>Because whilst those companies are slower to respond (due to the volume of content their users generate) they are quite active at removing hateful and harassing content.
This is goalpost-shifting. Kiwifarms does moderate threats and such. It does not moderate mere "hateful and harassing content" (whatever that means, I assume a lot of "harassing" content does overlap with illegality), but if the clear no-no actions are swatting and such then clearly KF actively tries to prevent that.
I fear that the implicit argument here is that in order for these efforts to be treated as "good faith", then forum owners need to not only remove clear legal threats but all sorts of speech (particularly "hate speech") that can be stochastically linked with the generation of those threats. That having sufficient moderation entails that moderators address not only actual threats but in removing a wide range of speech.
