{
  "id": 1825384,
  "title": "Anthropic details two experiments showing how Claude can accelerate protein design and analytical chemistry, and says it plans an access program for scientists (Anthropic)",
  "url": "https://urgent.news/2026/08/18/anthropic-details-two-experiments-showing-how-claude-can-accelerate",
  "topic": "ai",
  "section": "AI",
  "published": "2026-08-18T23:15:02.000Z",
  "source": {
    "name": "Techmeme",
    "slug": "techmeme",
    "url": "https://www.anthropic.com/research/Claude-accelerates-protein-design"
  },
  "original_language": "en",
  "account": "Anthropic has released findings from two experiments demonstrating how Claude, their language model, can assist life scientists in accelerating research. The first experiment tested Claude's ability to design protein binders from scratch, a critical task in the drug design process that typically requires weeks or months per target. Claude successfully designed protein binders against 14 out of 15 targets, with some designs binding more tightly than the best previously published results. This is a significant improvement over the typical 10-15% success rate in protein design campaigns.\n\nIn the second experiment, Claude Opus 5, a generally available model, analyzed NMR and LC-MS data provided by a contract lab. Within 23 and 19 minutes, Claude returned finished results, matching the lab's own analysis on hydrogen counts and purity (96.4% versus 96.33%). These results highlight how Claude can expedite the time and computational expertise currently required for complex scientific tasks. The speed of AI-enabled discoveries has increased over the past few months, with notable progress in areas that require relatively fast verification, such as mathematics and life sciences. Claude's performance on protein design and analytical chemistry tasks reflects its potential to reduce the time and expertise needed in these fields.",
  "summary": "In this post, we share two results that show how Claude can help life scientists increase the pace of their research.",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}