That was a problem with the reactor designs of the 1950s to 1990s. Over the past half century a lot of smart people have put a lot of thought into idiot proofing nuclear reactors to prevent another three mile, chernobyl, or fukushima. Reactor designers no longer make optimistic assumptions about the operator and assume they are a shortsighted idiot that cannot be trusted to do the right thing.

In modern reactors temperature coefficients tuned to automatically prevent meltdowns is something regulators care a lot about when approving designs. Rather than focusing on building safety mechanisms that the operator can trigger (ie control rods), a natural safety mechanism is built into the formulation of the fuel so if it gets too hot it is no longer capable of nuclear fission.

This is especially the case in small modular reactor (smr) designs meant to be used in commercial applications where no one actually trusts the operator to be responsible. The fuel is formulated to sacrifice some efficiency in exchange for the reactor automatically SCRAMing even if the operator does everything in their power to keep it running.

There is also a push for SMRs to use things like the thorium fuel cycle because it makes the reactor pointless for terrorists or other bad actors to target. The thorium fuel isn’t useful for radiological attacks or bomb making, the only reason it even works as a fuel is because it can produce small amounts of uranium that are immediately reacted upon forming. This was the entire reason governments ignored these fuel cycles for decades, they didn’t create waste that could be used for weapons making. As a result terrorists are better off getting a shovel and collecting natural uranium off the side of highways in the deserts of North Africa or North America.

https://nuclearsafety.gc.ca/eng/resources/news-room/feature-articles/positive-void-coefficient-of-reactivity-CANDUs.cfm

https://en.wikipedia.org/wiki/Void_coefficient

https://en.wikipedia.org/wiki/Fuel_temperature_coefficient_of_reactivity