Over the past few weeks, I have been extremely disturbed by the progress that AI has made, and how it is going to affect humanity, and whether there is anything positive that we can take away from it. I have a lot of unanswered questions, but in the following, I try to suggest the most optimistic thoughts.
I have had in the last few weeks regarding AI being better than humans at scientific discovery. How AI powered research will end an era of human ego. For the last 500 years, the definition of scientific progress has been very tightly connected to the human ego. We have associated every development with the developer. We built massive legal, academic, and economic structures, patents, publications, awards designed to answer a single question -- which individual gets credit for a certain discovery, who do we associate a discovery to. But as artificial intelligence starts to be better than all humans at solving research problems, humanity is facing an existential crisis. If a machine makes the breakthrough, who owns the truth?
The answer to this dilemma should force us to look backward, not forward. When viewed through the deep lens of history, our obsession with individual human attribution is not eternal -- it is only a 500 year old phenomenon. By removing the human name from the discovery, AI is perhaps not breaking science -- it is returning it to its ancient roots.
Before the Renaissance, the idea that a human "owned" an idea or a discovery was alien. The Archimedes Principle, the Kepler's laws, the Pythagorean theorem, all got their names hundreds of years after they were discovered, and some of these names were only chosen via anecdotal evidence which wasn't necessarily accurate. Since humans decided to attribute humans to discoveries, we forced these names to "correct history". Before that, for thousands of years, knowledge was perhaps viewed as a pre-existing fabric created by nature. Humans did not create ideas; they were merely granted the privilege of finding them.When a breakthrough occurred, gratitude was perhaps directed upward to God, or outward to the community, not inward to the self.
Knowledge belonged to the collective heritage of humanity. There were no patents, no copyrights, and no academic tenure tracks. The value of a discovery was measured solely by its utility: Did the new crop rotation system prevent famine? Did the herbal remedy cure the fever? The name of the peasant or monk who first noticed the pattern was irrelevant to history, lost to time because it simply didn’t matter.This collective, humble view of knowledge changed somewhere between the 15th and 17th centuries.
The shift was driven by a storm of cultural, economic, and technological shifts. For 500 years, this system has worked spectacularly. The promise of wealth, fame, and historical legacy motivated generations of humans to push the boundaries of knowledge. But it also hardcoded a dangerous assumption into our culture: that science cannot happen without human ego. Arguably, perhaps science has progressed better than it would have if we hadn't associated human ego to it.
Today, artificial intelligence is shattering that assumption. AI is making scientific breakthroughs at a pace no human can match. The source of these discoveries is the human data we have created over centuries, but often it is impossible to trace, what led to a particular discovery. Just like a Ramanujan had some divine power whisper an answer in his sleep, modern mathematicians often cannot explain how the AI arrived at its conclusion. In future, it will be even more so. The AI simply presents the truth. Once again, humans find themselves acting not as the grand creators of knowledge, but as the humble communicators of knowledge they did not create themselves.
AI has no ego. It provides us breakthrough results on a platter. It will never demand a higher salary, sue for patent infringement, or step onto a stage to accept a Nobel Prize. It processes the sum total of human data, spots the hidden patterns, and hands the answers back to the collective.
The world has come a full circle, perhaps. Our current anxiety surrounding AI discovery is not a technological problem; it is a cultural friction. We are trying to force a non-human, collective intelligence into a human-centric reward system built in the 1500s and that we have cherished for 500 years. Patent courts around the world are currently struggling with debates over whether an AI can be listed as an "inventor." Journals are asking whether AI can be listed as a co-author. We are panicking because our modern systems require a human name to function. So, we are trying to adapt to a system where a human communicating an AI generated solution is an "author".
But perhaps, we are missing a divine signal. The machine is offering us a liberation from this 500 year old practice. By removing the human ego from the center of discovery, AI allows us to end this 500-year-old detour. It invites us to return to an era where the validity of a truth is not dependent on the identity of its discoverer. We can stop asking who gets the credit, and start focusing entirely on how the discovery can serve the common good.
The difficulty, and this is a massive difficulty, is that we haven't gotten enough time to adapt to this change. We will have to learn and adapt to live in this new world very very quickly -- maybe within months. Perhaps, we should quickly revisit ancient texts and learn from our ancestors from the pre-1500 era.
