| Pages in topic: [1 2 3] > | Poll: In AI-assisted translation, where do you see the biggest risk today? Thread poster: ProZ.com Staff
|
|---|
This forum topic is for the discussion of the poll question "In AI-assisted translation, where do you see the biggest risk today?".
View the poll results »
| | | |
All of the above + the highest risk is relying on AI alone...
| | | |
"Hallucination" should be added to the list.
| | | |
Inconsistency thrives in the (yawn...) 'AI Translation' world...
| | |
|
|
|
Daryo
Local time: 06:22
Serbiyaan to Ingiriisi
+ ...
| There would be no 'risks' of any kind | Jan 28 |
if it wasn't for HUMANS spreading nonsense hype and other HUMANS all too eager to swallow it hook, line and sinker.
AI 'translation' in itself is no more 'dangerous' than Google translate, or even some old paper dictionary, for that matter.
What can make any of them (most literally) dangerous is when 'users' start having blind faith it the tool.
In that regard even the most ordinary paper dictionary CAN be very 'dangerous' in the hands of some... See more if it wasn't for HUMANS spreading nonsense hype and other HUMANS all too eager to swallow it hook, line and sinker.
AI 'translation' in itself is no more 'dangerous' than Google translate, or even some old paper dictionary, for that matter.
What can make any of them (most literally) dangerous is when 'users' start having blind faith it the tool.
In that regard even the most ordinary paper dictionary CAN be very 'dangerous' in the hands of someone who obstinately maintains that the first meaning given in it for a word is the only one that word could possibly ever have.
Put a tool like web search in the hands of the same kind of people and then you get lectured that you got it wrong because the first answer to a clumsily formulated query is a different one from yours - never mind that 'the authoritative Google answer' has less chance of fitting in the ST than a square peg in a round hole.
As long as you are aware what some tool CAN do and more importantly what it CAN NOT do, there is no danger in using it.
Unfortunately, after being told few times in the past by non-translators that my (researched and triple checked) translation is rubbish because Google Translate gives some other one I have now the believe-it-or-not experience of translators quoting AI output as some kind no-way-it-could-ever-be-wrong ultimate reference'. ▲ Collapse
| | | | Luis M. Sosa 
Ecuador
Local time: 00:22
Ingiriisi to Spanish
+ ...
"As long as you are aware what some tool CAN do and more importantly what it CAN NOT do, there is no danger in using it". Strongly agree with you.
[Edited at 2026-01-28 17:12 GMT]
| | | | | Other: possible contamination with false logic | Jan 28 |
I have no idea how and if the AI answers are used in the global system, but I see that sometimes AI delivers common wrong views (from traditional values point of view). That means that the algorithms don't allow him to be creative or inquisitive, combining the traditional and the innovative. Thus, instead of building a colorful landscape, I am afraid he might build millions of identical Empire State high scrapers.
| | | | Luis M. Sosa
Ecuador
Local time: 00:22
Ingiriisi to Spanish
+ ...
| There is still a long way to go | Jan 28 |
for AI to justify the prominent status it has reached over the last couple of years. I have made several questions in the fields of economics and finance to some AIs, and have gotten -not a few- of utterly wrong answers (hallucinations in AI lingo). And one can conclude that the rate of failure is still high. Makes me wonder if that's just a matter of further training the model.
| | |
|
|
|
| Fluent output creates the illusion of accuracy | Jan 29 |
Fluent output creates the illusion of accuracy, making errors harder to spot and easier to accept without proper review.
| | | | Philip Lees
Greece
Local time: 08:22
Greek to Ingiriisi
Monopolisation and monetisation of services by a small number of big tech companies that set profits ahead of all those other issues.
| | | | Alex Lichanow
Germany
Local time: 07:22
Ingiriisi to German
+ ...
I mean, even the term "AI-assisted translation" is misleading.
At least as far as I understand the word, "to assist" means to be helpful. AI-enshittified "translation" is neither helpful nor actual translation.
| | | | Liena V.
Latvia
Local time: 08:22
Member (2014)
Faransiis to Latvian
+ ...
As I recently saw somebody saying in an online post unrelated to translation, having to use AI is like dealing with an incompetent and annoying assistant who only keeps messing things up (while spying on you, I would add), but you can't even fire it because it is the CEO's nephew.
[Edited at 2026-01-29 10:35 GMT]
| | |
|
|
|
Zea_Mays
Italy
Local time: 07:22
Ingiriisi to German
+ ...
Philip Lees wrote:
Monopolisation and monetisation of services by a small number of big tech companies that set profits ahead of all those other issues.
The good news is: There are no profits, and none are expected anytime soon. Time is on our side.
| | | | | Overreliance on AI without human (specialized) review | Jan 29 |
Overreliance on AI without human (specialized) review is, in my view, the biggest risk today.
AI output can sound fluent and convincing while still missing nuance, context, or specialized terminology.
When companies skip human review, they’re not just risking quality: they’re risking accuracy, compliance, and credibility.
AI is a productivity tool, not a decision-maker. Without qualified human translators in the loop, especially subject-matter specialists, errors become har... See more Overreliance on AI without human (specialized) review is, in my view, the biggest risk today.
AI output can sound fluent and convincing while still missing nuance, context, or specialized terminology.
When companies skip human review, they’re not just risking quality: they’re risking accuracy, compliance, and credibility.
AI is a productivity tool, not a decision-maker. Without qualified human translators in the loop, especially subject-matter specialists, errors become harder to detect and easier to spread.
That’s when problems come in: subtle mistranslations, legal or cultural missteps, brand damage… all invisible until it’s too late! ▲ Collapse
| | | | Daryo
Local time: 06:22
Serbiyaan to Ingiriisi
+ ...
| It's not a bug - it's an unavoidable feature | Jan 29 |
expressisverbis wrote:
Overreliance on AI without human (specialized) review is, in my view, the biggest risk today.
AI output can sound fluent and convincing while still missing nuance, context, or specialized terminology.
When companies skip human review, they’re not just risking quality: they’re risking accuracy, compliance, and credibility.
AI is a productivity tool, not a decision-maker. Without qualified human translators in the loop, especially subject-matter specialists, errors become harder to detect and easier to spread.
That’s when problems come in: subtle mistranslations, legal or cultural missteps, brand damage… all invisible until it’s too late!
Overreliance on AI without human (specialized) review is not a bug - it's an unavoidable feature.
For one very simple reason:
the moment you include a proper check by qualified / expert humans [be it an 'AI translation' or any other AI output] ALL the 'savings' in time and/or money vanish in thin air.
| | | | | Pages in topic: [1 2 3] > | To report site rules violations or get help, contact a site moderator: You can also contact site staff by submitting a support request » Poll: In AI-assisted translation, where do you see the biggest risk today? | Protemos translation business management system | Create your account in minutes, and start working! 3-month trial for agencies, and free for freelancers!
The system lets you keep client/vendor database, with contacts and rates, manage projects and assign jobs to vendors, issue invoices, track payments, store and manage project files, generate business reports on turnover profit per client/manager etc.
More info » |
| | Draftsmith | Edit Translations Faster With Affordable AI You Can Trust
Draftsmith is an affordable AI editing tool for busy translators. It works directly in Microsoft Word, supporting the way you work
ProZ members receive a 20% discount on a single user subscription of Draftsmith.
More info » |
|
| | | | X Sign in to your ProZ.com account... | | | | | |
Login to reply/comment 
