At the GALA 2024 Valencia session, the Special Interest Group (SIG) on MTPE Tr-AI-ning explored the impact of AI translation on various key stakeholders, including academics, buyers, language service providers (LSPs), and post-editors. The overall goal of the session was to explore both the challenges and opportunities presented by AI, with a focus on how these groups can adapt and evolve in the rapidly changing translation landscape. This chapter outlines the key areas of concern and opportunity for each stakeholder group as discussed during the presentation.
The advent of AI in translation has dramatically changed the landscape of translation studies. Traditional methodologies are now merging with cutting-edge AI technologies, opening up new avenues for research. Academia faces both opportunities and challenges in integrating AI translation into its programs. There is great potential for collaboration between human language skills and artificial intelligence, a concept often referred to as "co-creation". However, the rapid pace of technological advancement has created significant training gaps, making it difficult for academic institutions to keep up and effectively train students for the future of translation.
Quality and ethical considerations are of paramount importance to academia. With the unpredictable and dynamic behavior of AI models such as Large Language Models (LLMs), ensuring the accuracy and ethical use of AI translation tools remains an important area of research. As academia continues to explore the output quality of AI models, the focus remains on how to ensure that future linguists are prepared to meet these evolving challenges.
For buyers of translation services, AI translation offers numerous opportunities for cost-effective solutions. Integrating AI into translation workflows allows companies to streamline operations and increase efficiency, making it an attractive option. However, it comes with its own set of challenges. Buyers are faced with the need to maintain high levels of quality assurance while ensuring the ethical use of AI technology. Concerns about the accuracy, reliability, and cultural appropriateness of AI-generated translations remain key issues for buyers to address.
In addition, training and expectations management are important for buyers. As AI translation becomes more prevalent, buyers need to understand both its limitations and its strengths. Proper training for those working with AI translation tools is essential to ensure that workflows are optimized and that both human- and AI-generated content meet required standards.
LSPs are at the forefront of the AI translation revolution. AI offers a transformative shift in service offerings, moving from traditional translation services to more specialized post-editing tasks. In this new environment, LSPs must adapt to new collaborative working models that combine human expertise with AI-driven workflows. The synergy between AI tools and human translators enables greater productivity, but also requires ethical considerations regarding bias, cultural sensitivities, and data security.
AI translation also reshapes the role of post-editors. As AI systems increasingly handle repetitive translation tasks, post-editors are expected to focus on refining the more complex aspects of translation that AI systems may struggle with—such as idiomatic expressions, cultural nuances, and domain-specific terminology. Post-editors play a crucial role in ensuring that AI-generated translations are both accurate and appropriate for their intended audience.
Training and skill development for both LSPs and post-editors are critical in this evolving landscape. LSPs must invest in upskilling their workforce to effectively leverage AI tools, while post-editors need to adapt to new responsibilities that require higher-level decision-making and problem-solving skills. Additionally, ensuring ethical standards are upheld in the post-editing process is crucial to maintaining the integrity and quality of translated content.