A computer scientist is questioning the feasibility of artificial superintelligence (ASI), arguing that the concept of recursive self-improvement (RSI) is speculative and not grounded in current engineering realities. Despite this skepticism, prominent figures at AI labs like Anthropic and OpenAI have expressed concerns about the existential risks posed by pursuing ASI, with some researchers even resigning over these fears. The debate highlights a divergence between theoretical risks and practical engineering challenges, with potential harms arising even if ASI never materializes. AI
IMPACT Questions the feasibility of ASI, potentially recalibrating industry focus and risk assessment.
RANK_REASON Article discusses the theoretical feasibility of ASI and the differing opinions on its risks, rather than announcing a new model or product.
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →