Halting illegal economic activities requires effective law enforcement, including credible prosecution. Signs of weakening political support for enforcement efforts can reduce the expected costs of illegal behavior. This paper investigates the impact of anti-conservation statements by political leaders on subsequent deforestation in the Brazilian Amazon. We use monthly municipality data from Twitter (currently known as X) in 2019 – during Bolsonaro’s first presidential year – to track anti-conservation political information signals and build a forest-related social-media penetration index. Relying on a shift-share approach, we identify the effect of these anti-conservation signals on deforestation. High exposure to such signals increases forest loss by 2.2–6.6%. Effects are stronger in areas with high opportunity costs of conservation but insensitive to measures of political allegiance. Political cycles with fluctuating conservation commitments undermine otherwise effective enforcement mechanisms and threaten sustainable tropical forest protection.