Manifold MarketsSem KYCDados de resolução verificadosRazoavelmente calibradoAviso de qualidade dos dadosDados da fonte degradados
Dados de 19 de out. de 2026, 14:10 UTC · política pm-quality-4
Previsão da comunidade atual2035 69.7%Líder entre 6 opções
Previsores
14
Tipo de pergunta
multiple choice
Metodologia
Play-money forecasting platform
Tipo de fonte
Previsão
Variação 7d
0.0 pts
Dados do mercado
Atualizado há 10 horas
Desatualizado16/10/25, 14:41 UTC2/01/35, 03:04 UTC
Tendências
Mercados Relacionados
Perguntas relacionadas para dar contexto, não uma comparação de contratos equivalentes.
Which events will happen before AGI is released (Hendrycks definition >95%)?
Manifold MarketsWill ASI be developed 5+ years after AGI?
Manifold MarketsWill a paper clip maximizing AGI be developed 5 years after AGI is acheived
Manifold Markets10 years before AGI is built, debate about AGI risk will still be less mainstream than global warming was in 2015
Manifold MarketsBest AI at the end of 2026?
KalshiWhich company has the best AI model end of October?
PolymarketRegras
A new highly-detailed operationalization of AGI has been proposed by Hendrycks et al. based on the standard that "AGI is an AI that can match or exceed the cognitive versatility and proficiency of a well-educated adult" (www.agidefinition.ai/).
- Their definition is inspired by Cattell-Horn-Carroll theory and calculate a model’s “AGI Score” as a composite of ten equally-weighted cognitive categories.
- GPT-4 is judged to score 27% according to the definition, and GPT-5 is scored at 58%.
- This market resolves YES for all years that lie AFTER the first date when an AI system, with an AGI Score eventually judged to be above 95%, is deployed and accessible to customers external to the AI developer for commercial or consumer purposes.
- The score would either need to be judged by the authors, or by a third party in a manner credibly commensurate with the original evaluations of GPT-4 and GPT-5.
- The score would need to be announced within 6 months of model release.