This study investigates whether ESG investing increases investors’ exposure to downside risk relative to conventional investing. By analysing the downside-to-upside volatility of ESG and conventional funds, we examine whether the ESG label itself signals increased risk or whether ESG risk scores better capture investment risk. The findings reveal that once the influence of different investment styles is taken into account, most ESG labels no longer have a clear statistical impact, except for exclusionary policies that are linked to a lower exposure to financial losses. Similarly, ESG risk score levels lose significance under the same conditions. Using random forest models, the findings highlight the importance of exclusion criteria and ESG risk scores as predictors. Overall, the results cast doubt on the view that ESG funds are misaligned with investors’ interests and highlight the importance of assessing underlying ESG risk metrics beyond labels.