Will ChatGPT Eliminate Translation Jobs?

by David Grunwald

Jobs across numerous fields could be replaced, according to the company ChatGPT. The tech company, which has sparked debate in recent months, said it could “potentially replace jobs that involve written communication,” such as translation services and social media managers.
(Yahoo Finance)

In a recent interview, IBM CEO Arvind Krishna stated that A.I. is going to replace many clerical white collar jobs. Krishna stated that A.I. will eliminate 7,800 jobs at IBM over the next five years.

Should professional translators be worried and start to look for another line of work? Nobody knows the answer for sure but here are some of my own thoughts on the future of the industry and job market.

History is on the side of translators

The debate on whether machine translation will replace human translators has been going on for decades. And despite the technological advances and improvements in MT over the past years, the demand for translators has grown. According to the Department of Labor, the number of people employed in the translation and interpretation industry has doubled in the past 7 years (courtesy of the University of Arizona website). According to the US Bureau of Labor Statistics, the job outlook for translators and interpreters is very strong. (Disclaimer: this chart was updated in September 2022, before the launch of ChatGPT).

It is difficult to see at this stage how ChatGPT will be a bigger disrupter than Google Translate, since it essentially does the same thing (machine translation).

ChatGPT’s Position

I asked ChatGPT-4 this question: Will AI replace professional translators? This was the response.

Advancements in artificial intelligence and natural language processing have led to the development of increasingly sophisticated language models, like GPT-4. As a result, it is possible that many jobs involving written communication could be affected, including translation services. Some ways in which AI might impact these industries include:

Automated translation: AI-powered translation tools, like Google Translate, have already made significant progress in providing fast and accurate translations for many languages. As AI continues to improve, these services are likely to become even more accurate and reliable, potentially reducing the demand for human translators.

Content creation: AI can generate written content in multiple languages, saving time and resources for businesses that need to produce multilingual content. This capability could reduce the need for human content creators and translators, particularly for more repetitive or formulaic content.

Proofreading and editing: AI can be used to automatically detect and correct grammar, spelling, and punctuation errors, as well as provide suggestions for improving style and clarity. This might impact the demand for human proofreaders and editors, especially for more routine tasks.

Real-time translation services: AI-powered devices and applications can provide real-time translations for conversations and meetings, potentially replacing human interpreters in some situations.

Language tutoring: AI-powered language learning apps can help users learn new languages, practice their skills, and receive personalized feedback. This might reduce the demand for human language tutors, especially for more basic language learning.

However, it is important to note that AI will not completely replace human professionals in these fields. Human translators and content creators bring unique skills, such as cultural understanding and creativity, that are difficult for AI to replicate. Additionally, AI-generated translations may still require human oversight and editing to ensure accuracy and appropriateness, particularly in more complex or specialized contexts.

The key to minimizing job displacement will be to adapt and find ways to work alongside AI. Professionals in the translation and content creation industries should focus on developing their expertise in areas where AI struggles, such as nuanced language, idiomatic expressions, and culturally specific content.

The Status Quo and things to come

After reading the ChatGPT response, these are the thoughts that come to my mind:

1. Comparisons to Google Translate. ChatGPT says that as AI continues to improve, it has the potential to eliminate some translator jobs. But as I already stated, MT has not eliminated any jobs in the last 20 years. In fact, it has added many MT post-editing jobs.
2. Content creation in multiple languages. What this means is that companies will be able to generate original content in other languages, eliminating the need for translation of marketing material, website content, product descriptions. This actually makes sense and could eliminate some translation jobs. But even if companies use AI to generate content in other languages, language experts will still be needed to review the content.
3. Proofreading and editing. This is an extension of the previous point. A person can feed MT or a rough draft into ChatGPT and get a rewrite which is linguistically accurate. How will ChatGPT be able to compare the source and the target? I am not sure but it does seem possible.
4. Real-time translation. This is another discussion which has been going on for years. Will technology render the job of interpreters obsolete? Advances in speech-to-speech translation can theoretically make this possible. And all of the big techs like Microsoft, Google, Facebook and Amazon have been working on this for years. It is hard to see this happening in the next few years. And again, history is on the side of skepticism.

The best course of action for professional translators

ChapGPT suggests that professional translators work alongside AI. This is good advice. Translators should keep abreast of the advances in AI and harness its strengths to make their job more efficient. Just like every translator uses MT to a point (and any translator who says they never use it are lying), translators should research and use AI tools to similar benefit. Translation work has some repetitive tasks which could potentially be eliminated by AI. Many of these tasks are already covered by word processing software and CAT tools like applying TM matches, using term bases, spell and grammar checking. AI may bring more efficiency to the process and enable translators to focus on the areas in which they bring the most added value.