Welding is a process with an uneven temperature distribution: heat is supplied from the arc side and heat is dissipated from the component and electrode side. This creates temperature gradients that depend on the thermal conductivity of the component and the electrode. The higher the thermal conductivity of the material, the more energy is required to melt it. At the same time, tungsten has a very high melting temperature and a very high thermal conductivity, which is why it cannot be melted by an arc (the temperature at the tip of the electrode remains below the melting point for all arc parameters).

You can also observe the same effect of good heat dissipation from the side of the component when you are welding copper: despite the relatively low melting temperature of copper, you need more energy for welding compared to carbon steel, because copper removes heat from the welding zone 10 times better.

Vadim Verlotski In addition to your comments, tungsten tip is being cooled by the inert gas supply in TIG welding process.

By the way, tungsten electrode would melt, albeit slowly, this is why in GTAW process, we use DCEN to reduce the "heat" to the tungsten electrode. If you use DCEP, the electrode tip would start melting.