The 2021 ESG Summit will be held in Ireland on 31 May, while the World Environment Day is set on 5 June. Welcoming the two, we’ll be looking at the ESG reports of listed companies in Hong Kong and the challenges and future direction of translation of these reports in three consecutive feature articles.

In the last two weeks, we’ve talked about what Environmental, Social and Governance (ESG) reports are and had an idea of the challenge of translating ESG reports, which is the limited timeframe against a massive amount of copywriting-translation content. In the last chapter of this ESG trilogy, we’re going to look forward—how can AI translation engine help break the bottleneck?

Among the translation of an ESG report, the first draft is the most time-consuming step. Meanwhile, manual vetting (or some translators like to call it reviewing, while laymen often refer to it as proofreading) is much quicker than manual translation word-by-word. Therefore, the most efficient way would be to let artificial intelligence complete the tedious first draft, followed by manual vetting and polishing. As time is saved from word-by-word translation and instead spent on brushing up and copywriting, the entire workflow will be more productive with better outcome in both speed and quality. In particular, as we mentioned last week, quite some listed companies from Mainland China draft their ESG reports in Chinese and ask for English translation. Practicing translators are, however, generally less competent in Chinese-to-English translation, which restricts manpower allocation and makes it harder for translation firms to take up C-E translation tasks of ESG reports under such constraint. I’ve just done a quick test by importing a Chinese ESG report of 84 pages into DeepTranslate’s AI platform and it only took 6 MINUTES to generate an English version! Even with manual vetting, it should take no more than a few days; in contrast, a word-by-word, sentence-by-sentence manual translation in the traditional way can probably take two to three weeks of manhours. A number of clients have given feedback that it’s much more time effective to translate ESG reports with AI, even when labour vetting is taken into account.

Without a doubt, despite a slight shift in the role of translators, their instinct for language aesthetics, insight about tone and manner, choice of words for the appropriate context and understanding of clients’ preference are still irreplaceable. Nonetheless, the advancement in technology can hopefully be a painkiller that share some of their burden of translating ESG reports during peak annual reporting seasons.