[R] How R calculates SE of prediction for Logistic regression?

[Christofer Bogaso]

Hi,

I have below logistic regression

Dat =
read.csv('https://raw.githubusercontent.com/sam16tyagi/Machine-Learning-techniques-in-python/master/logistic%20regression%20dataset-Social_Network_Ads.csv')
head(Dat)
Model = glm(Purchased ~ Gender, data = Dat, family = binomial())

How I can get Standard deviation of forecasts as

head(predict(Model, type="response", se.fit = T)$se.fit)

My question: given that in Logistic regression, logit link is used,
how R calculate SE for the predicted probability from the VCV matrix
of estimated coefficients?

Does R uses some approximation like delta rule?