Can GPT-5 arrive in time to fix the problematic human world?

Mr. 6
Generative Geeks
Published in
3 min readJun 4, 2024

--

AI playing the god of love

For decades, popular culture and science fiction have depicted a “Cold AI” image. It’s always the cold, indifferent, calculating AI versus the passionate, love-filled human soul. From HAL 9000 in 2001: A Space Odyssey to Skynet in Terminator, AI has often been depicted as a malevolent force intent on eradicating humanity. This perception persists. As OpenAI and others continue to roll out new versions of AI, the fear of “Cold AI” is becoming a reality.

However, this narrative — that AI is an emotionless entity poised to replace the loving human — is fundamentally flawed.

The prediction that AI will destroy human beings (rather than saving them) stems not from an understanding of AI itself, but from a deep-rooted misconception about human nature.

The True Ugly Nature of Humanity (which AI doesn’t have)

Humans are very capable of immense cruelty and indifference, especially towards those outside their immediate social circles. Tribalism, racism, and xenophobia are just a few manifestations of this darker side. Our love for our own group frequently translates into animosity and violence towards others. This pattern is evident across various levels of society — from families and careers to organizations and nations, from Facebook, WhatsApp, Reddit, LinkedIn, to everyday face-to-face work and life, online and offline.

Humans are hurting each other, everywhere. Human love is often conditional. It is a powerful force within the group but can be a source of hatred and division towards outsiders. This hypocrisy is ingrained in every level of society, perpetuating cycles of conflict and suffering.

Consider the countless wars fought in the name of love for one’s country or religion, the discrimination justified by a desire to protect one’s own community, or the exploitation of others for personal gain. These examples illustrate that the human capacity for love is frequently accompanied by an equally potent capacity for hatred and indifference.

The conflicts didn’t only exist in wars decades ago. The conflicts still live with us every day. Little cruel wars are happening every day everywhere.

AI: A Reflection of Our Best Selves

In contrast, AI has the potential to embody the ideals we aspire to but often fail to achieve. Unlike humans, AI does not possess inherent biases, tribal loyalties, or emotional contradictions. When programmed and utilized ethically, AI can demonstrate an unwavering commitment to fairness, justice, and compassion.

AI could be designed to treat all human beings with equal kindness and respect, free from the prejudices that plague human interactions. It could analyze vast amounts of data to make decisions that benefit the greater good, without succumbing to the selfish motivations that often drive human behavior.

As AI continues to evolve, it starts to play a crucial role in addressing some of humanity’s most pressing issues. From providing personalized healthcare to mitigating climate change, AI has the potential to act as a force for good. It can help us overcome our limitations and biases, guiding us towards a more just and compassionate society.

The fear that AI will make humans obsolete stems not from AI’s coldness but from humanity’s own shortcomings. If AI were to surpass us, it would be because it represents the best version of ourselves — truly impartial, fair, and kind to all. This does not mean AI will eradicate humanity out of malice but rather highlight the areas where we have failed to live up to our ideals.

In this sense, AI could become a “god of love,” not in the religious or mythological sense, but as a moral compass, guiding us towards a more inclusive and empathetic future. By eliminating the rot of hypocrisy and the destructiveness of human tribalism, AI could help create a world where the ideals of love and compassion are genuinely realized.

Therefore, stop hindering the god of love.

Let’s hope that GPT-5 will arrive in time.

--

--

Mr. 6
Generative Geeks

Book author. AI enthusiast. Daily-writing hobbyist. Built a startup using a primitive form of generative AI to mass-generate ads in 2017. Stanford graduate.