ICYMI: Best of XTM Live: Large Language Models – A New Era for Localization

AuthorAleix Gwilliam

“Look at how [AI] was five years ago and how it is now. Take the difference and propagate it forward. That’s scary”

This is a quote by Geoffrey Hinton, a computer scientist who is known as the ‘Godfather of AI’. In terms of accuracy, it’s hard to argue with it. Recent AI advancements in the past 12-24 months have revolutionized the world, not least the arrival of GPT in the localization industry, an industry in constant search for increased speed, quality, and efficiency. For this reason, AI advancements are a hot topic in the world of localization, with their hype matched in proportion by the uncertainty of how to deploy them successfully.

Andrzej Zydroń is one of the leading and most trusted voices in the industry when it comes to new technologies, having architected the first cloud-based TMS among many other achievements. Back in April, he presented the talk Large Language Models: A New Era for Localization at XTM Live, in London. The presentation received rave reviews from attendees for its concise and insightful findings on AI, providing clarity in a field where there seem to be more questions than answers. Given the excellent feedback and the growing importance of the topic, we asked Andy to present his talk once again as a webinar for those who could not be at XTM Live. And in case you missed it (ICYMI), we’ve covered the main takeaways in here for you.

Big change is coming, and the pace of change is dramatic. We are still a long way away from AI technology plateauing.”

Andrzej Zydroń

Chief Innovation Officer at XTM

In the webinar, you will find simple answers to questions like:

What are large language models and how do they relate to AI?
What exactly is GPT?
What other models exist now and how big are they all?
What does GPT mean for the future of machine translation and neural machine translation (NMT)?
How does the deployment of AI technologies affect the role of humans in localization?

There is no doubt that GPT has marked a turning point in the localization industry. An uncertain economic landscape has resulted in companies focusing heavily on cost efficiency — yet without wanting to give up their current levels of output volume and quality. A few years ago, the idea of increasing the volume of localized content while simultaneously enhancing its quality and reducing time-to-market would’ve been unthinkable. However, with the arrival of large language models, this seemingly impossible goal has become a reality.

Nevertheless, there is a big difference between theory and practice, and deploying AI is not as simple as pressing a button and starting to reap the benefits. “What AI features do I need? How do I deploy them? How can I measure the ROI that will justify investing in them?” All of these questions are sprouting in the business world on all company levels, and answers are not always as straightforward and easy to come by as one would want. For this reason, learning about AI, GPT, and LLMs and what they can (and cannot) do, now and in the future, is key to being able to successfully deploy them and maximize their potential and benefits.

If you’d like to find out more about AI, LLMs and how they are shaping the present and localization, watch the recording of Andy’s webinar and gather all the information you need to understand how these technologies can drive growth in your localization program.

Watch the webinar