JAKARTA - Gemini this week introduced the seventh extension of Data Commons, a Google initiative aimed at organizing "public data available in the world and making it more accessible and useful."

Since 2017, the Commons Data team has sought to standardize and process thousands of data sets from publicly and reliably available sources. This includes data from the Intergovernmental Panel on the United Nations Climate Change, the Brazilian Institute of Geography and Statistics, and the United States Department of Commerce, with more than 200 sources to date.

There are thousands of data sets that include demographics, economy, education, housing, public health, climate, sustainability, and biomedicals. Google notes that "Commons data doesn't collect or have any data."

In September 2023, Google began implementing a big language (LLM) model to allow users to ask questions in a natural way. However:

"The results come directly from Commons Data, including links to the original data source; thus the output is not generated by the LLM. This approach allows Commons Data to avoid some of the known limitations of the LLM related to accuracy in some cases."

You can now access @Data Commons using Gemini. This can specifically "visualize data on complex topics such as climate change, occupation, economy and health." For example: "Show me details of the type of business in Florida." This was announced in May as part of a wider launch of Gemini Education.

This extension is enabled by default, and joins Google Flights, Google Hotels, Google Maps, Google Workspace, YouTube, and YouTube Music.

Speaking of Gemini Extensions, Google made it "available in Beta versions for additional Google Workspace Gemini subscribers" today. It includes Gmail and Drive, with Google noting that "the Workspace data will not be used to train Gemini's public models."

This new extension brings strong data visualization capabilities to Gemini, allowing users to better understand and analyze data from various sources and efficiently


The English, Chinese, Japanese, Arabic, and French versions are automatically generated by the AI. So there may still be inaccuracies in translating, please always see Indonesian as our main language. (system supported by DigitalSiber.id)