Pages in topic:   < [1 2 3 4 5] >
Poll: How do you feel about AI and LLM?
投稿者: ProZ.com Staff
Tom in London
Tom in London
英国
Local time: 14:16
2008に入会
イタリア語 から 英語
Me too Jun 18, 2023

Yetta Jensen Bogarde wrote:

but Moderately Concerned would be a better option.
Not so much for me personally, at my age I do not envision that AI and LLM are going to snatch my work away in my lifetime.
I am more concerned about the world in general, especially since it is unrealistic to expect that everyone is going to adhere to any new regulations or ethical rules.


I agree. As soon as I leave, the whole world is going to fall apart.


Nikolay Novitskiy
 
Lieven Malaise
Lieven Malaise
ベルギー
Local time: 15:16
2020に入会
フランス語 から オランダ語
+ ...
Abbreviations Jun 19, 2023

I don't understand the fuzz about the use of the abbreviations in the question. I suppose everybody knows what AI is and it took me less than 1 minute to Google the meaning of LLM. It probably costs more time to write a post complaining about the use of those abbreviations.

[Edited at 2023-06-19 05:29 GMT]


Mr. Satan (X)
expressisverbis
 
Philip Lees
Philip Lees  Identity Verified
ギリシャ
Local time: 16:16
ギリシャ語 から 英語
Respect Jun 19, 2023

Lieven Malaise wrote:

I don't understand the fuzz about the use of the abbreviations in the question.

Have you never come across an unfamiliar abbreviation in a source text and had no idea what it means? Have you never had to spend an inordinate amount of time finding out?

I've worked in IT and I keep up with developments, so I know both the abbreviations in the topic title, but a lot of people probably don't know what LLM means, and the inclusion of "I'm not sure what either AI or LLM means" among the answers suggests that person who proposed the topic was expecting just that.

In my view, undefined abbreviations are one of the banes of a translator's existence, and to construct a topic title in this deliberately mysterious way shows a lack of respect for one's colleagues.

It seems I'm not the only one who feels this way.


P.L.F. Persio
Josephine Cassar
Maria Teresa Borges de Almeida
Evgeny Sidorenko
Christopher Schröder
IrinaN
Jennifer Levey
 
Lieven Malaise
Lieven Malaise
ベルギー
Local time: 15:16
2020に入会
フランス語 から オランダ語
+ ...
Digital age Jun 19, 2023

Philip Lees wrote:
I've worked in IT and I keep up with developments, so I know both the abbreviations in the topic title, but a lot of people probably don't know what LLM means


I had never heard of it and as I already mentioned it took me seconds to find out what it meant (first hit after entering "AI LLM": "What is a Large Language Model"?).

Philip Lees wrote:
and the inclusion of "I'm not sure what either AI or LLM means" among the answers suggests that person who proposed the topic was expecting just that.


It's a poll and polls ask for information. One bit of information might be how familiar translators are with AI technology.

Philip Lees wrote:
In my view, undefined abbreviations are one of the banes of a translator's existence, and to construct a topic title in this deliberately mysterious way shows a lack of respect for one's colleagues.


I would have agreed on this if we were still in the age of paper reference books and forced trips to libraries. But if you read a post on Proz you are only a few clicks or taps away from Google.

[Edited at 2023-06-19 08:29 GMT]


 
Maria Teresa Borges de Almeida
Maria Teresa Borges de Almeida  Identity Verified
ポルトガル
Local time: 14:16
2007に入会
英語 から ポルトガル語
+ ...
@Philip Jun 19, 2023

Many years ago, I had a customer who had the bad habit of inventing abbreviations and acronyms, so I decided to start building a glossary, then during my years working in the EU the thing "blow up" and now I have a glossary with 493 pages (!!!???), written in Times New Roman 10…

[Edited at 2023-06-19 11:09 GMT]


Jean Dimitriadis
expressisverbis
 
Denis Fesik
Denis Fesik
Local time: 16:16
英語 から ロシア語
+ ...
From "what do you think" to "how do you feel" Jun 19, 2023

I guess "post-truth" is just that: manipulating people through feelings. Today, feelings sell better and help achieve outcomes more efficiently. And bring more profits to Big Pharma (where are my anxiety pills?). Of course, I did feel some curiosity when faced with an avalanche AI-related content: some of it enthusiastic, some of it alarmist, some of it dismissive. That curiosity is now satisfied, so I'm left with no feelings towards the subject-matter. Maybe AI developers will come up with some... See more
I guess "post-truth" is just that: manipulating people through feelings. Today, feelings sell better and help achieve outcomes more efficiently. And bring more profits to Big Pharma (where are my anxiety pills?). Of course, I did feel some curiosity when faced with an avalanche AI-related content: some of it enthusiastic, some of it alarmist, some of it dismissive. That curiosity is now satisfied, so I'm left with no feelings towards the subject-matter. Maybe AI developers will come up with something really interesting later on. I just don't feel like turning to the machine for lists of life goals, personal productivity tips, or blog posts that sell (people who create that kind of stuff for money are called info-gypsies in Russia, and I can see how ChatGPT could make one perfect info-gypsy). I don't trust it much with my translations either. Now, I'm sitting here in front of my work PC and reading a book (because the managers have no jobs for me just yet, and I've just been told that I'm supposed to be on my last day of vacation). Nothing's leaping or yelling at me from the screen. What if this AI thing is quite tame at this day and age? Will it become more aggressive over time? Anyway, if it does, I'll see what I can do about itCollapse


P.L.F. Persio
Tom in London
Liena Vijupe
Christine Andersen
 
jyuan_us
jyuan_us  Identity Verified
米国
Local time: 09:16
2005に入会
英語 から 中国語
+ ...
Totally agreed Jun 19, 2023

Philip Lees wrote:

In my view, undefined abbreviations are one of the banes of a translator's existence, and to construct a topic title in this deliberately mysterious way shows a lack of respect for one's colleagues.

It seems I'm not the only one who feels this way.


If a survey contains abbreviations that most potential respondents wouldn't understand, they would have sufficient grounds to stay away.

[Edited at 2023-06-19 11:42 GMT]


Maria Teresa Borges de Almeida
 
IrinaN
IrinaN
米国
Local time: 08:16
英語 から ロシア語
+ ...
@ Lieven :-) Jun 19, 2023

Lieven Malaise wrote:

I had never heard of it and as I already mentioned it took me seconds to find out what it meant


It's plain impolite to greet others with abbreviations. Humanitarians thriving on literature and the beauty of languages should not be doing it. That used to be the way for a few millennia before us. Maybe, "Old people of Proz" feel it more acutely simply because we were raised the old-fashion way in every way. It's not even a show-off anymore; to the contrary, it's just a new norm and no second thoughts are given...

How about I invite you for dinner (honestly, I'd love to:-) ) but offer you to cook your own meal since with modern Air Fryers it will only take you a few minutes, just enough for an aperitif in the meantime (or two:-) ).

Oh, the eternal clash of generations... Those darn grumpy old men and women, when will they finally occupy their rocking chairs and leave the office ones for us:-) Be careful what you (plural) wish for...


Lieven Malaise
P.L.F. Persio
Christopher Schröder
Liena Vijupe
 
Kay Denney
Kay Denney  Identity Verified
フランス
Local time: 15:16
フランス語 から 英語
. Jun 19, 2023

Lieven Malaise wrote:

Of course I'm talking about non-creative texts, but one worry does come to my mind about creative texts and ChatGPT, although it doesn't really concern me because I hardly translate any creative texts : apparently ChatGPT is able to make texts sound very fluently, so wouldn't that be a big threat for translators of creative texts ?


In my humble opinion as the translator of creative texts, CheatGPT's style is very bland. My clients like my work because I go to great lengths to find a pithy word that rhymes, fitting idiomatic expressions, and puns, and to adopt the right tone (upbeat, high-brow, business casual etc.) It's the opposite of bland. Sometimes I include a list of links to urban dictionary or competitor websites so that the client won't balk at certain terms (a notable instance was "badass" in a fashion translation). So I don't think CheatGPT will be making me redundant any time soon.
(I only have another four years before retirement anyway.)


Maria Teresa Borges de Almeida
Lieven Malaise
P.L.F. Persio
Christopher Schröder
Philip Lees
Peter Dahm Robertson
Christine Andersen
 
Lieven Malaise
Lieven Malaise
ベルギー
Local time: 15:16
2020に入会
フランス語 から オランダ語
+ ...
Specific styles Jun 19, 2023

Kay Denney wrote:
In my humble opinion as the translator of creative texts, CheatGPT's style is very bland.


I suppose you are talking about the style that is applied if you ask ChatGPT a random question. But apparently you can ask it to apply any specific style on any text you enter. My question is more if ChatGPT would be capable of nearing the creativity that is applied by human translators if you ask it specifically to be creative.


 
Tom in London
Tom in London
英国
Local time: 14:16
2008に入会
イタリア語 から 英語
Yes- that didn't last long Jun 19, 2023

Denis Fesik wrote:

..... I'm left with no feelings towards the subject-matter.....


 
Lingua 5B
Lingua 5B  Identity Verified
ボスニア・ヘルツェゴビナ
Local time: 15:16
2009に入会
英語 から クロアチア語
+ ...
Audience Jun 19, 2023

Lieven Malaise wrote:

Kay Denney wrote:
In my humble opinion as the translator of creative texts, CheatGPT's style is very bland.


I suppose you are talking about the style that is applied if you ask ChatGPT a random question. But apparently you can ask it to apply any specific style on any text you enter. My question is more if ChatGPT would be capable of nearing the creativity that is applied by human translators if you ask it specifically to be creative.


There is audience out there that can't tell a difference between styles (and they don't care either).

Yes, GPT will apply a style if you ask it to, in a robotic way. Style is not only technique, it goes beyond it (includes visceral experiences).

I liked the Kay's typo "Cheat GPT" (or was it intentional?), it really fits it perfectly.


P.L.F. Persio
Philip Lees
Christine Andersen
Kay Denney
Sebastian Witte
 
Lingua 5B
Lingua 5B  Identity Verified
ボスニア・ヘルツェゴビナ
Local time: 15:16
2009に入会
英語 から クロアチア語
+ ...
It thinks stereotypically (like a little child) Jun 19, 2023

Me: Write a short poem (1 verse) about a little mouse. I want three versions of the poem. Version 1: black emotional tone, Version 2: red emotional tone, Version 3: yellow emotional tone

Chat GPT:

Version 1 (Black Emotional Tone):
In shadows it hides, a mouse so small,
Lost in darkness, it fears the fall,
Whispered sorrows, its heart feels sore,
A tiny soul, searching for more.

Version 2 (Red Emotional Tone):
A fiery spirit, th
... See more
Me: Write a short poem (1 verse) about a little mouse. I want three versions of the poem. Version 1: black emotional tone, Version 2: red emotional tone, Version 3: yellow emotional tone

Chat GPT:

Version 1 (Black Emotional Tone):
In shadows it hides, a mouse so small,
Lost in darkness, it fears the fall,
Whispered sorrows, its heart feels sore,
A tiny soul, searching for more.

Version 2 (Red Emotional Tone):
A fiery spirit, this little mouse,
With passion burning, it breaks the house,
Unyielding courage, it takes its stand,
Defiant heart, against all demand.

Version 3 (Yellow Emotional Tone):
A sunny sprite, the mouse so small,
Spreading joy with every scurry and crawl,
Bright as daffodils, its spirit aglow,
A tiny marvel, the world's golden glow.

- - - -

As you can see, it used all the stereotypical elements from big data (black=darkness, red=passion, yellow=sun).

It shouldn't work that way. We should be able to see/perceive/feel black, red and yellow subtones without these stereotypes, but just by skilled phrasing and word manipulation. The stereotypes could work for junior beginner writers (10 year olds).
Collapse


 
Lieven Malaise
Lieven Malaise
ベルギー
Local time: 15:16
2020に入会
フランス語 から オランダ語
+ ...
@Lingua Jun 19, 2023

I'm pretty sure it doesn't think at all .

But your exemple is irrelevant (creative writing comprises also much less sophisticated communication than poetry). My initial concern was and is if ChatGPT is creative enough, or on its way to be creative enough, to threaten the position (or rates) of writers/translators of creative texts. That's a different perspective. You see, it doesn't need to be better than humans, it wo
... See more
I'm pretty sure it doesn't think at all .

But your exemple is irrelevant (creative writing comprises also much less sophisticated communication than poetry). My initial concern was and is if ChatGPT is creative enough, or on its way to be creative enough, to threaten the position (or rates) of writers/translators of creative texts. That's a different perspective. You see, it doesn't need to be better than humans, it would already be enough that the result is close to have an impact on human translation/copywriting.

[Edited at 2023-06-19 14:29 GMT]
Collapse


 
P.L.F. Persio
P.L.F. Persio  Identity Verified
オランダ
Local time: 15:16
2010に入会
英語 から イタリア語
+ ...
It's just a machine in fancy knickers Jun 19, 2023

Lieven Malaise wrote:

Kay Denney wrote:
In my humble opinion as the translator of creative texts, ChatGPT's style is very bland.


I suppose you are talking about the style that is applied if you ask ChatGPT a random question. But apparently you can ask it to apply any specific style on any text you enter. My question is more if ChatGPT would be capable of nearing the creativity that is applied by human translators if you ask it specifically to be creative.


A machine – or "a natural language processing tool driven by AI technology" – cannot be creative, no matter how much input you feed into it. It only processes information. It's not sentient. You can ask it specifically to be creative until the cows (or the electric sheeps, apparently dreamt of by androids) come home, but you'd just waste your time. Give it all the madeleines in France, it still won't produce the Recherche.

Many of you have probably read about what happened at Clarkesworld, a sci-fi magazine: https://www.nytimes.com/2023/02/23/technology/clarkesworld-submissions-ai-sci-fi.html
It was a case about applying AI to creative writing, to which creative translation (and the same goes for localisation and transcreation) is closely related. The chatbot-generated stories were easily spot, the writing "bad in spectacular ways."

Imitating a style (or even the style of every single novelist, poet, journalist, non-fiction writer, copywriter, content writer, and so on, who ever walked/is walking this Earth) is not the same thing as coming up with something new, or slightly original. Imitation is definitely not creation.

The fact is, we can't say yet what creativity is, and presumably never will be able to grasp it. In case we will in the future, then we'll see what happens, but I wouldn't hold my breath.


Lieven Malaise
Lingua 5B
Tom in London
Maria Teresa Borges de Almeida
Becca Resnik
Philip Lees
Kay Denney
 
Pages in topic:   < [1 2 3 4 5] >


To report site rules violations or get help, contact a site moderator:

このフォーラムのモデレーター
Jared Tabor[Call to this topic]

You can also contact site staff by submitting a support request »

Poll: How do you feel about AI and LLM?






Wordfast Pro
Translation Memory Software for Any Platform

Exclusive discount for ProZ.com users! Save over 13% when purchasing Wordfast Pro through ProZ.com. Wordfast is the world's #1 provider of platform-independent Translation Memory software. Consistently ranked the most user-friendly and highest value

Buy now! »
Trados Business Manager Lite
Create customer quotes and invoices from within Trados Studio

Trados Business Manager Lite helps to simplify and speed up some of the daily tasks, such as invoicing and reporting, associated with running your freelance translation business.

More info »