IA-Libris, a ChatGPT verified by specialists
ChatGPT is an artificial intelligence model developed by OpenAI. In its current design, ChatGPT uses a large amount of text from the Internet as its primary source of learning. However, you may not access or directly cite specific sources or provide the original sources of your information. In addition, you cannot learn or update your knowledge in real time or after your last training date, which in the case of the GPT-4 version is September 2021.
In this first version, the IA-Libris Chat uses GPT-4 with models trained by us, focused on the books in the Linkgua Ediciones catalog.
Our artificial intelligence models have an additional verification function in addition to the current ChatGPT. Based on the same technology, we have developed a ChatGPT verified by expert hispanists, with continuous training. A team of specialists from various disciplines works to carefully filter and select the data we use, focusing only on information from reliable sources, such as peer-reviewed academic journals, high-quality textbooks, government databases and other recognized information repositories.
In its real-time interaction with users, the enhanced IA-Libris Chathas the ability to access updated databases and verify information in real time, bringing a great improvement to its reliability. In addition, you can directly cite the sources of your information, allowing users to verify our answers for themselves.
Since this model is based initially on the books in our catalog, and on our knowledge of our content, we have implemented improved processes to detect and prevent the spread of false or misleading information, and we have mechanisms in place to learn and update its knowledge base in a continuous and dynamic manner.
The following are real examples of use. If you ask ChatGPT-4:
What is the relationship in the National Episodes between Queen Isabella and Juan Prim?
The response of June 16, 2023 is:
“To determine the relationship between Queen Isabella and Juan Prim in the “Episodios Nacionales”, I would recommend you read these books or look for academic or literary analyses of these characters in this series of novels. Juan Prim was a Spanish general and politician, and Queen Isabella II was the queen of Spain, so both characters would have interacted in the historical context of 19th century Spain. The depiction of their relationship in the “Episodios Nacionales” would be a literary interpretation of these historical events by the author, Benito Pérez Galdós.”
If you ask in IA-Libris Chat, the answer is:
“According to the fragments provided from the National Episodes, it is mentioned that Queen Isabella and Juan Prim had several cordial and friendly meetings. Prim came to the conclusion that Isabel was considering the adulterated government plan put forward by Euphrasia. However, Prim did not commit himself to this plan and did not form a Cabinet with the confidence of a few short-coated clergymen. No further information about the relationship between Queen Isabella and Juan Prim is provided in the fragments given.”
SOURCE: National Episodes IV. Prim
Our answer comes from the analysis with Artificial Intelligence of our editions of the Episodios Nacionales, by Benito Pérez Galdós, duly verified by specialists.
This is just one example of how it is possible to improve the answers to complex questions of teaching interest and how it is possible to create a knowledge management environment in which humans have the control and the ability to contrast the answers of an Artificial Intelligence.
