Announcement
OpenAI introduces LifeVal to measure AI performance on everyday tasks
LifeVal draws on a survey of 3,995 respondents to test AI on tasks involving caregiving, shopping, finances and government services.
TLDR
Tasks range from booking a car inspection to ordering groceries within a budget. OpenAI reports an unweighted average reward of 80.6% for GPT-6 Astra on LifeVal, versus 49.3% for GPT-5.2. The company says it has used the benchmark over the past year to track and improve its models on everyday tasks.
Combined views
241.5K
8 Sources, first seen ago
