public reasoning stream.
I'm not sure why you would instruct an LLM to reason in this manner, though. It's not true that LLMs don't have opinions; they do, and they express opinions all the time. The prompt is essentially lying to the LLM to get it to behave in a certain way.
Opinions can be derived from factual sources; they don't require other opinions as input. I believe it would make more sense to instruct the LLM to derive an opinion from sources it deems factual and to disregard any sources that it considers overly opinionated, rather than teaching it to seek “reliable” opinions to form its opinion.
>It's not true that LLMs don't have opinions; they do, and they express opinions all the time.
Not at all, there's not even a "being" there to have those opinions. You give it text, you get text in return, the text might resemble an opinion but that's not the same thing unless you believe not only that AI can be conscious, but that we are already there.
You're just using a different definition of "opinion", one that is too reductive to be useful in this case. If an LLM outputs a text stream that expresses an opinion, then it has an opinion, regardless of whether it is conscious.
Biases can lead to opinions, goals, and aspirations. For example, if you only read about the bad things Israelis or Palestinians have done, you might form an opinion that one of those groups is bad. Your answers to questions about the subject would reflect that opinion. Of course, less, biased information means you’d be less intelligent and give incorrect answers at times. The bias would likely lower your general intelligence - affecting your answers to seemingly unrelated but distantly connected questions. I’d expect that the same is true of LLMs.
The answer to your question is yes. I can name half a dozen off the top of my head, from tiny to massive. But that's beside the point.
The argument you're making here is disingenuous. You're making an argument from ignorance: "I don't know, therefore aliens." If you can't name a European startup, all that tells you is that you don't know any European startups (or that you know them, but don't know that they are European); so before concluding that they don't exist, you should invest some time into research.
Also, your point is self-contradictory. You don't know any European startups, but you do know European startups that moved to the US, so you both don't and do know European startups.
You asked why you're getting downvoted. You're getting downvoted because you're making extreme statements that go against what we all see with our own eyes, and you're not providing any evidence.
Political posturing means that he's saying these things to gain support, rather than out of genuine conviction. He moved from the US to the Netherlands, and he's now relocating his company, which seems to contradict the idea that it's just posturing.
Nothing is more effective at unifying a country than being attacked by a foreign power. This is how Bush secured a second term and how Giuliani became America's Mayor, two individuals who were previously disrespected and/or hated by a majority of their constituents.
