Companies are embracing generative artificial intelligence to produce marketing and corporate content at scale, even as most say they do not trust the technology to understand the cultural and emotional nuances needed to connect with audiences across global markets.
According to a survey commissioned by RWS, a provider of language, content and localization technology, 94% of senior content executives have little or no confidence in AI’s ability to handle cultural nuance, yet many are deploying it across international markets regardless.
The findings, based on a survey of 200 senior content leaders in the United States, Britain and Asia-Pacific, underscore a growing challenge for multinational companies: while AI is accelerating content production, it is also creating new costs and operational bottlenecks when content must be adapted for local audiences.
Speed gains, but localization lags
Some 86% of respondents said generative AI had accelerated content creation, helping organizations produce material faster and in greater volume, the survey found.
But those gains have not extended evenly across the content supply chain. About 65% said AI had slowed localization — the process of adapting content for different languages, markets and cultural contexts — as human teams spend more time reviewing and refining machine-generated output.
Fewer than one in 10 respondents said they were confident AI could handle cultural nuance effectively, according to RWS.
Rework adds to enterprise costs
RWS said the need to revise AI-generated content is consuming roughly 21% of enterprise localization budgets annually, creating what it described as a growing financial and operational burden.
For illustration, the company said an enterprise spending $5 million a year on localization could incur more than $1 million in rework costs.
The gap is particularly evident between translation and localization. While 71% of respondents said they use generative AI for translation, only 20% use it for localization, which requires a deeper understanding of local idioms, preferences and cultural context.
Teams under pressure
More than half of those surveyed described their organizations as “managing but stretched,” while only around one in six said they were handling current content demands well.
Even so, more than half said they expected their organizations to cope better within three years without significant structural changes.
RWS cautioned that this may underestimate the scale of the challenge as AI-generated content expands across more languages, formats and channels.
Human oversight remains critical
The company said the solution is not to slow AI adoption, but to deploy systems that better account for context, culture and brand intent.
“As AI-generated content volumes continue to grow, the complexity tax will only compound,” RWS said in the report, arguing that enterprises should build cultural intelligence into their content operations earlier in the workflow.
The survey offers an indication of how large global enterprises are adapting to AI, though its findings are based on company-sponsored research rather than an independent industry-wide study.
