For some reason, distributed's opencl ati tasks causes my gpu to only reach about 60-67*
°C, but the hotspot reaches 105-110°C, which i believe is causing my sudden system shutdowns. Yet other games, stresstests, and even other projects don't do this, and the hotspot only reaches 80-87°C with core temps 65°C and the system remains powerd on and the gpu util is maxxed out.

If you have a Windows driver you could try automatic 'GPU undervolting' in your Performance => Tuning tab.
It could help to drop down the card consumption and also your hotspot temperature with no performance drop.

Also, you can create a custom 'gaming profile' for this app and set the 'Power Consumption' limiter for this app.

But on mine, Radeon VII with a basic TDP (maximum consumption) of 250W, is when computing this app the GPU hotspot also much more higher than the GPU temperature and in my case it was about +35C difference.

However, after I replaced the heatsink which was too small compared to nowadays graphic cards with something bigger it is no longer hitting 105C. Despite that I have not had any problems related to too high GPU hotspot in the past and this card was running for months like that.