{
    "statement": "task success for AI agents on OSWorld, a test of real computer tasks, up from 12%; agents still fail about 1 in 3 attempts",
    "value_display": "About 66%",
    "value_numeric": 66,
    "unit": "%",
    "data_label": "2025-2026",
    "data_period_end": "2026-03-31",
    "region": "Global",
    "size": "all",
    "source": {
        "name": "AI Index Report",
        "publisher": "Stanford HAI",
        "type": "research",
        "url": "https://hai.stanford.edu/"
    },
    "source_url": "https://hai.stanford.edu/ai-index/2026-ai-index-report",
    "source_quote": "AI agents made a leap from 12% to ~66% task success on OSWorld, which tests agents on real computer tasks across operating systems",
    "verification_note": "Read on hai.stanford.edu (2026 AI Index Report, top takeaways) on 2026-10-23. Benchmark score, not real-world adoption.",
    "verified_at": "2026-10-23T00:00:00+00:00",
    "margin_of_error": null,
    "dont_know": null,
    "data_table": null,
    "revisions": [],
    "page_url": "https://digitaltreed.com/statistics/ai-agents/agent-success-osworld",
    "csv_url": "https://digitaltreed.com/statistics/ai-agents/agent-success-osworld.csv"
}