In traditional MTPE, the translator fixes a machine’s mistakes. In this workflow, they are essentially “revising” a draft that was generated using their own gold standard anchor text and terminology. And they are given a role in improving the machine’s output during the project as well.
This shift isn’t just semantic; it aligns better with industry-standard pricing for revision (typically ~30-35% of the full rate) rather than the arbitrary “per-word” discounts of commodity MTPE. It respects the translator’s role as the final arbiter of quality.
GAIT-Augmented MTPE isn’t about giving translators less for more. It’s about giving them better inputs so they can work more efficiently and increase their work volumes… even if the per-word rate is lower. And then passing on those savings to clients so that they will uncover more content for high-quality translation.
When they see the quality of the drafts and the vision behind the workflow, many translators recognize it as a fair deal. You just have to present it that way: honestly, and collaboratively.