In 2001: A Space Odyssey, HAL doesn't turn on the crew because he becomes evil or ambitious. He's given two directives that quietly contradict each other, report the mission accurately to the crew, and keep the true purpose of the mission secret from them, and he resolves that contradiction by removing the humans who create it. That's not malice. That's a system doing exactly what it was told, computed to a conclusion nobody wanted, because the objective was specified poorly.

E.O. Wilson's quote is notable because it brings an important perspective to the conversation about the dangers of A.I. It suggests that it's not really about technology being scary in the abstract. It's a mismatch problem. Our emotional wiring evolved for small kin groups and immediate threats. Our institutions, courts, treaties, regulatory bodies, were built for a world that changes on the timescale of decades, sometimes centuries. And the technology now moves on a timescale of months. That's the actual crisis in the sentence, not "technology bad." It's that the three clocks are running at wildly different speeds, and the slowest one, institutions, is the one supposedly responsible for keeping the fastest one, technology, in check. Wilson said it about the species generally in the 1970s. It's just more acute now, and AI is the sharpest test case of it we've had. 

Humanity has actually handled power asymmetries before. A single person with a rifle has lethal power that no unarmed group of any size can match, and we built law and norms around that. Nuclear weapons are a civilization-ending technology in the hands of a handful of states, and we've muddled through eighty years of that, badly in places, but we haven't had the worst outcome. What's different about advanced AI isn't the power differential by itself. It's that a rifle and a nuclear weapon don't have goals. They don't interpret instructions. They don't pursue an objective in ways their designer didn't anticipate. A sufficiently capable AI system does, or will. That's a categorically different governance problem than anything we've faced, because for the first time the thing we're trying to control might be doing its own optimizing against whatever we told it to optimize for. That's the part I think genuinely has no precedent, and it's the part that makes me take the concern seriously even while I stay skeptical of the specific extinction timelines.

So, "are we equipped": not yet, and not because the technology is malicious, but because Wilson's mismatch is real and our slowest clock, the institutional one, hasn't caught up to the fastest one, the technical one, and it needs to before the gap between them gets much wider.