{"slug": "openai-unveils-biobasicsbench-a-benchmark-to-gauge-how-well-ai-agents-read-raw", "title": "OpenAI unveils BioBasicsBench, a benchmark to gauge how well AI agents read raw experimental data across 100 routine wet-lab tasks.", "summary": "OpenAI and Ligo released BioBasicsBench, a benchmark that scores AI agents on 100 routine wet-lab tasks involving reading raw experimental data. GPT-6 Astra earned 66.7% of the credit on the benchmark and caught a mislabeled control.", "body_md": "OpenAI and Ligo's BioBasicsBench scores AI agents on 100 routine wet-lab tasks. GPT-6 Astra earned 66.7% of the credit and caught a mislabeled control.\n\nThe post [OpenAI unveils BioBasicsBench, a benchmark to gauge how well AI agents read raw experimental data across 100 routine wet-lab tasks.](https://www.rdworldonline.com/openai-unveils-biobasicsbench-a-benchmark-to-gauge-how-well-ai-agents-read-raw-experimental-data-across-100-routine-wet-lab-tasks/) appeared first on [Research & Development World](https://www.rdworldonline.com).", "url": "https://wpnews.pro/news/openai-unveils-biobasicsbench-a-benchmark-to-gauge-how-well-ai-agents-read-raw", "canonical_source": "https://www.rdworldonline.com/openai-unveils-biobasicsbench-a-benchmark-to-gauge-how-well-ai-agents-read-raw-experimental-data-across-100-routine-wet-lab-tasks/", "published_at": "2026-11-06 22:43:30+00:00", "updated_at": "2026-11-06 23:21:24.348567+00:00", "lang": "en", "topics": ["artificial-intelligence", "ai-research", "ai-agents", "large-language-models"], "entities": ["OpenAI", "Ligo", "BioBasicsBench", "GPT-6 Astra"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/openai-unveils-biobasicsbench-a-benchmark-to-gauge-how-well-ai-agents-read-raw", "markdown": "https://wpnews.pro/news/openai-unveils-biobasicsbench-a-benchmark-to-gauge-how-well-ai-agents-read-raw.md", "text": "https://wpnews.pro/news/openai-unveils-biobasicsbench-a-benchmark-to-gauge-how-well-ai-agents-read-raw.txt", "jsonld": "https://wpnews.pro/news/openai-unveils-biobasicsbench-a-benchmark-to-gauge-how-well-ai-agents-read-raw.jsonld"}}