Discover how AI and LLMs are revolutionizing business innovation with insights from Welocalize’s conversations with over 40 global business leaders.

Artificial intelligence (AI) and large language models (LLMs) are changing the world as we know it. From content creation to supply chain management, AI and LLMS are reshaping the speed of innovation and emerging as monumental catalysts for change, redefining the boundaries of what is conceivable. What was once deemed impossible is now not only possible but expected.

Welocalize conducted conversations with leaders from over 40 global companies across diverse sectors to explore how AI and LLMs are redefining innovation.

 

Revolutionizing Industries

The innovation that AI and LLMs, such as OpenAI’s ChatGPT and Google’s ‎Gemini, bring to the table is reshaping multiple industries. Giants like Meta, IBM, Microsoft, Baidu, and Huawei are at the forefront, developing and refining a plethora of LLMs, heralding widespread change and rapid adoption of generative content within the coming months.

Adoption and Challenges

Darin Goble, Head of Solutions at Welocalize, anticipates significant changes as businesses, from software to life sciences, are feeling the pressure to implement LLMs. “We expect wide-scale change, with rapid adoption of generative content within 12 to 18 months,” said Goble.

Despite being in the early stages, companies are facing a dilemma between launching LLM output and ensuring compliance with legal standards. An executive from the software industry shared the same predicament, as their leadership is pushing to go live with LLM output despite the legal team’s recommendation to wait.

A Shifting Tide: The Disruption of NMT

Neural Machine Translation (NMT) has become a cornerstone in localization projects for language service providers (LSPs). This operating model has been repeatedly proven successful when combined with human post-editing.

However, the rise of LLMs poses a potential threat to dethrone NMT. Welocalize’s experiments indicate a shifting tide, with LLMs gradually outperforming NMT in localization due to the relentless pace of innovation.

During a discussion, Goble clarified, “NMT still currently outperforms LLMs for localization, but we expect this to change due to the pace of innovation.”

It’s only a matter of time before this shifts. A localization manager from a multinational internet brand and services provider shared that they produced positive results for a proof of concept with AI as a machine translation (MT) tool.

Co-Pilot Authoring: A Paradigmatic Shift

Major software companies such as Microsoft and Google have already integrated GenAI co-pilots, paving the way for easier and faster content generation and translation in multiple target languages.

“This is the paradigmatic shift of this new wave of technology, and we see it as an eventual inevitability for some language pairs and content types,” shared Goble. Companies should expect larger volumes of content as LLMs improve the speed and efficiency of content generation and translation.

Updated Workflows and Transforming Roles

Adapting to this technological wave means rethinking workflows and evolving job functions. Welocalize is pioneering new workflows, moving from “source to target” to “prompt to target,” addressing concerns like quality evaluation, human involvement in translations, and maintaining a brand’s tone and voice in translated content.

The customer uses prompts to generate multilingual content and AI quality estimates to assess its effectiveness; if it fails, the LLM is adjusted, and prompts are modified.

However, there were several questions and concerns, such as:

• The cost difference between using MT with post-editing and generative AI with post-editing
• The estimated time saved by adopting the new workflow
• Combining and integrating all elements from a tech stack perspective
• How generated content can pass quality evaluation without human evaluation
• Whether humans are no longer needed for translations
• How to customize prompts for each target language
• Who will be responsible for ensuring the brand’s tone and voice remain consistent in translated content
• The timeline for transitioning this new workflow from pilot to full-scale production.

“It’s not only workflows, but job functions will need to evolve rapidly,” Goble noted. “Localization leaders will need to move upstream. And localization management will evolve into global content management.”

 

Risks and Countermeasures

While LLMs offer numerous opportunities, they come with their set of challenges, including accuracy, data protection, and regulation.

Establishing trust in AI-generated content and addressing security concerns are paramount.

The output generated by GenAI can be unpredictable and problematic. Biases, falsehoods, and hallucinations are some of the issues that can arise from flawed training data. Therefore, ensuring that the training data used in developing AI models is reliable and unbiased is crucial.

A translator working for a multinational pharmaceutical and biotechnology company raised concerns about the possibility of AI generating false statements. Similarly, a quality management leader in the same company expressed concerns about the use of LLMs beyond their intended purposes. In addition, a senior localization manager from an international fashion enterprise questioned the reliability of AI’s quality estimation.

Security is a top priority that encompasses safeguarding data privacy and ensuring protection. “Security concerns will be overcome by major tech players, with regulations likely to differ by region. We are working closely with Microsoft and other suppliers. We are keen to work with our clients to map out new workflows pertaining to GDPR (General Data Protection Regulation) and other infosec considerations in their industries,” Goble assured.

Here are some practical ways to ensure enterprise security and data privacy:

• Use AI/LLM services hosted by major cloud providers incorporating advanced enterprise security and data protection measures for their AI offerings.
• Ensure these providers use data encryption and isolate the foundational LLMs from client data.
• Educate employees on the proper and approved use of AI. Warn against non-approved platforms.
• Form an internal AI task force to actively evaluate the new technical, legal, and business impacts of LLMs.

Integrating AI and LLMs into Localization and Content Services

Welocalize’s mission is not just to leverage the transformative capabilities of GenAI but to be the vanguard in integrating it into localization workflows, content supply chains, and the broader, ever-evolving business terrain. This is not just about keeping pace with change but defining it.

To delve deeper into GenAI’s transformative role in the evolution of content and translation and harnessing machine learning to redefine workflows, watch our Emerging AI webinar. Let’s explore the limitless possibilities and chart the course toward a future where technology and innovation converge to create a reshaped and redefined world.