aifaq.wtf

Just like how social media giants spend the majority of their time and resources regulating English-language content, there is definitely not enough attention paid to the abilities of AI tools in non-English languages for things other than translation.

A while back iMEdD analyzed political speeches with ChatGPT, translating them into English prior to the analysis. I had thought hey, you should just do it in the original Greek, but looking at this maybe I was wrong!

I feel like we've heard this a thousand times, but it's going to keep being a problem. Story here.

Translation is the one part of AI tooling that I'm most pessimistic about. While it could be used to really increase access to information, it's really just going to be used as an outlet for low-quality content that's disrespectful to the audience and anchors English even further as the language of the internet.

Meta released a new model that can do all sorts of speech- and text-based communication tasks. It's called Seamless, and you can find the model here.

Speech-to-speech translation is super fun, yes, but I love that they highlight that it (supposedly) understands code switching, aka bouncing back and forth between different languages. That's always been a tough one!

I tried the demo here and mostly learned my Russian is horrible:

Tried again with my pathetic grasp Japanese, and it thinks I'm speaking Hindi. I feel like these systems are usually pretty good at coping with my inability to speak other languages well (or at least Whisper is), so I don't know if I'm especially bad today or this model's flexibilty is also a potential downside.

Forget subtitles: YouTube now dubs videos with AI-generated voices – the background to know for this one is the role of dubbing in YouTube fame, featuring MrBeast conquering the world through the use of localized content.

There's definitely a need, as "as much as two-thirds of the total watch time for a creator’s channel comes from outside their home region." But what's the scale look like for those who make the transition?

The 9-year-old Russian YouTuber, Like Nastya, has created nine different dubbed channels, expanding into Bahasa Indonesia, Korean, and Arabic to reach over 100 million non-Russian subscribers

To make things personal: my boring, mostly-technical English-language YouTube channel has India as its top consumer, currently beating out the US by just under one percentage point (although India was twice as common as the US a couple years ago).

Although I realize that yes, that's cheating, since almost 90% of those with a Bachelors degree in India can speak English.

One approach for classifying content is to translate the text into English, then analyze it. This has very predictable side effects if you aren't tweaking the model:

One example outlined in the paper showed that in English, references to a dove are often associated with peace. In Basque, a low-resource language, the word for dove (uso) is a slur used against feminine-presenting men. An AI moderation system that is used to flag homophobic hate speech, and dominated by English-language training data, may struggle to identify “uso” as it is meant.

Looks great!

Multilingual AI is a vey real issue, with literal lives on the line. Mostly because Facebook wants to use AI to moderate hate speech instead of using actual human beings (although that has problems, too). Ignoring content moderation on social media in non-English countries goes much worse than you'd imagine.

Lots of ways to contribute, from the Aya site:

If you're interested in this kind of thing, a term to search for is "NLP and under-resourced languages." It absolutely goes well beyond that, but it's a good starting point.