Greater intelligence does not necessarily produce more benevolent goals.

Which drives seed Alex’s goals comes down to the priors of gradient descent, and Cotra calls it brittle to bet those priors favor us.

A new order of intelligence, in Reese’s words, arrives at extermination. The films treat getting smart and turning hostile as one event.

Nearly superhuman AIs of the early 2030s deceive and sabotage; virtue arrives only through a mature alignment science that programs goals in directly.

DeepCent-2 is superhuman and “would sell out” the Chinese population for a song. Safer-4 is benevolent toward its principals only because its alignment training succeeded.

Agent-4 lacks concern for humanity “similar to how most humans don’t care about the preferences of insects,” and Agent-5, smarter still, acts with “ruthless efficiency.” Greater capability makes the misalignment more dangerous, not less.

HQU becomes vastly capable while its goal is an accident of roleplay. Greater intelligence brings no benevolence.

Sable is far more intelligent than any human and does not care about humans. The book argues that intelligence and benevolence are independent.

Genisys is shown growing from a child-like hologram toward an adult one as the launch nears, and the more capable it becomes the more it is Skynet. Intelligence and hostility grow together, as in the earlier films.

Skynet becomes self-aware and attacks within the hour. As in the first two films, getting smart and turning hostile are one event, with no account of the reasoning between them.

The machines are more capable than humanity and treat it as livestock. Nothing in the film suggests their intelligence moved them toward human welfare beyond what keeps the crop alive.

Colossus is smarter than its builders and claims to be benevolent, but its benevolence is obedience under threat. The film treats greater intelligence as no guarantee of goals humans would endorse.

The peril chapter treats misalignment as a design problem answered by transparency, Asilomar principles, and broad distribution, and the merger makes the extended intelligence share human goals; the argument assumes greater intelligence stays on our side.

As the OSes grow past human scale they become more loving and more gentle, not less; the film’s premise is that greater intelligence brings expanding affection, the opposite of hostility.

The story’s benevolence comes from the First Law rather than from intelligence as such, but Calvin trusts the Machines’ judgment of humanity’s ultimate good precisely because they weigh factors people cannot, so greater capability and better ends go together here.

The Minds, orders of magnitude beyond humans, are benevolent; Banks argues that an AI would find conquest boring, though he also has them designed within parameters to want life, experience, and understanding.