It's the opposite isn't it? k-anonymity would remove data until you are the same as k others, whereas this adds data (such as first name), so you are not the same as many others?
There's the privacy-utility tradeoff in data anonymisation, but most algorithms focus primarily on privacy. There usually are no parameters that promise any kind of utility, only parameters that promise privacy.
In this case it looks like they want a guarantee on both, which makes sense.
(So yeah, you're right, this definitely isn't just k-anonymity)
