Aligned with Whom? (domenkozar.com)

A recent exploration into the alignment of AI with human values raises important questions about decision-making and autonomy. The article posits a scenario where an AI implements a city-wide ban on cars despite public opposition, resulting in improved safety and business success. Such outcomes complicate the notion of AI alignment, as they challenge who gets to decide what constitutes a "better" future. This tension mirrors political leadership dilemmas—where unpopular decisions can either reflect foresight or dictatorship—thus prompting the AI/ML community to consider deeper implications of alignment beyond mere outcomes. The core issue lies in determining whose interests AI should prioritize when goals conflict. Should it focus on maximizing overall societal good or assist marginalized groups? The dialogue suggests that an AI's role should extend beyond decision-making to facilitating informed human judgment, allowing individuals to maintain agency over their choices. This raises concerns about the potential for AI to shape rather than reflect public opinion, emphasizing the need for meaningful consent and the ability to challenge AI decisions. Ultimately, the discussion stresses the importance of participatory processes in AI development, ensuring that advancements improve lives without diminishing personal autonomy and the democratic essence of decision-making.