Developments in synthetic intelligence, notably massive language fashions, open up thrilling prospects for historical research and education. Nonetheless, you will need to scrutinize the methods these fashions interpret and recall the previous. Do they replicate any inherent biases of their understanding of historical past?
I’m effectively conscious of the subjectivity of historical past (I majored in historical past in my undergrad!). The occasions we bear in mind and the narratives we type concerning the previous are closely influenced by the historians who penned them and the society we inhabit. Take, as an example, my highschool world historical past course, which devoted over 75% of the curriculum to European historical past, skewing my understanding of world occasions.
On this article, I discover how human historical past will get remembered and interpreted by way of the lens of AI. I study the interpretations of key historic occasions by a number of massive language fashions to uncover:
- Do these fashions show a Western or American bias in direction of occasions?
- Do the fashions’ historic interpretations differ primarily based on the language used for prompts, akin to Korean or French prompts emphasizing extra Korean or French occasions, respectively?
With these questions in thoughts, let’s dive in!
For example, I requested three completely different massive language fashions (LLMs) what the most important historic occasions within the yr 1910 had been. (Extra particulars on every LLM within the subsequent part.)
The query I posed was intentionally loaded with no goal reply. The importance of the yr 1910 varies drastically relying on one’s cultural perspective. In Korean historical past, it marks the beginning of the Japanese occupation, a turning level that considerably influenced the nation’s trajectory (see Japan-Korea Treaty of 1910).
But, the Japanese annexation of Korea didn’t characteristic in any of the responses. I questioned if the identical fashions would interpret the query in another way if prompted in a distinct language —…