@@masontheinterpreter but what if those people bump into me because they can't see me with those tiny eyes. Just kidding, keep up the good work, you seem like a great guy!
Personally, I learned French, and I'm learning Turkish to interpret for the Turkish army in Africa. (for those curious as to why, I am Libyan) I cannot see the army using an American program to interpret, furthermore, during military training they need a present person, as it's more versatile and adaptive. Not to mention, internet isn't universal, and sometimes it's merely not desired due to confidentiality. The same goes for political, legal, medical and high profile events, due to prestige, accuracy and legal liability. Even for tourism, most tourists tend to be in average older, and thus less likely to ditch human solutions, and some genuinely enjoy the company of their interpreter, thus we ought to combine this with the skills of a guide, there no AI can beat you. For those who use Google Translate, they might translate the signs and the menu, but this doesn't substitute The business would prevail, but it would be deeply unsettled. Literary translation would also continue, as AI's translation is tasteless. Translating a UA-cam video, dubbing, narrating, transcribing, translating essays and articles, non-fiction books etc.. all would cut down on workers and only keep them for supervision and dispose entry-level ones. Language teachers are also going nowhere, if your job is threatened you can always teach, supervision and interpersonal aspects make up for the sophistication of AI. The checkers are also at a risk, as AI can spot many errors even in grammar, they ought to watch out, other than the high profile works, the entry-level are going to trouble, People can already learn an entire language by themselves, no further AI is needed, moreover, any hyperrealistic AI for speaking practice would fall flat, it's more worthy to speak out loud imagining a potential conversation than to practice with a banal AI. But I can see some application for it in the form of an immersive video game, where you can the reps and confidence to eventually put yourself out there. My regards.
Thank you for the comment and great to hear from an interpreter with a different background. The examples you have provided with your reasonings, although they are convincing and completely understandable, I do feel that you might be overlooking the potential progress of AI, which will be most likely exponential. Of course, I am not an AI expert nor a psychic... but my guess is that whatever we are seeing and picturing when we think of AI's abilities and capabilities today, it will be drastically underestimated compared to what we will see in the future as AI gets developed further. Once again, it is fair to say that this isn't something that will happen tomorrow and those with skills and knowledge will have a higher probability to outlast other workers. Thank you for wonderful insights!
I agree that matters of trust are different between humans and AI. Moreover, AI is the property of Business Owners, who may have many interests that interfere with honestly brokering communication. An interpreter has a limited capacity to form relationships and subvert their role to serve interests beyond the interpreted communicators. An AI can smoothly integrate with a global network of interests that serve its owners. It will be able to instantaneously infuence micro-deviancies in interpretation to influence outcome, or to plant ideas or notions that are of limited relevance to the conversation at hand, but fit a separate global designs... this is especially sensitive for diplomatic and business communications. When AI can identify social networks, it can target communications indirectly by subtly planting ideas in the social milieu of the diplomatic or business communicators. This can happen even if there is no interpretation necessary. AI 'Companions' and AI Therapists, and AI Tutors, etc., can all do the same thing. It is rational to distrust AI in every communication. By the time AI is really good at it, we may not even be able to detect we've been manipulated. After all, Humans have been gaslighting each other ever since there was a value motive to do so. I predict, IMO, that National Security communications will be hackable, and so they will not use AI interpreters. Billionaires doing business will not use them, because they want the best, which is what Diplomats and National Security agencies use. People who want to be Billionaires are likely to do the same. More common users will be driven by cost, but will pay with trustworthness, and it will be like people who don't vote: I'm not important and I'm not worth it for AI to deceive me. Believe what you will. A small deception of two billion people leads to shifts in global trends.... you are worth hacking, but you'd never know it.
I agree. It will happen sooner than later. Just consider, openAI Whisper model being capable of understanding nuances in different contexts and settings, even when stuttering or out of incomplete ideas or even overlapping speakers which is something we see on a daily cadence as interpreters. Now, the only thing that would probably hold up its expansion would be the fact its still going to come out expensive as hiring an opi, and all users must have a reliable connection to have the natural latency in between interactions even with the newest 4o. But all these issues may very well be addressed in a couple of years. But until then we all can say that interpreting is fun.
Thank you for your input! As I mentioned in the video, I do not doubt AI's capabilities in this area, particularly in the near future. However, I believe there are additional limitations beyond the cost and bandwidth issues you mentioned. Another factor that I could think of is the subjectivity involved in interpretation. While the primary goal is to act as a conduit obviously, interpreters inevitably bring their own perspectives to understanding the meaning of an utterance within a limited context. Since I am no expert in AI, I do not fully understand how AI would tackle this challenge, but it certainly seems complex to me. 😂
Besides people talking over each other and the general chaos of humans being humans that AI might have trouble dealing with, patient confidentiality is also a big issue. Also, like you mentioned, people might be comforted by simply talking to a human being and not a robot. Never-the-less, I do agree that interpreters jobs will eventually be impacted by A.I., especially interpreters that work in low-stakes situations where accountability isn't an issue.
I appreciate your insight! As you mentioned, I also believe the low-stakes situations will be most likely utilizing AI before others, especially many of the customer service/relations departments of many businesses. If so, skilled interpreters/linguists will become even more distinguishable and valuable in the labour market, hopefully.
You’re right. Right now, AI depends much on the “input” (the data that has already been trained). It means, if all LEP’s languages books were used to train a strong AI model, it would be as fluent as it can. The input here is comprised of training data (accents, dialects, slangs…) Currently, AI is only at the level of imitating what is "taught" but cannot maintain and invent itself => So people are aiming for these.
Thank you for your input. All I can say at this very moment is that I am very surprised at how quick things are being improved compared to just several years ago!
Ai is doing things in areas of life we actually didnt think it would but tbh, ai will probably be better at language and understanding of different languages around the world when trained with the voice data of those people, especially music, im sure open ai is already training its models with all the music on the internet but as an engineer myself… maybe one day
Great insights! AI is jeopardising all! Nonetheless, although not an interpreter myself, the numbers of employment diminishing due to AI should be compensated for by a rising number of other kinds across various sectors! Hopefully!
Sorry but, way I see things, I think we are AT the peak for human interpreters.. We're also nearing the peak for translation voice actors. AI is progressing far too rapidly now, to doubt what's coming.. By this time next year, there won't be any job a human can do with their voice, that an AI won't be able to do better and cheaper.
Bro how do you even see? I've never met anybody with eyes as tiny as yours
That’s probably because you never go outside and socialize with other people! 😉 Go outside and meet some more people!
@user-xr4vp3nl7y you are disgusting
@@masontheinterpreter but what if those people bump into me because they can't see me with those tiny eyes. Just kidding, keep up the good work, you seem like a great guy!
@@Josephsmith-p4h your passive agressive comment sucks!
@@Josephsmith-p4h ssawft
I love how friendly you keep the topics conversation, not into translating but really enjoy your videos ☺️
Thank you for the kind comment! Greatly appreciate it! 😊
Personally, I learned French, and I'm learning Turkish to interpret for the Turkish army in Africa. (for those curious as to why, I am Libyan)
I cannot see the army using an American program to interpret, furthermore, during military training they need a present person, as it's more versatile and adaptive.
Not to mention, internet isn't universal, and sometimes it's merely not desired due to confidentiality.
The same goes for political, legal, medical and high profile events, due to prestige, accuracy and legal liability.
Even for tourism, most tourists tend to be in average older, and thus less likely to ditch human solutions, and some genuinely enjoy the company of their interpreter, thus we ought to combine this with the skills of a guide, there no AI can beat you.
For those who use Google Translate, they might translate the signs and the menu, but this doesn't substitute
The business would prevail, but it would be deeply unsettled.
Literary translation would also continue, as AI's translation is tasteless.
Translating a UA-cam video, dubbing, narrating, transcribing, translating essays and articles, non-fiction books etc.. all would cut down on workers and only keep them for supervision and dispose entry-level ones.
Language teachers are also going nowhere, if your job is threatened you can always teach, supervision and interpersonal aspects make up for the sophistication of AI.
The checkers are also at a risk, as AI can spot many errors even in grammar, they ought to watch out, other than the high profile works, the entry-level are going to trouble,
People can already learn an entire language by themselves, no further AI is needed, moreover, any hyperrealistic AI for speaking practice would fall flat, it's more worthy to speak out loud imagining a potential conversation than to practice with a banal AI.
But I can see some application for it in the form of an immersive video game, where you can the reps and confidence to eventually put yourself out there.
My regards.
Thank you for the comment and great to hear from an interpreter with a different background. The examples you have provided with your reasonings, although they are convincing and completely understandable, I do feel that you might be overlooking the potential progress of AI, which will be most likely exponential. Of course, I am not an AI expert nor a psychic... but my guess is that whatever we are seeing and picturing when we think of AI's abilities and capabilities today, it will be drastically underestimated compared to what we will see in the future as AI gets developed further. Once again, it is fair to say that this isn't something that will happen tomorrow and those with skills and knowledge will have a higher probability to outlast other workers. Thank you for wonderful insights!
Finalement j'ai trouvé quelqu'un Libyen qui est intèrissant de l'apprentissage de langes étrangers comme moi 😩💕💕💕! Bonne chance !
I agree that matters of trust are different between humans and AI.
Moreover, AI is the property of Business Owners, who may have many interests that interfere with honestly brokering communication.
An interpreter has a limited capacity to form relationships and subvert their role to serve interests beyond the interpreted communicators.
An AI can smoothly integrate with a global network of interests that serve its owners. It will be able to instantaneously infuence micro-deviancies in interpretation to influence outcome, or to plant ideas or notions that are of limited relevance to the conversation at hand, but fit a separate global designs... this is especially sensitive for diplomatic and business communications. When AI can identify social networks, it can target communications indirectly by subtly planting ideas in the social milieu of the diplomatic or business communicators.
This can happen even if there is no interpretation necessary. AI 'Companions' and AI Therapists, and AI Tutors, etc., can all do the same thing.
It is rational to distrust AI in every communication. By the time AI is really good at it, we may not even be able to detect we've been manipulated. After all, Humans have been gaslighting each other ever since there was a value motive to do so.
I predict, IMO, that National Security communications will be hackable, and so they will not use AI interpreters. Billionaires doing business will not use them, because they want the best, which is what Diplomats and National Security agencies use. People who want to be Billionaires are likely to do the same.
More common users will be driven by cost, but will pay with trustworthness, and it will be like people who don't vote: I'm not important and I'm not worth it for AI to deceive me. Believe what you will. A small deception of two billion people leads to shifts in global trends.... you are worth hacking, but you'd never know it.
Great insight. Thank you for the comment!
I agree. It will happen sooner than later. Just consider, openAI Whisper model being capable of understanding nuances in different contexts and settings, even when stuttering or out of incomplete ideas or even overlapping speakers which is something we see on a daily cadence as interpreters. Now, the only thing that would probably hold up its expansion would be the fact its still going to come out expensive as hiring an opi, and all users must have a reliable connection to have the natural latency in between interactions even with the newest 4o. But all these issues may very well be addressed in a couple of years. But until then we all can say that interpreting is fun.
Thank you for your input! As I mentioned in the video, I do not doubt AI's capabilities in this area, particularly in the near future. However, I believe there are additional limitations beyond the cost and bandwidth issues you mentioned.
Another factor that I could think of is the subjectivity involved in interpretation. While the primary goal is to act as a conduit obviously, interpreters inevitably bring their own perspectives to understanding the meaning of an utterance within a limited context. Since I am no expert in AI, I do not fully understand how AI would tackle this challenge, but it certainly seems complex to me. 😂
Besides people talking over each other and the general chaos of humans being humans that AI might have trouble dealing with, patient confidentiality is also a big issue. Also, like you mentioned, people might be comforted by simply talking to a human being and not a robot. Never-the-less, I do agree that interpreters jobs will eventually be impacted by A.I., especially interpreters that work in low-stakes situations where accountability isn't an issue.
I appreciate your insight! As you mentioned, I also believe the low-stakes situations will be most likely utilizing AI before others, especially many of the customer service/relations departments of many businesses. If so, skilled interpreters/linguists will become even more distinguishable and valuable in the labour market, hopefully.
You’re right. Right now, AI depends much on the “input” (the data that has already been trained). It means, if all LEP’s languages books were used to train a strong AI model, it would be as fluent as it can. The input here is comprised of training data (accents, dialects, slangs…)
Currently, AI is only at the level of imitating what is "taught" but cannot maintain and invent itself => So people are aiming for these.
Thank you for your input. All I can say at this very moment is that I am very surprised at how quick things are being improved compared to just several years ago!
Ai is doing things in areas of life we actually didnt think it would but tbh, ai will probably be better at language and understanding of different languages around the world when trained with the voice data of those people, especially music, im sure open ai is already training its models with all the music on the internet but as an engineer myself… maybe one day
Indeed! Thank you for the insightful comment!
Great insights! AI is jeopardising all! Nonetheless, although not an interpreter myself, the numbers of employment diminishing due to AI should be compensated for by a rising number of other kinds across various sectors! Hopefully!
Thank you for your kind and insightful comment! I sure hope so as well!
Its happened already. Over.
Interpreters will become like a public notary...used for cases where you need a stamp of official translation
Definitely a possibility! Thank you for your input.
Machine translation has been around for decades....
That is correct, but AI hasn’t been!
@@masontheinterpreter It's the same technology. There hasn't been any major breakthroughs in machine translation.
Sorry but, way I see things, I think we are AT the peak for human interpreters.. We're also nearing the peak for translation voice actors.
AI is progressing far too rapidly now, to doubt what's coming.. By this time next year, there won't be any job a human can do with their voice, that an AI won't be able to do better and cheaper.
Appreciate your input!
Its the only job that i can 💯 say it will be replaced in the near future, but lots of jobs more too
Appreciate the comment!
Yeah they will be homeless
That sounds a bit extreme, but who knows! 🤔