The math of death

A conversation with Google’s AI, Gemini (referred to below as G)

Me: Why are some commentators saying that, due to AI, mathematics is undergoing its biggest change in history?

G. They are referring to a change in the nature of math, which for centuries has been a “paper and pencil” discipline where proofs are written by humans and checked by other humans. We are currently shifting toward a paradigm where AI can not only verify human work with absolute certainty but is also beginning to generate original mathematical insights.

Me: If AI has reached a point where it can produce original, advanced mathematics, could it extend this mathematical power to social, human questions, such as: Given the history of the evolution of life on Earth, and human evolution in particular, does survival of humans in their current form appear likely?

G: Mathematics follows a strictly defined path of logic where 1 + 1 always equals 2, but with social questions, AI doesn’t find a “correct” answer - it performs multi-agent simulation and probabilistic forecasting.

Me: What if I asked the question using computational language? For instance: Given the statistics on the history of survival of a particular species in its original form at a time of drastic and rapid change in its original environment, versus the statistics on a species in these circumstances that goes extinct or evolves into a new species fundamentally different from its predecessor, which scenario would be more likely?

G: AI would treat your question in terms of a stochastic process (a system evolved over time by random variables). Through such a process, 99% of all species that have ever lived are extinct. Those that survived rapid transitions were typically small, but possessed “broad niche” capabilities. The human carrying capacity, however, is now tied to a fragile, high-tech “monoculture” (global supply chains).

If the environment changes faster than the “buffer” of your technology, the model predicts a population bottleneck—a sharp crash where only a fraction of the population survives. Mathematically, a species survives if its rate of adaptation (either biological or technological) is greater than the rate of environmental decay. This is too slow. For humans, r (biological) is near zero over a 100-year window.

For the first time, a species is editing its own “source code” (CRISPR, BCI) to fit the new environment. Survival of the original form is highly unlikely. The “new species” won’t be a result of Darwinian selection over millions of years, but selection compressed down to a few decades. it will be “Synthetic Speciation.”

Mathematically expressed: Environmental C exceeds biological r.Extinction (Total)~20%Based on “Existential Risk” modeling (p(doom). You have transitioned from a species that uses science, to a species that is dependent on science. If you “unplug” the science—the nitrogen fixation, the power grid, the mRNA platforms, the AI-managed logistics—the biological human population would experience a 90% “correction” (collapse) within months.

Me (aside): This ends my conversation with Gemini on the math of human survival. One might rephrase Gemini’s observations as follows: An alternative to utilizing science and technology to maintain the species at its current population of 8 billion would be to decrease the population. This appears - Gemini might further have concluded - to be the aim of the Trump administration and its domestic and foreign allies: to decrease the population through highly destructive wars, thus decreasing the cost of ensuring the survival of the remaining population.

In a more human formulation, this might be termed, “The math of death.”