Globalizing perspectives on AI safety

Artificial intelligence (AI) safety is an emerging field dedicated to ensuring that AI systems operate reliably, ethically, and beneficially. However, mainstream narratives surrounding AI safety often reflect the objectives and perspectives of Western institutions, prioritizing technical risks like alignment and misuse while neglecting broader societal and contextual harms. As global AI safety efforts gain momentum—through initiatives like the AI Safety Summit, specialized institutes, and new benchmarks—critical gaps remain. Many current evaluation methods lack coverage of non-Western languages, cultural norms, and societal contexts, leading to AI systems that perform poorly in diverse environments and reinforce systemic biases. These limitations highlight the urgent need for globalized approaches to AI safety.

On February 19, the Center for Technology Innovation at Brookings hosted a webinar featuring a panel of experts from across the globe to examine how Western-centric assumptions in AI safety frameworks can perpetuate inequities and bias. Panelists explored region-specific challenges, such as linguistic and cultural barriers as well as innovative frameworks, technical measures, and human-centered approaches to redefine what it means for AI to be “safe” on a global scale.

Read more from the series “AI Safety and the Global Majority.“

Viewers submitted questions for speakers by emailing [email protected] and via Twitter at @BrookingsGov by using #AISafety.