Will most people stop studying Chinese now that AI is so good?

[realmayo]

On the other AI thread I wrote the following, which makes me want to ask whether people think the number of people studying foreign languages is set to decline dramatically.
 

Quote

 

I'm starting to wonder if there won't be any long-lasting AI language learning programmes, for the simple reason that most people will stop learning languages.

 

If you go to a foreign country you will surely be able to just wear a regular bluetooth earphone with microphone: when you speak your phone will output your words into the local language, and send any replies into your ear, translated into your own language.

 

I assume this technology already exists.

 

 

[Moshen]

"Very high quality translation" is not anywhere close to understanding another language and being able to do things in it.

 

I don't think it's even relevant to why most people who want to (vs. are forced to) study foreign languages do so.

[PerpetualChange]

No. For starters, it's not even good. 

[Woodford]

Whenever I'm looking at a video of a person speaking a foreign language, and the video features things like English subtitles or an English overdub, I still think, "Wouldn't it be cool to understand the language without the aid of those things?" Whenever I've attempted to speak Chinese with a Chinese person without the aid of technology, there's been a really great element of personal connection ("Wow, you took the time to learn my language!").

But what would it take to discourage me from learning a foreign language? Probably if Elon Musk's Neuralink company successfully allows people to "download" languages into their brains. You could possibly argue that the experience of language learning would still be worth the effort, but that would at least give me reason to pause. At any rate, I think technology is lightyears away from that point.

[matteo]

I don't believe such a  technology would have a big impact on someone's motivation if they are really into the language and culture and want to establish a deeper connection etc.  If we want to make a comparison: I like drawing and painting and I'm not going to stop just because now I can use Midjourney to create a better watercolor than I will ever be able to in no time (by the way, is it ethical to trawl the internet for art so we can replace humans? I don't think so). Convenience is not always the point - although it seems like we are being pressured really hard to believe it is. 

 

However, I do think it will be a further hurdle to overcome for intermediate students; I can imagine many situations in which an impatient listener will just prefer shoving a phone with its simul-translation app in your face rather than having to deal with your less-than-perfect Mandarin/English/whatever.

[becky82]

If anything, I would guess that AI will increase people's motivation to study foreign languages.  At the very least, it's polite and motivating, unlike some of the teachers I've encountered.

 

I'm not sure if I'm using the same AI as others, but I'm flabbergasted at how good Sage is at Chinese.  Sure, it's imperfect, but so is literally every resource, and you overcome this by cross-checking resources against each other.  I probably have 10+ videos on my YouTube channel now where I chat with Sage in Chinese in imaginative ways (conducting interviews, writing choose your own adventure stories, playing word games, and so on).

 

I have about 4 Chinese friends who are now using ChatGPT or some variant for studying English.  This is going to affect my life in China: fewer people are blatantly trying to exploit me as a free English teacher, as ChatGPT gives better responses and has better knowledge of IELTS etc.

[YuehanHao]

Great question.

A lot of "no"s so far. Not too surprising given the theme of this forum and the time and effort many have already invested. At this point, I would honestly say no too.
Yet if I imagine myself as a youth starting all over again tabula rasa, and if I genuinely accept the premise of the question that a translation method exists that is immediate and high quality (which some responses above explicitly deny), no doubt I'd be less motivated to perform the drudgery (rewarding as it might be in the end) of learning a language -- in the same way as I have "learned" to avoid solving a math problem by hand that I could simply punch into a calculator or whip with a computer program.

It may go too far, however, to suggest that AI-based learning tools might therefore not exist in the future. As long as humans are studying anything, I believe language will be among the topics of interest. Indeed, there are a select few who even today practice as hobbies time-honored skills (perhaps quilting or candle-making or ... ) that, at the moment, may appear to some as obsolete or antiquated?

[rongminshan]

In my opinion, no. AI is no replacement for humans. Even the best AI will make mistakes because the models used to train the AI doesn't cover every single possibility (among other things).

 

Plus, people learn languages for reasons that popping a sentence into an AI and having that pop the answer out does not fufill.

 

I won't argue that AI is a fabulous tool to help learn a new language though

[dakonglong]

This is an interesting question that I would also like to know the answer to, but I don't think you're going to get a balanced results here.

 

Generally, I think that all technology-based language tools (for example, Google Translate) reduce the unique value associated with learning a foreign language. Put another way, they narrow the range of activities that actually require you to know the language. Before Google Translate, if I wanted to order a cup of coffee at a Chinese Starbucks, I would probably have had to know at least a few words of Chinese. Now I can type my order into Google Translate, show it to the cashier and get the exact same result. Ordering coffee is no longer a task that requires actual knowledge of the Chinese language. AI tools have done the same kind of thing, but now generation of short-form text (essays, blog posts, emails) no longer requires knowledge of the Chinese language either, so the field of potential applications for Chinese has narrowed further. I suspect that it will continue to narrow over time as technology advances.

 

Will that reduce the percentage of people that decide that the benefits of learning Chinese are worth the costs? Probably, yes (my opinion). If the number of applications for a skill decreases, but the costs of acquiring the skill stay roughly the same, it should make acquisition of the skill less appealing. The most similar example I can think of is the case of a watchmaker. I suspect the percentage of people in the world who become watchmakers has declined from 100 years ago. There is still a use for the skillset (making high-end status-symbol watches like Rolexes) but now that you can machine-make 99% of watches, it's probably no longer worth the intense training required to acquire the skill (unless it's a passion).

 

Now for the question that was asked. Will I, specifically, be less likely to learn another language now. Probably not, because language learning is a hobby of mine and I suspect that is true for a lot of others here as well. My Chinese skills are already almost useless from a real-world perspective because I can 1) use Google Translate or ChatGPT to serve most of my actual needs in Chinese, and 2) there are no material benefits to me of learning Chinese because there are probably 100 million Chinese people who speak better English than I do Chinese, so I doubt I will ever find a job or business opportunity that requires it. I just continue to learn it because I love it!

 

TLDR: I think these tools will reduce the number of people that decide to start learning the language in the first place, but not stop those with a passion for learning languages from continuing to learn.

[Moshen]

Why do lots of people prefer to read a novel or poetry in the original language if they can?  The availability of excellent translations in their own language has little or no bearing on that desire.

 

Similarly I'm thinking back to my Ph.D. studies where we were required to have a mastery of at least one foreign language relevant to the discipline (French, German, Greek, Latin).  I again doubt very much that AI would put a dent in that idea.  There are simply too many important distinctions between languages to believe that AI would fill the gap.

 

I agree that for people who have purely transactional interest in languages, AI may put a dent in their desire to study languages.  But is that the main reason most people study languages - when that's a choice?

[Jan Finster]

The answer to your question "Will most people stop studying Chinese now that AI is so good? " is a resounding "yes"!

 

This board does not represent "most" people and "most people" would not spend 10+ years studying Chinese so that one day they acn read Tang dynasty poetry in its original form. People look for convenience and look at opportunity costs. If we one day have AI translation down to almost perfect and seamlessly integrated like a Babelfish*, then "most" people would invest their time in other endeavours.

 

Since we have lighters nowadays, most people no longer bother learning how to make fire from scratch. Thousands of years ago, virutally everyone knew how to do this ?

 

*The name "Babelfish" comes from the science-fiction classic The Hitchhiker's Guide to the Galaxy," referring to a creature which could translate brain waves, and in effect, translate all languages for anyone who had a Babelfish placed in the brain

[mouse]

On 5/29/2023 at 7:26 AM, Jan Finster said:

If we one day have AI translation down to almost perfect and seamlessly integrated like a Babelfish*, then "most" people would invest their time in other endeavours.

 

If.

[Tomsima]

On 5/29/2023 at 7:26 AM, Jan Finster said:

Since we have lighters nowadays, most people no longer bother learning how to make fire from scratch.

 

But interestingly, we do still have matches, which are actually better than lighters in a lot of situations, especially lighting fires. If you don't have an understanding about how to build a fire, a lighter is also completely useless. 

 

The poll is also skewed by the fact that many people who choose to learn Chinese do it knowing that it is supposedly more 'difficult' than other languages. There is a motivation to learn there that goes beyond your average business language class. It seems to me that AI is more a challenge to the Michel Thomas or Pimsleur models, as it will remove the need for phrasebook style learning.

[becky82]

I guess the other thing we should think about here is: Who are the people learning Chinese as a foreign language?

• People who learn Chinese for fun or self-improvement (or to flex).
• People wanting to go to, or are in, a Chinese-speaking country, perhaps for travel or work or to accompany a partner.
• People learning to pass some exam (HSK, TOCFL, etc.), as required by e.g. a Chinese university.
• People who are, or have relatives who are, ethnically Chinese or Taiwanese.
• Ethnic minorities in China.
• Singaporeans and Malaysians.

Is there any reason to think these people will give up learning Chinese due to AI?

[Jan Finster]

On 5/29/2023 at 11:05 AM, mouse said:

If.

When....(!)

 

It is just a matter of time. I am actually surprised I come across as somewhat of an "early adopter" on this forum. Typically I am a late adopter (I typically start loving music bands or fashion when they are already considered "out") ?

 

But, the discussion here reminds me a lot of the discussion I had with friends around the year 2000, when digital cameras came of the market. I told me buddies not to buy an analog SLR camera since digital cameras were the future. He said they would never get as good as analog models. I saw my buddy 2 weeks ago and asked him if he was taking his analog SLR on his trip to Africa. He just laughed...

[Jan Finster]

On 5/29/2023 at 1:10 PM, becky82 said:

People learning to pass some exam (HSK, TOCFL, etc.), as required by e.g. a Chinese university.

 

You are right, this post is very West-centric. I guess a significant proportion of students at Chinese unis are Asians or Africans. For them, studying in China is often a great choice because maybe education in their home country  is inferior (?!). 

[realmayo]

On reflection I kind of squashed two different questions together. Most of us here are enthusiastic about language learning so it's not surprising that we generally expect to keep learning languages.

 

One scenario I had in my head: say I was going to visit Vietnam for a few weeks, and was also quite interested in the language. In the past I would have considered learning some Vietnamese for say six months prior to visiting, for two reasons: interest, and use.

 

But what if, in the very near future, people are very comfortable using the microphone/speaker functions of their phones as high quality real-time interpreters? I think it would mean that speaking intermediate Vietnamese in Vietnam would bring no "use" benefits. People would be impatient at listening to your modest Vietnamese when you could just use the interpreter function. A bit like someone from England insisting on speaking poor Dutch when in the Netherlands.

 

For me one of the pleasures of speaking a foreign language is having conversations with foreigners that would not otherwise be possible. That will no longer be the case - or at least, won't be the case until you've reached quite a high level of fluency.

[Moshen]

I wondered if any research had been done that would shed some light on this question, and in this article...

https://www.cusef.org.hk/en/cusef-blog/our-research/the-factors-fueling-americas-interest-in-learning-mandarin-chinese

...I found a couple of things.

 

1)The article noted that technology has helped fuel an increase in learning Chinese because various apps etc. make language learning easier.

 

2)Xi Jinping and Pres. Obama agreed on a political push to increase the learning of Chinese in the US.

 

3)The article also noted the increased interest in Chinese from heritage speakers.  

 

Of these factors, #1 and #3 are likely to continue to increase for some time, I believe, unaffected by AI.  #2 probably not in the current political climate.

[Jan Finster]

On 5/29/2023 at 2:35 PM, Moshen said:

1)The article noted that technology has helped fuel an increase in learning Chinese because various apps etc. make language learning easier.

 

I am not sure you can lump AIs like ChatGPT into the same group as Duolingo and other apps. Sure, apps made language learning easier, but not unnecessary. Good AIs might not only make language learning easier, but somewhat unnecessary.

[Moshen]

Quote

But what if, in the very near future, people are very comfortable using the microphone/speaker functions of their phones as high quality real-time interpreters? I think it would mean that speaking intermediate Vietnamese in Vietnam would bring no "use" benefits. People would be impatient at listening to your modest Vietnamese when you could just use the interpreter function. 

 

Maybe I have a poor imagination.  But I was thinking back to my adolescence when I was hitchhiking in Europe and feeling great while using my French or Spanish with the locals in the cars.  (Presumably they had warm fuzzy feelings from the exchanges also - I have no reason to believe otherwise.) When I try to imagine the same feelings being produced by conversation with strangers in a car through holding up phones to translate, this does not compute.