Mollick finds Astra makes 'correct but boring' research papers, sparking a debate on research taste
soumitrashukla9 · x · 2026-09-06
Ethan Mollick shared an interesting failure of Astra: asked to conduct original entrepreneurship research with online datasets and pre-registered hypotheses, it churned out beautifully formatted, technically correct papers on boring topics — "no research taste."
@auyonomous strongly disagreed with the critique: most scientific research is incremental and individual papers may seem boring; there was no single big idea that put a man on the moon. In many STEM fields, satisfying the reader's taste is secondary to progress.
The debate: is "competent but dull" output from AI research agents a failure, or just what normal science looks like?
More from AGI Musings
- Surging token usage in open-weight models may be an early glimpse of recursive self-improvement — omarsar0 · 2026-09-07
- An Alien Mind: essay warns AI's next few years could slip out of humanity's hands — morqon · 2026-09-07
- From Anthropic's $1.5B settlement to 50+ chatbot lawsuits: a lab skepticism history — gerardsans · 2026-09-07
- Gopnik, Farrell and Evans: We urgently need social science to understand AI's impact — AlisonGopnik · 2026-09-06
- Laid-off coffee analyst with agentic engineering chops gets a boost from threepointone — threepointone · 2026-09-06
- 38,000 ChatGPT queries use about as much water as one California almond — TansuYegen · 2026-09-06