---
format: "aidr-story-markdown/v1"
id: "aca951de3e07af39364eb1d43ae67a3526bd174964f73a83093106cee78e90f8"
canonical_url: "https://aidr.today/aca951de?lang=en"
title: "No AI agent scores above 12 percent on SemiAnalysis' new real-world work benchmark"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-11-05T16:46:24.000Z"
category: "Agents"
topics: ["agent"]
source_urls: ["https://the-decoder.com/no-ai-agent-scores-above-12-percent-on-semianalysis-new-real-world-work-benchmark/"]
summary: "No AI agent scored above 12 percent on SemiAnalysis' new GDPx benchmark. Claude Opus 5.5 led narrowly over Gemini 4 Carbon. Critics point to the small number of tasks per domain and uneven testing conditions."
---

# No AI agent scores above 12 percent on SemiAnalysis' new real\-world work benchmark

> [Open the canonical story](<https://aidr.today/aca951de?lang=en>)

**Published:** 2026-11-05T16:46:24.000Z
**Category:** Agents
**Topics:** agent

## Summary

No AI agent scored above 12 percent on SemiAnalysis' new GDPx benchmark\. Claude Opus 5\.5 led narrowly over Gemini 4 Carbon\. Critics point to the small number of tasks per domain and uneven testing conditions\.

## Sources

- [Story source](<https://the-decoder.com/no-ai-agent-scores-above-12-percent-on-semianalysis-new-real-world-work-benchmark/>)