This thesis analyzes Instagram’s content moderation policy documents to look into how the platform frames moderation and its stance on harassment and harm toward marginalized communities. Using Fairclough’s three dimensional model for critical discourse analysis, I study what the text says (its language and wording), how it can be interpreted, and why it might be written that way.

The findings suggest that Instagram uses vague language to avoid accountability while maintaining power behind discourse, and exert considerable control over user expression, by defining what is acceptable and what is not. This control also includes the representation of bodies, particularly through policies that regulate nudity in gendered ways, reinforcing patriarchal norms and cultural discomfort with female bodily autonomy.

On the basis of the findings, it can be concluded that Instagram presents itself as progressive and inclusive, but at the same time avoids strong commitments to DEI in its actual content policies. It allows expression of political and religious views that may stigmatize LGBTQ+ people. In using such a strategy to please both progressive and conservative audiences, Instagram appears neutral on the surface, avoiding clear, ethical stances, which indicates a priority in platform growth and reputational risk management over marginalized users’ safety. In the end, Instagram’s version of neutrality tends to protect dominant ideologies more than the people those ideologies often harm.