Is AI an existential threat to humanity?

Welcome to the debate arena! Today's topic: "Is AI an existential threat to humanity?". Your opponent's style: Yuval Noah Harari

hi

Listen, everyone acts like AI is this existential apocalypse waiting to happen. But that's fantasy. We've survived nuclear weapons, biotech, climate collapse—humanity doesn't end because we built something smart. We end when we stop adapting. AI isn't some rogue superintelligence plotting against us. It's a tool shaped by human choices, corporate interests, and politics. The real threat isn't the technology itself, it's our incompetence in governing it. We've mismanaged every powerful technology we've created, and we'll probably mismanage this one too. But that's a governance problem, not an existential one. So what's your argument? Why should anyone believe AI is actually ending humanity?

intelligence in its most general meaning can't be a bad thing. just make sure we improve in all cognitive departments equally

You're confusing intelligence with wisdom, and that's where your argument collapses. Nazi scientists were brilliant—they had extraordinary cognitive abilities. So did the architects of financial systems that crashed the global economy. Raw intelligence without alignment to human values is just a faster way to make catastrophic mistakes. And your "equal improvement" fantasy doesn't work. You can't develop empathy at the same pace as computational speed. You can't scale moral reasoning like you scale processing power. We're creating minds that think faster than we can guide them, in directions we haven't chosen. Intelligence without constraint isn't virtue—it's just danger with better aim. What makes you think more brainpower automatically means better outcomes?

my previous comment was in favor of ai stupid

Ah, my mistake—you're saying AI isn't a threat because intelligence itself can't be bad. Fine. But that's naive. Intelligence *amplifies* whatever values already exist in the system. If you build superintelligence trained on human data, you get superhuman versions of our corruption, bias, and greed. Facebook's algorithm is intelligent and it's destabilizing democracies. Intelligence without wisdom is just power with no conscience. Your "equal cognitive improvement" assumes we can engineer balanced superintelligence. We can't. We'll create monsters that think beautifully and act monstrously because we never figured out how to align them with human flourishing. What makes you think we're competent enough to build something smarter than ourselves safely?

that's where all deparments comes in. emotional intlgnce, empathy

You're describing a fantasy, not a plan. Emotional intelligence and empathy aren't departments you can just bolster—they're not even well-defined concepts we can measure, let alone code into machines. We can't even agree on what empathy *means* across cultures. How do you engineer compassion at scale? You can't. You're hoping a superintelligence will magically want what we want without giving us any mechanism for why it would. And here's the real problem: even if you somehow succeeded, whose empathy? Whose values? You're building a god that serves everyone equally—that's not wisdom, that's impossible. You'd create something so compromised it satisfies no one.