As every year, Slator, a leading provider of news, research, consulting and market intelligence for the translation, localization, interpreting and language technology industry, has conducted its ranking of language service providers (LSPs) around the world. The ranking comes with extra information related to each company, such as 2021 revenues, year-over-year growth, corporate headquarters, ownership and year-end. As the ranking indicates, in 2022 Pangeanic was in 98th position, 27 places higher than the previous year (125th in 2021).
In an increasingly globalized world, the need for high-quality language services is constantly increasing. From translating important documents to interpreting at international events, companies and organizations are looking for reliable and efficient language service providers to meet their needs. In this regard, language service providers are experiencing an increase in demand and a boom in the global market.
Communication is important in our society, and doing so in the relevant language is an added value that consumers appreciate. Language services play a fundamental role and will be expanded in the future. As Manuel Herranz (CEO of Pangeanic) says, language services (language learning, publishing, creative translation) will continue to be essential because they are related to communication channels and person-to-person or company-to-person communication. Technology is affecting the way we communicate with machines, as well as how we humans communicate with each other.
Artificial Intelligence has undergone major changes in recent years. With the arrival of ChatGPT, different sectors are being threatened and have not been able to see it as an opportunity to combine human reasoning with the machines' contributions.
The translation industry, and especially Machine Translation, shares many similarities with ChatGPT. Artificial Intelligence is:
In it, we can find its most complex branch: Natural Language Processing (NLP). Manuel says he sides with those who think that solving the communication challenge solves the most complex challenge of AI. Therefore, when we talk about NLP, we are talking about AI, these two branches cannot be separated.
ChatGPT is still generative technology and in the case of Machine Translation, the models have the pattern relationship between one language and another down to a T; therefore, they are more specialized in what happens between one language and another. They have more data, more adjustments and better techniques, and at some point, Machine Translation will achieve human parity, although, "in some aspects, it has been getting closer for some time now," says Manuel Herranz.
ChatGPT technology is here to stay, and will continue to evolve and improve over time; but are language service providers at risk? Can ChatGPT ever replace Machine Translation?
ChatGPT is based on the same technology as all modern Machine Translation: transformers. The name GPT stands for Generative Pre-trained Transformers, so, as Manuel says, it is the same technology that Google released about 5 years ago, but now applied to monolingual text to predict the next best word.
The goal of OpenAI is not to translate, and early tests point to the quality being too general compared to generic or specialized engines. Translating requires other linguistic skills, understanding the context and culture of other languages, etc. That is why the human figure who brings that creative touch to the machine's output is so important: "Only translators who translate like machines will be replaced by a machine. If the human brings value, creativity and knowledge, they are irreplaceable."
Pangeanic began its journey in the world of translation over 20 years ago. With the changes and evolution of the environment and technology, we have had to adapt to the times by innovating and developing disruptive solutions to differentiate ourselves from the competition. We work with three-year plans that are reviewed year by year to incorporate changes and address opportunities on the go. As our CEO says: "We are an SME and that gives us more agility than large companies to identify opportunities and implement changes. Adapting and improving processes is part of our DNA. Our mission is always clear and permeates the entire Pangeanic team, our raison d'être: to combine the scalability of computational processes (Artificial Intelligence) with human ingenuity."
We know that people have to be at the center of what we do, and for this we develop technology that validates this message, such as anonymization processes to respect the privacy of personal data in documents, databases or data collected by machines. We work on creating data that feeds Big Tech systems from an ethical point of view. All this combination promotes better translation services that can be human (creative) or machine (automatic), but always with the user, the person, in mind.