s AI rapidly reshapes the global economy and society, open–source AI has emerged as a potential catalyst for making our future infrastructure, economies and societies better off—from inclusive innovation to economic opportunity, increased privacy, and decentralized technology governance. Why? Among other things, open–source AI allows for interoperability, open weights, and smaller economies, start–ups, civil society, and academia to inspect, iterate, and build upon foundational models.
Ideally, open–source AI will enable individual users and communities across the globe to shape their digital and economic future. By offering increased transparency and adaptability, these models can be used to build locally driven innovation that reduces dependency on gatekeepers and authoritarian technologies.
For example, China’s DeepSeek took the AI world by storm in early 2025, but it also revealed the pitfalls of open–source models developed under censorship laws and surveillance regimes that require data sharing with the Chinese Communist Party. Open–source AI under democratic standards and international collaboration provides a means of inclusion and a counterbalance to state–controlled innovation.
Closed–source models invariably consolidate control and reduce competition in the marketplace. While this approach is often framed as a security measure against malicious actors (often backed by malign states), such groups will inevitably find access points, regardless of whether the broader AI ecosystem is open or closed. What centralized control does effectively limit, however, is the global community’s ability to leverage, audit and fix flaws in new models and localize the potential of the AI economy. After all, AI models are built upon our collective digital footprint.
Advanced open–source models are already widely accessible and OpenAI released two new open–source models last month alone. Attempts to restrict distribution through overregulation or export controls are simply impractical. We should instead focus on offering better alternatives—from models and safeguards to partnerships and governance approaches.
Here are some things we can do to bring about those better alternatives.
Support open access. Promote open-source AI initiatives that incentivize education and skilling, inclusivity, transparency, and ethical design.
Establish guardrails. Develop frameworks for safety evaluations, use restrictions, and responsible release of model weights.
Enable global participation. Partner with emerging economies, start–up clusters, academia, and civil society to support local AI infrastructure and development capacity.
Promote AI governance. Create balanced, flexible standards and sandboxes that encourage innovative new entrants and consumer trust while mitigating harms and dependency on authoritarian technology.
Open–source AI is already reshaping the digital landscape in ways big and small. Inclusion requires more AI diffusion with collaboration across democratic governments, industry, and the public to build trust and leverage this technology locally, maximally and securely.
a global affairs media network
The case for more open–source AI
Image via Adobe Stock.
September 23, 2025
As emerging technologies continue to reshape our economies and societies, open–source AI is emerging as one way to mitigate many potential ills associated with the future of AI, writes Louisa Tomar.
A
s AI rapidly reshapes the global economy and society, open–source AI has emerged as a potential catalyst for making our future infrastructure, economies and societies better off—from inclusive innovation to economic opportunity, increased privacy, and decentralized technology governance. Why? Among other things, open–source AI allows for interoperability, open weights, and smaller economies, start–ups, civil society, and academia to inspect, iterate, and build upon foundational models.
Ideally, open–source AI will enable individual users and communities across the globe to shape their digital and economic future. By offering increased transparency and adaptability, these models can be used to build locally driven innovation that reduces dependency on gatekeepers and authoritarian technologies.
For example, China’s DeepSeek took the AI world by storm in early 2025, but it also revealed the pitfalls of open–source models developed under censorship laws and surveillance regimes that require data sharing with the Chinese Communist Party. Open–source AI under democratic standards and international collaboration provides a means of inclusion and a counterbalance to state–controlled innovation.
Closed–source models invariably consolidate control and reduce competition in the marketplace. While this approach is often framed as a security measure against malicious actors (often backed by malign states), such groups will inevitably find access points, regardless of whether the broader AI ecosystem is open or closed. What centralized control does effectively limit, however, is the global community’s ability to leverage, audit and fix flaws in new models and localize the potential of the AI economy. After all, AI models are built upon our collective digital footprint.
Advanced open–source models are already widely accessible and OpenAI released two new open–source models last month alone. Attempts to restrict distribution through overregulation or export controls are simply impractical. We should instead focus on offering better alternatives—from models and safeguards to partnerships and governance approaches.
Here are some things we can do to bring about those better alternatives.
Support open access. Promote open-source AI initiatives that incentivize education and skilling, inclusivity, transparency, and ethical design.
Establish guardrails. Develop frameworks for safety evaluations, use restrictions, and responsible release of model weights.
Enable global participation. Partner with emerging economies, start–up clusters, academia, and civil society to support local AI infrastructure and development capacity.
Promote AI governance. Create balanced, flexible standards and sandboxes that encourage innovative new entrants and consumer trust while mitigating harms and dependency on authoritarian technology.
Open–source AI is already reshaping the digital landscape in ways big and small. Inclusion requires more AI diffusion with collaboration across democratic governments, industry, and the public to build trust and leverage this technology locally, maximally and securely.
