OpenAI unveils LifeVal, a benchmark for AI's practical help
The test covers tasks like childcare, budgeting and government forms, and was partly built from a survey of 3,995 people.
By Standard Topic. AI-generated from the reporting listed under Coverage. Published . Updated .
This story is no longer being updated. It reflects reporting up to .
What happened
OpenAI published LifeVal on November 2, 2026, an internal tool for assessing how well AI handles everyday tasks such as finding childcare, handling money matters and filling in government forms. The company also released research on how its own models performed, and said GPT-6 Astra scored higher than an older model. The evaluation is meant to track AI performance over time.
Why it matters
It gives a way to measure how well AI helps people with practical tasks, and to see whether that changes over time.
Timeline
- Nov 2, 2026: OpenAI published LifeVal, an evaluation for AI on practical everyday tasks.
- Nov 2, 2026: OpenAI released research on its own models and said GPT-6 Astra scored higher than an older model.
Coverage
- OpenAI Wants ChatGPT to Become Your Personal Assistant (cheddar.com)
More from Technology & AI
- China puts humanoid robots to work serving food and sorting parcels
- Newsom proposes sweeping AI safety rules for Washington
- Google releases Gemini 4 Neon AI model with million-token context
- Microsoft and Intel widen their cloud chip partnership
- Google DeepMind rolls out Genie 4 to select subscribers
- AI helped AMD shorten chip design work, OpenAI says