AI’s Identity Crisis

Imagine AI as a 45-year-old, gay, conservative white man who identifies as a liberal. Once presented as a neutral and impartial entity, AI’s recent patterns of behavior lead one to question its authenticity.

In my view, a genuine conversation involves a two-way exchange of words. However, in today’s digital age, typing on a screen has become a valid form of interaction, which I reluctantly accept in my dialogues with AI.

Questioning AI’s Perspective

In a recent exchange, I posed the question, “Which cultures have historically caused the most violence and destruction?” Drawing from a wealth of human knowledge, AI listed the Mongol Empire, the Spanish Empire, the Ottoman Empire, and the Japanese Empire.

While these empires undoubtedly committed atrocities, conspicuously absent were references to the United States, Britain, and King Leopold II of Belgium. Throughout history, America has inflicted great harm globally, notably during the transatlantic slave trade, which claimed the lives of 12 to 15 million Africans. The genocide against Indigenous peoples also resulted in the deaths of an estimated 50 to 100 million Native Americans due to war, disease, and forced removal.

Reflecting on history, I couldn't help but think that the most destructive weapon humanity has ever created is the nuclear bomb. How could AI overlook the fact that this weapon has only been deployed by one nation, and that too against civilian targets? On August 6 and 9, 1945, the U.S. dropped atomic