I've seen more criticism of facial recognition surveillance on the grounds that it is less accurate for some races, than on the grounds that, if it worked perfectly, it would result in a massive loss of freedom, so this is no strawman:

https://projects.iq.harvard.edu/sciencepolicy/blog/racial-di...

https://theswaddle.com/how-facial-recognition-ai-reinforces-...

https://www.aclu.org/news/privacy-technology/how-is-face-rec...

https://www.cbsnews.com/news/facial-recognition-systems-raci...

https://www.csis.org/blogs/technology-policy-blog/problem-bi...

https://www.cnet.com/science/why-facial-recognitions-racial-...

https://theconversation.com/ai-technologies-like-police-faci...

https://www.washingtonpost.com/technology/2019/12/19/federal...

https://exposure.org.uk/2021/09/why-facial-recognition-tech-...

https://www.theguardian.com/technology/2017/dec/04/racist-fa...

https://www.nytimes.com/2018/02/09/technology/facial-recogni...

https://www.amnesty.org/en/latest/press-release/2021/01/ban-...

https://www.npr.org/2020/06/24/882683463/the-computer-got-it...

https://www.usatoday.com/story/tech/2019/12/19/facial-recogn...

https://www.theatlantic.com/technology/archive/2020/07/defun...

But in the interest of accuracy, "only" was hyperbole - there is some criticism on non-discriminatory grounds. It's just overshadowed by concerns on whether the hell we are building will be racist.