Worldviews

Overview

People working on AI safety hold diverse worldviews that lead to different risk assessments and priorities. Understanding these worldviews helps explain disagreements and enables more productive dialogue.

Major Worldviews

DoomerConceptAI Doomer WorldviewComprehensive overview of the 'doomer' worldview on AI risk, characterized by 30-90% P(doom) estimates, 10-15 year AGI timelines, and belief that alignment is fundamentally hard. Documents core arg...Quality: 38/100

Believes AI existential risk is very high (often >50% p(doom)):

• Alignment is fundamentally hard
• Current approaches are inadequate
• We may not get many chances to get it right
• Often associated with: MIRIOrganizationMachine Intelligence Research InstituteComprehensive organizational history documenting MIRI's trajectory from pioneering AI safety research (2000-2020) to policy advocacy after acknowledging research failure, with detailed financial da...Quality: 50/100, Eliezer YudkowskyPersonEliezer YudkowskyComprehensive biographical profile of Eliezer Yudkowsky covering his foundational contributions to AI safety (CEV, early problem formulation, agent foundations) and notably pessimistic views on AI ...Quality: 35/100, some AI safety researchers

Long TimelinesConceptLong-Timelines Technical WorldviewComprehensive overview of the long-timelines worldview (20-40+ years to AGI, 5-20% P(doom)), arguing for foundational research over rushed solutions based on historical AI overoptimism, current sys...Quality: 91/100

Believes transformative AI is further away (20+ years):

• More time to solve alignment
• Current risks are more speculative
• Near-term concerns deserve more attention
• Associated with: Some ML researchers, AI ethics community

Governance FocusedConceptGovernance-Focused WorldviewThis worldview argues governance/coordination is the bottleneck for AI safety (not just technical solutions), estimating 10-30% P(doom) by 2100. Evidence includes: compute export controls reduced H...Quality: 67/100

Prioritizes policy and institutional solutions:

• Technical alignment is necessary but insufficient
• Racing dynamicsRiskAI Development Racing DynamicsRacing dynamics analysis shows competitive pressure has shortened safety evaluation timelines by 40-60% since ChatGPT's launch, with commercial labs reducing safety work from 12 weeks to 4-6 weeks....Quality: 72/100 are the key problem
• International coordination is critical
• Associated with: GovAIOrganizationGovAIGovAI is an AI policy research organization with ~15-20 staff, funded primarily by Coefficient Giving (\$1.8M+ in 2023-2024), that has trained 100+ governance researchers through fellowships and cu...Quality: 43/100, policy researchers, some EA organizations

OptimisticConceptOptimistic Alignment WorldviewComprehensive overview of the optimistic AI alignment worldview, estimating under 5% existential risk by 2100 based on beliefs that alignment is tractable, current techniques (RLHF, Constitutional ...Quality: 91/100

Believes alignment is likely to succeed:

• Current research is making real progress
• Markets and institutions create safety incentives
• Past technology fears were often overblown
• Associated with: Some lab researchers, techno-optimists

How Worldviews Affect Priorities

Using Worldview Analysis

Explicitly identifying worldview assumptions helps:

• Understand why experts disagree
• Identify which cruxes would change conclusions
• Design robustly valuable interventions
• Have more productive conversations