OpenAI introduces BioBasicsBench, highlighting limits in lab-data interpretation
OpenAI introduced BioBasicsBench on November 6, evaluating AI agents on 100 routine wet-lab data-analysis tasks. It cautions that better scores do not establish reliable laboratory judgment. This is research, not an app or API launch. We have not tested it; this article is not medical or laboratory-procedure advice.
The announcement is a public English webpage; local access was not tested. Taiwan appears on ChatGPT web/mobile and API support lists; Hong Kong and Macau do not. This research does not guarantee model access. Model plans, prices and Traditional Chinese support require separate verification.
https://help.openai.com/en/articles/7947663-chatgpt-supported-countries
https://help.openai.com/en/articles/5347006-openai-api-supported-countries-and-territories
https://help.openai.com/en/articles/7947663-chatgpt-supported-countries
https://help.openai.com/en/articles/5347006-openai-api-supported-countries-and-territories
Usage conditions
- Hong Kong
- The announcement is a public English webpage; local access was not tested. Taiwan appears on ChatGPT web/mobile and API support lists; Hong Kong and Macau do not. This research does not guarantee model access. Model plans, prices and Traditional Chinese support require separate verification.
- Macau
- The announcement is a public English webpage; local access was not tested. Taiwan appears on ChatGPT web/mobile and API support lists; Hong Kong and Macau do not. This research does not guarantee model access. Model plans, prices and Traditional Chinese support require separate verification.
- Taiwan
- The announcement is a public English webpage; local access was not tested. Taiwan appears on ChatGPT web/mobile and API support lists; Hong Kong and Macau do not. This research does not guarantee model access. Model plans, prices and Traditional Chinese support require separate verification.
Official source: OpenAI: Measuring how AI agents handle wet-lab science