Scale AI News: STAGING - AI Models Struggle With Expert-Level Benchmark

Scale AI Unicorn News - January 24, 2025

Scale AI and the Center for AI Safety have released findings from 'Humanity’s Last Exam', a new AI benchmark revealing that current AI models answer fewer than 10% of expert-level questions correctly. This highlights the limitations of AI in complex reasoning and knowledge tasks, emphasizing the need for further research and development.

Sources
https://scale.com/blog/humanitys-last-exam-results
https://qz.com/ai-benchmark-humanitys-last-exam-models-openai-google-1851745995

Outline
(00:00:00) Introduction
(00:00:27) Scale AI Unveils Results of Humanity's Last Exam Benchmark

About Linqto
At Linqto, we empower investors with real-time insights and access to private AI companies, ensuring you stay ahead in the fast-evolving tech landscape. Our platform provides unparalleled opportunities to invest in high-potential startups and unicorns before they go public.

View Linqto Products: https://www.linqto.com/products/
Find all the Unicorn News on Linqto: https://www.linqto.com/unicorn-news/
View Scale-ai on Linqto: https://www.linqto.com/products/scale-ai