In March 2025, a U.S. Court of Appeals ruled (PDF) that purely AI-generated works cannot receive copyright protection.

AI translation is — of course — an example of AI-generated works, and with intellectual property (IP) in translation long being a hot potato in the US and beyond, translators are increasingly crying foul over the use of AI trained on their work to create content.

Before language AI, disputes tended to revolve around disagreements on royalties between translators and publishing houses. AI adds another layer to the squabble when their works are used to train large language models (LLMs) without any legal protection.

This point was illustrated in a letter (PDF) sent by several organizations representing translators, writers, and journalists in March 2025 to the European Commission (EC), in which they oppose the copyright enforcement measure laid out in the third draft of the EU Code of Practice for the implementation of the AI Act because it fails to protect their IP. 

“As to the requirements of transparency and copyright enforcement measures laid down in the AI Act and the related Directives on copyright, the third draft is toxic in its entirety,” reads the letter.

We asked readers if they think AI-generated translations should qualify for copyright protection, and the largest cohort (41.1%) believe they should, but only if there’s a human-in-the-loop. Over a third (35.7%) disagree. For the rest, it depends (12.5%), and the smallest group actually agrees (of course, 10.7%).

Voice Cloning Consent

YouTube’s venture into automatic dubbing, first announced in September 2004 and subsequently rolled out to qualifying creators from December and on, marks a significant, albeit potentially fraught, step in democratizing global audiovisual content.

The “hundreds of thousands” of qualifying creators that would get access to automatic dubbing capabilities in their channels started seeing the dubbing service roll out as a feature turned on by default, though: unless the feature is disabled, dubs are automatically generated when a new video is uploaded and published — for “non-experimental languages,” explains a YouTube support article.

The nine “non-experimental” languages for automatic dubs are English, French, German, Hindi, Indonesian, Italian, Japanese, Portuguese, and Spanish. YouTube detects the original language and generates dubbed versions to and from English and the other supported languages. 

Creators now face the dual challenge of managing the inevitable imperfections of AI-translated audio and navigating the predicament of voice cloning permissions. YouTube’s default-on approach, while prioritizing ease of use, may well be treading a delicate line between access, accessibility, and the rights of individuals to control the use of their likeness and voice in AI-dubbed content.

Among Slator readers, obtaining explicit consent for voice cloning is a must (84.0%). The rest of the poll respondents are undecided (4.0%), consider it not practical (5.3%), or think this consent exists via the creators themselves (6.7%).

Hiring for AI-Related Roles is Getting Easier

The latest installment of the “State of AI” report released by McKinsey & Co. on March 12, 2025, offers a particularly intriguing glimpse into AI’s impact on the translation sector: an easing in the once reportedly challenging task of recruiting translators.

For the past three years, the consulting outfit’s analysis has grouped translators among AI specialists, a tacit acknowledgment of the profound shifts the technology has wrought upon the field. This year’s findings are based on a July 2024 survey of over 1400 respondents across many countries and industries who were asked about the hiring difficulty for twelve “AI-related roles.”

What was once a hurdle, with over 70% of respondents reporting recruitment difficulties in 2022, has softened to roughly 55% in the latest survey. This trend, echoed across several AI-adjacent roles, is examined in the report’s “state of AI-related hiring” section. 

Respondents were asked to describe the level of difficulty in hiring for roles by choosing among options like “very difficult,” “difficult,” “neither difficult nor easy,” or “easy.” 

McKinsey’s own take on the influence of upskilling in the downward trend aligns with other data indicating that a growing number of language professionals are acquiring AI skills, including Slator’s June 2024 survey of 260 translators and interpreters.

On the question of whether hiring translators had become easier, a slim majority (50.7%) of readers concur that it is indeed the case, yet a significant 26.7% believe it hinges on the translator’s specific skill set. This leaves a notable segment (22.6%) of those who see no such trend, reflecting perhaps the fluid nature of the AI language market as it was in Q3 2024 and continues to be.

Localization Roller Coaster

Despite economic shifts and relentless technological strides, the language services industry has demonstrated remarkable resilience and consistent growth over the past 20+ years. Sustaining this expansion is a level of globalization that took just as many years to solidify and that is now under threat by the announced, then suspended, then announced again new US tariff regime. 

Global by nature, local by design, a great number of language services providers (LSPs) are concentrated in the US. These are mostly small companies already feeling the effects of language AI and other Trump administration policy shifts, and these new tariffs have the potential to wreak havoc across every localization-buying vertical.

International trade is like the visible stitches in a well-made leather bag: they reassure the wearer that the bag will not fall apart and will hold its shape (and value!) for years to come. Can buyers and LSPs aspire to profitability as the thread is pulled? We asked readers their opinion on the tariffs’ impact on the language industry, and close to half (46.2%) think it will be immediate and severe.

To one in three respondents (33.8%), the effect is net negative, but it will take time. To the rest (12.3%), the impact will be immediate but mild.