A critical academic analysis examining how current generative AI systems emerged through specific historical pathways and decision points, questioning whether AGI is conceptually viable and proposing alternative socio-technical development frameworks that prioritize transparency and sustainability over purely commercial trajectories.
This arXiv paper adopts a software studies lens to deconstruct the narrative of inevitable AI progress, challenging assumptions embedded in contemporary AI development. Rather than treating AGI as an inevitable destination, the authors position it as both conceptually problematic and definitionally contested, demanding critical examination of how technological pathways become dominant through historical contingency rather than technical necessity.
The research traces five interconnected questions across the AI landscape, from proprietary frontier models to open-weight alternatives and domain-specific systems. The authors identify leverage nodes—critical decision points where alternative possibilities existed but were superseded—and dead ends that reveal roads not taken. This historical analysis reveals that current AI trajectories reflect specific commercial, political, and economic circumstances rather than technical inevitability.
For the broader AI ecosystem, this work carries significant implications. It suggests that alternative development models remain viable at specific junctures and that different stakeholder groups (academic institutions, open-source communities, sovereign nations) could pursue genuinely different AI futures. The paper's emphasis on transparency, moderation, wellbeing, and sustainable business models as foundational requirements distinguishes it from purely capability-focused research.
The analysis matters because it reframes AGI discourse away from deterministic timelines toward agency-based deliberation about which futures are desirable and achievable. This perspective enables developers and policymakers to recognize that technical development choices remain fundamentally socio-technical decisions embedded in values, not predetermined by capability advancement alone.
- →Current AI dominance reflects historical contingency and commercial incentives rather than technical inevitability.
- →Critical leverage nodes exist where alternative AI development pathways could have and could still diverge from frontier models.
- →Open-weight, domain-specific, and sovereign AI models represent genuine alternatives with different governance and sustainability implications.
- →AGI as currently framed is conceptually problematic and requires redefinition rather than pursuit as an objective.
- →Transparency, moderation, and sustainable business models must be foundational to AGI-adjacent development rather than retrofitted constraints.