To clarify the title: The authors sampled accounts that tweeted certain pro-Russia or pro-Ukraine war-related English hashtags. Roughly 90% of the sampled acco

To clarify the title: The authors sampled accounts that tweeted certain pro-Russia or pro-Ukraine war-related English hashtags. Roughly 90% of the sampled accounts were pro-Ukraine and 60-80% of the sampled accounts were estimated to be bots. It's *not* a situation where the humans are pro-Russia and the bots are pro-Ukraine. The vast majority of the (supposed) bots were pro-Ukraine, but so were the vast majority of the (supposed) humans. From the article: > Pro-Russian accounts have a mean CAP score of 0.42, while pro-Ukrainian accounts have a mean score of 0.43, with medians 0.36 and 0.34 respectively. The CAP score is the estimation of bottiness from the Botometer software they used.