Risks and drawbacks of the digital dictatorship The overarching risk of AI in Myanmar is the centralisation of arbitrary surveillance and the automation of repression. In a fragile State where the rule of law has collapsed, AI tools are not being used to enhance development, but to enforce a digital dictatorship. The military has integrated AIpowered technologies, specifically facial recognition and data processing systems, into its security apparatus to create a pervasive climate of fear. This transforms the public square (both physical and digital) from a space of cultural expression into a space of risk, effectively dismantling the right to participate in cultural life without fear of persecution. Algorithmic bias as structural exclusion Algorithmic bias in Myanmar acts as a structural barrier to the right to non-discrimination in the context of development. Major AI models trained predominantly on Western datasets remain culturally and linguistically blind to Myanmar's diverse reality. This bias manifests firstly as linguistic exclusion. Low-resource ethnic languages like Shan, Kachin, and Rakhine are underserved by AI tools for translation and moderation. This creates a discriminatory tier of development where digital benefits are accessible only to Burmese and English speakers, further marginalising minority cultural expression and undermining their right to use their own language. Secondly, a pervasive engagement bias drives polarisation. As most clearly documented during the Rohingya crisis, social media algorithms favour content inciting anger or fear over peace-building. This design bias systematically amplifies hatred while burying cultural rights advocacy, actively undermining the State's obligation to prohibit advocacy of national, racial, or religious hatred that constitutes incitement to discrimination, hostility, or violence. Algorithmic silence and automated censorship While State censorship is overt, AI-driven content moderation introduces a form of algorithmic silence that acts as an arbitrary interference with the right to freedom of expression. Social media platforms, which serve as the primary archive of Myanmar’s resistance history, rely on blunt AI tools to moderate content at scale. These algorithms often lack the cultural context to distinguish between violent content and documentation of war crimes. Consequently, valid human rights evidence is routinely flagged and removed, or significantly downranked, effectively reducing its reach to near zero. This creates a reach gap where sanitised, promilitary propaganda spreads freely, while the harsh reality of the conflict, which is essential for historical truth, justice, and the right to remedy, is rendered invisible. This is not just a moderation error but is the erasure of a people's history by an automated system prioritising so-called safety metrics over human rights obligations.