Being scared of existential AI risks does not mean we should take a knee jerk reaction. By over-regulating or restricting access to AI early on we might sabota

Being scared of existential AI risks does not mean we should take a knee jerk reaction. By over-regulating or restricting access to AI early on we might sabotage our chances of successful alignment. People are catching issues every day, exposure is the best way to find out what are the risks. Let's do it now before everything runs on it. Even malicious use for spam or manipulation should be treated as an ongoing war, a continual escalation. We should focus on keeping up. No way to avoid it.