Fetching from the wire…
01
02
03
04
05
06
07
08
Research2026-08-28 · source-backed
Edison Scientific hands an agent a research objective, methodological guidance and raw data from a published study, then asks it to run the whole analysis chain (arXiv 2608.25286). Across 13 frontier models on 20 paper-derived tasks generating 138 artifacts, scores ran from 0.00 for Gemini 3.1 Flash Lite to 0.48 for GPT-5.6 Sol, where 0.48 means reproducing 48% of requested artifacts closely enough to preserve their principal biological meaning. Most frontier models sit near the floor.
Each link below shares sources, entities, or timing with this story.
Claude Code benchmarked against GPT / Shared entities / Same source domain / Shared topic / Earlier coverage / Tension
Linked by a graph relationship (Claude Code benchmarked against GPT); both cover Gemini, GPT; reported by the same outlet (arxiv.org).
Linked by a graph relationship (Claude Code benchmarked against GPT); both cover GPT, Most; reported by the same outlet (arxiv.org).
Gemini competes with Claude / Shared entities / Same source domain / Shared topic / Earlier coverage / Tension
Linked by a graph relationship (Gemini competes with Claude); both cover Gemini, GPT; reported by the same outlet (arxiv.org).
Gemini built by Google / Shared entities / Shared topic / Earlier coverage
Linked by a graph relationship (Gemini built by Google); both cover Gemini, GPT, Most; overlapping topics (frontier, model).
Vercel uses Gemini / Shared entities / Shared topic / Earlier coverage
Linked by a graph relationship (Vercel uses Gemini); both cover Gemini, GPT, Most; overlapping topics (agent, model).
Gemini competes with Claude / Shared entities / Same source domain / Shared topic / Earlier coverage
Linked by a graph relationship (Gemini competes with Claude); both cover GPT, Most; reported by the same outlet (arxiv.org).
Gemini competes with Claude / Shared entities / Shared topic / Earlier coverage / Tension
Linked by a graph relationship (Gemini competes with Claude); both cover Gemini, GPT; overlapping topics (agent, frontier, model).
Gemini competes with Claude / Shared entities / Same source domain / Shared topic / Earlier coverage
Linked by a graph relationship (Gemini competes with Claude); both cover Gemini, GPT; reported by the same outlet (arxiv.org).