"The shift towards an MT-first-and-only scenario in translation profoundly impacts workflows, quality control, and post-editors' roles. With the metric "Time to Edit" (TTE) decreasing, post-editors' roles are transitioning from faster translations to refining quality-critical content. As MT quality improves, there's an increase in content translated without budget increments, sparking higher demand for translations in more languages. Customer data, alongside AI-based technologies, helps continually refine MT systems, turning MT from a competitor into a collaborator and enhancing, not replacing, human expertise."

Marina Pantcheva, RWS

“The recent rise of LLMs and their emerging capabilities has sparked the temptation to utilize them for translations. Research has demonstrated, though, that LLMs perform on par with NMT when it comes to segment-level translation. Their true contribution lies in tackling linguistically complex tasks that have traditionally been reserved for human translators".

Richard Sikes, memoQ

“Although LLMs have the potential to change the language industry in remarkable ways, we at memoQ remain positive about the future of purpose-designed CAT tools in our chosen target markets".

"The question of what impact GenAI will have on what and how we translate in the future can best be answered by looking at what we don’t and do know; we don’t know how much can we trust the output, or if it will take our jobs away, or what the risk are – but what we do know is we want to adopt it as fast as we can!"

Sung Cho, Amazon

“With the AI revolution, the business model for the localization industry is shifting. I would like to share the buyer’s perspective on what the AI revolution means for future industry requirements".

Michael Reeves, Google

“I will discuss how Google pivoted to a MT first approach where over 92% of our content is translated by an MT workflow with over 50% of our overall volume flowing through MT with no human editing".

"Uber's Localization journey: Embracing touch-less translation through MT and Human-AI collaboration."

"The Future of Localization: Machine Translation vs. GAI - Competition or Collaboration?"

"Exploring the combination of existing and emerging LLM and machine translation solutions to create the optimal experience for the global customer."

"The future of machine translation depends on our ability to accurately evaluate its output."

Giovanna Conte, Centific

“Discover how collaboration between LLMs and human translators enables the best of both worlds for quality evaluation".

"Utilizing LLMs for automating the quality estimation of Machine Translation".

"A vital factor for producing quality AI-driven MT concerns the LLM training data’s characteristics. To enhance the utility of resources for model training, we apply a multi-layer cross-lingual approach, incorporating typical linguistic patterns in meticulously constructed monolingual datasets, which can extend multilingually."

"As the use of generative AI tools for text generation proliferates, enterprise localization programs will need to develop observability tools and processes to ensure that they're not just buying ChatGPT output."

"In the wake of GenAI's transformative impact on translation, the ultimate test for technology companies lies in our ability to design intuitive, flexible tools - it will be the deciding factor for how linguists adapt, succeed and survive in this next era.”

"Showcasing how organizations can now make sure their localized content is inclusive and appropriate through AI-powered quality assurance checks and thus safeguard their brand image in their global markets."

"Showing that GenAI can become a useful tool to simplify the current Linguistic Quality Assessment workflow."

"Mitigating known machine translation issues around lack of context and repeated errors with LLMs by using document-level translation and Reinforcement Learning-based human-in-the-loop input."