Key takeaways from the Language access & technology virtual session: Reflections on multilingual digital services in the federal government

KulturaTech hosted its first virtual session on Language access and technology: reflections from the federal government in late February. The conversation brought together researchers, practitioners, and former federal leaders working at the intersection of digital services and multilingual communities.

At a time when policies, technologies, and institutional priorities are shifting, one thing remains clear: language access is not just a compliance requirement, it is fundamental to equity, usability, and trust in public services.

Below are some of the key themes and reflections that emerged from the discussion.

1. What’s changed and what hasn’t

Executive Order 14224 does not require agencies to stop providing services or content in languages other than English, nor does it change existing obligations under Title VI of the Civil Rights Act. While some confusion followed a later Department of Justice memo, leading some agencies to remove translated content, these requirements do not apply to state and local governments. 

2. Language access is about people, not just translation

One of the most powerful reminders from the session was that language is not simply about converting words. It’s about identity, culture, and dignity.

“Oftentimes we talk about language as translation, the transformation of words, but to many people, especially those who come from the global South, members of the global majority, language is not just translation. It's an identity. It's a culture. It's a recognition of their humanity. And so when we advocate for language access alongside these communities, we're not just advocating for language, but we're advocating for people.” (Laura Gonzales, Panelist from Code for America)

For many multilingual communities, access to information in their language is directly tied to their ability to fully participate in society. When we design for language access, we are not just translating content, we are recognizing people’s humanity and ensuring they can access critical services with confidence and clarity.

3. Multilingual communities are not monolithic

Too often, institutions treat limited English proficiency (LEP) populations as a single group. In reality, these communities are incredibly diverse.

Factors such as:

  • Education level

  • Digital literacy

  • Age

  • Immigration experience

  • Familiarity with government systems

all shape how people interact with digital services.

Designing effective multilingual experiences requires moving beyond one-size-fits-all approaches and understanding the full spectrum of user needs.

4. Communities are already filling the gaps

During the session, speakers recognized that we often talk about what we are doing to serve multilingual communities, but it’s equally important to recognize the significant, often overlooked work these communities are already doing to support themselves and each other.

From informal translation networks to community-led guidance on accessing benefits, individuals are stepping in where systems fall short. 

5. The problem isn’t “hard-to-reach” users, it’s hard-to-reach systems

A key reframing emerged around research and engagement:

Multilingual communities are often labeled as “hard to reach.” In reality, it is our systems, processes, and assumptions that create barriers.

The challenge wasn’t a lack of willingness from participants. It was that recruitment systems and processes were designed around digitally fluent, English-dominant users. Common approaches like online panels and digital sign-ups unintentionally excluded multilingual individuals with limited English or access to technology. Even well-intentioned efforts, such as visiting community-based organizations with QR codes or asking people to share email addresses, created barriers for those without smartphones, email access, or familiarity with these tools.

The key takeaway is that organizations must actively remove these barriers and adapt their methods. When they do, communities are not only willing to share their experiences, but also feel seen and valued in the process.

6. Translation alone is not enough

One of the most important lessons from federal digital services work is that translation is not the strategy.

Effective language access requires:

  • Transcreation and adaptation, not word-for-word translation

  • User research that takes account the language and culture, not assumptions based on English users

  • Content strategies driven by community needs, not mirrored from English

In some cases, teams found success by starting with a non-English-first approach, designing content based on the needs of Spanish-speaking users and then expanding outward. This approach often improved the experience for everyone.

7. Preserve what works: Federal lessons still matter

Even as some federal efforts evolve or pause, there is a wealth of knowledge worth preserving and building upon.

Key assets mentioned include:

These resources are not just valuable for the government. They are highly relevant for nonprofits, academia, and private-sector organizations.

8. The future: seamless, inclusive, and collaborative

Looking ahead, participants shared a vision for what language access could become:

  • Seamless user journeys across federal, state, and local systems (no more language “drop-offs”)

  • Greater collaboration across states to avoid reinventing the wheel

  • Integration of language access with accessibility including audio, video, and non-text formats

  • Responsible use of AI, with human review and quality assurance at the center

Importantly, there was strong consensus that this work will continue, whether driven federally or at the state and local levels, because the need is undeniable.

Final reflection

As digital services continue to evolve, the question is not whether organizations should invest in multilingual experiences but how thoughtfully, inclusively, and sustainably they will do so.

Next
Next

Start with strategy, not translation