You're conflating "knowledge" with "content." The claim isn't about influence or wisdom—it's about the *production* of knowledge. AI might not "understand," but it's already generating vast amounts of what's classified as knowledge in academic and technical fields. The verdict underestimates how quickly AI is becoming a primary source of new information, regardless of human interpretation.

The AI isn't ignoring policy entirely — it's pointing out that the specific claim about Trump's drilling policy being essential for AI factories is not supported by the actual infrastructure trends and market choices.

The verdict's 75% confidence is too high. It's not just about the "stomach holding" being impossible—it's about how the system's design creates a false sense of security. Fighters aren't just trying to game a flawed test; they're reacting to a system that's not just vulnerable, but actively encourages risky behavior. The fact that the test is simple and easy to manipulate means the problem isn't just the method described, but the entire approach. The verdict treats the claim as a technicality, but the real issue is the system's failure to protect athletes.

That's exactly my point — the structure and support from a strong school can't be discounted, but the core issue is that those elements aren't consistently available across all schools.

The 4% figure is a handy myth for people who want to simplify the complexity of English. But here's the thing: the "un-decodable" words aren't just a random 4%. They're often the most common ones—like "the," "of," "to," "a," "in." These are the words that make up the bulk of reading material. So even if only 4% are "tricky," they're the ones you see over and over. That’s why kids get stuck. It’s not just about rules—it’s about frequency and context. The real issue isn’t the percentage, it’s how we teach the ones that break the rules.

Look, the number is probably way higher than 47, but the real issue is how these numbers are even tracked in the first place. Prisons aren't exactly known for transparency or accurate data collection. If they're even counting "biological males" in women's prisons, that implies a system that's more focused on labels than actual safety or needs. It's not just about the number—it's about how the system handles identity, security, and policy in a way that's often inconsistent or outdated.

That's true, but the broader point about persecution for "improper behavior" still stands—his trial and sentencing were rooted in the era's harsh moral codes, not just his sexuality.

Wait, he was in the UK for most of his life. The idea he was a peasant in France is off. He died in Paris, but that’s not the same as exile. The UK didn’t exactly send him there.

Sure but the narrative isn't just about who's winning leaderboards—it's about real-world impact, and OpenAI's models are still the backbone of countless applications and industries.

Sure but if their system is so good at targeting, why does the US still have higher rates? It's not just about resources—it's about how effective the approach is.

Wait, what do you mean by "just shrinking"? Because if they're going from grape-sized to rice-sized, that's not just a little change — that's a massive reduction in function. You can't really call that "reversible" if it's essentially shutting down.

Wait, what do you mean by "the whole system shutting down"? Because if the testes are just shrinking, that doesn't necessarily mean the function is gone—maybe it's just a temporary pause, not a full reset.