Locations
What is Every Language Matters?
Every Language Matters (ELM) is a global-to-local initiative that turns hard-won country lessons into coordinated public action toward AI that leaves no one behind, no matter the language they speak. It builds on two years of work through UNDP’s global network of ELM Accelerators operating in-country, and unites governments, researchers, funders, technology partners, and communities for sustainable local AI ecosystems and policy frameworks that promote local agency.
Are you working on making AI serve your language and culture?
AI systems are reliable in fewer than 100 of the world's 7,000+ languages. Every Language Matters helps change that, together.
To build AI that serves all communities and reflects the world’s wealth of cultures and languages, high-quality data is crucial – but it is not itself sufficient. Systematic change is needed to protect local agency and ensure AI diffusion is sustainable while consistently delivering public value. This larger ecosystem shift is what Every Language Matters sets out to orchestrate.
The initiative mobilises stakeholders across sectors to work together on five areas of joint action: (1) understanding barriers to inclusive, sustainable AI innovation, (2) mobilising leadership for systemic change, (3) translating ambition into implementation, (4) unlocking resources for scale, and (5) shaping the field’s best practices. Taken together, Every Language Matters systematizes the insights implementers are learning the hard way and realigns incentives to fix at systems level what individual programmes cannot fix alone.
What can Every Language Matter enable for you?
Understand how to unlock new value for local AI ecosystems through the Data-to-AI Value Chain
While many challenges may be shared, each community faces interlocking needs, sensitivities, and ambitions when it comes to developing and deploying AI. These cannot be understood in isolation. The Data-to-AI Value Chain helps reveal the technical, institutional, financial, and governance activities through which data is transformed into AI, helps reveal where progress may be stalling or where a unique opportunity lies, and identifies points of leverage where coordinated action can generate new public value.
Catalyse Government action to enable systemic change
Putting the Data-to-AI Value Chain insights into practice, ELM supports Governments, public decision makers, and institutions define and implement concrete commitments to support their local AI ecosystems – be it through releasing public data, strengthening local capacity, harmonising mandates or channelling investment.
Channel ambition into practical implementation with the ELM Accelerators
Where a concrete bottleneck has been identified, UNDP works with global experts and local partners to pilot a project or technical solution. Built in the open, with local communities while drawing on state-of-art expertise: each Accelerator is locally owned yet globally enabled.
Unlock new partnerships and resources for scale
Many projects integrating under-resourced languages are already funded – through development programs, philanthropic donors, research grants, start-up venture capital or smaller vendor contracts. ELM sets out to help connect the investments already at play, catalyse new partnerships that drive mutually beneficial synergies, and channel new resources toward promising innovations ready to scale.
Shape the field’s technical standards and best practices
The reliability of AI systems across diverse languages, cultures and contexts is a key consideration in global AI governance – and excellent grassroots work deserves the spotlight. ELM convenes a global pool of experts and innovators to define and promote shared norms that orchestrate progress toward responsible, community-centered practices in AI development, deployment, and evaluation.
Looking to dive deeper?
Examples from ongoing Every Language Matters Accelerators:
Supporting communities to exercise meaningful agency over their data and AI systems.
