Humans and AI: Balancing Mutualism and Parasitism

The concept of evolution is key to understanding both life and technology. Biological evolution, a cornerstone of natural sciences, involves the gradual development of organisms over millions of years through natural selection. In contrast, cyber evolution refers to the rapid advancement of technology and AI driven by human ingenuity and market forces. Despite differences in mechanisms and timescales, striking parallels between these evolutions provide valuable insights into their development.

One fundamental parallel is incremental progress and adaptation. Biological evolution involves small genetic changes over vast timescales, leading to species adaptation. Similarly, cyber evolution is marked by incremental technological advancements. Innovations build on previous developments, resulting in gradual improvements. Modern smartphones, for example, are the product of decades of incremental advancements and numerous design failures.

Adaptation is key in both biological and cyber evolution. In biology, organisms adapt to pressures like climate and food availability, developing survival-enhancing traits. Similarly, technologies and AI systems must adapt to their environments. AI/ML algorithms, for example, improve through iterative learning and are replaced if they fail. User needs, ecosystems, and regulations shape new solutions. For example, the Internet and mobile computing have driven the evolution of connectivity, data processing algorithms, and cybersecurity. While failed technologies don't feel pain, their creators and investors do. Think, for example, of WebCrawler (1994), Lycos (1994), Infoseek (1994), AltaVista (1995), Excite (1995), and Ask Jeeves (1996) — all of which were once prominent search engines and businesses but eventually faded away after losing to Google and a few others in the competitive market.

Both biological and cyber evolution create new forms through change and innovation. In biological evolution, new species emerge via speciation, driven by genetic divergence and environmental pressures, like the Galapagos finches. This adds to Earth's biodiversity. Similarly, cyber evolution produces new technologies and products by combining existing ones, such as the Internet of Things (IoT) from integrating internet connectivity, sensors, and computing. This mirrors the diversification in biological evolution, highlighting the creative potential of both.

Despite similarities, significant differences exist between biological and cyber evolution, particularly in timescale and intent. Biological evolution occurs over millions of years, with genetic changes and natural selection spanning geological timescales. In contrast, cyber evolution happens much faster, with technological advancements occurring within years or even months. Rapid progress in fields like computing, telecommunications, and AI leads to significant changes within a human lifetime, allowing swift adaptation and new technologies at an unprecedented rate. It is humans who struggle to adapt to these rapid changes.

Another key difference is in the guidance and intent behind each form of evolution. Biological evolution is an unguided natural process, while cyber evolution is directed by human intent and design. Engineers and innovators deliberately create and refine technologies to achieve specific goals, unlike the natural process of biological evolution. At the end, however, it is the often-unpredictable market (composed of humans) that decides which technology lives and which dies.

And now let’s talk about the most important consequence of the aforementioned parallelism: the entanglement of human and cyber evolution. This forms a complex symbiotic relationship where each influences the other profoundly.

As AI and technology advance rapidly, they integrate deeply into human life, enhancing capabilities, extending lifespans, and transforming societal structures. Humans use AI and other cyber technologies to develop medicines, manipulate genes, control implants, combat diseases, and more, pushing biological boundaries. Cyber technologies evolve rapidly, shaped by human needs, ethics, and regulations. This dynamic interplay accelerates progress in both domains, creating a feedback loop where technological advancements drive biological enhancements and vice versa. As this symbiosis deepens, the line between human and machine will blur, suggesting a future where the two may evolve together as a singular intertwined entity.

However, any symbiotic relationship has the potential to be either mutualistic, where both benefit; commensal, where one benefits and the other is neither helped nor harmed; or parasitic, where one benefits at the expense of the other. The same principle applies to symbiotic relationships between humans and AI. Like mistletoe plants that attach to trees and shrubs, extracting water and nutrients from their host and potentially weakening it, AI can similarly impact humans.

There is one more type of symbiotic relationship: competition. An example of such competition is the relationship between coral and sponges, which compete for the same resources, but sponges will not survive if the coral dies. This scenario doesn't seem applicable here, as both humans and AI can likely survive alone.

I am sure that humans have always viewed this relationship as commensal, where we gain all the benefits while continuing to feed the cyber revolution solely for our own advantage. Even if this were not true, we would be fine with having mutualistic relationships. And we never imagined the relationship would turn into competition. ?Regardless of the intentions, it is crucial to acknowledge the realities and potential risks that a parasitic relationship with AI can present.

Indeed, symptoms of a parasitic relationship are emerging in our interaction with AI and cyber technologies. Job losses, threats to intellectual work, increased surveillance, loss of privacy, intellectual property issues, and ethical concerns like algorithmic bias are real costs we are already paying. Yet, the numerous benefits continue to indicate a strong mutualistic relationship as we persist in investing heavily in AI and cyber development. However, the long-term trajectory of this relationship remains uncertain, particularly with the potential emergence of AGI (Artificial General Intelligence) or even a highly advanced AI.

To balance this relationship, proactive measures must be taken to ensure mutual benefits. Investing in better STEM education, targeted regulations, and clear ethical standards can align AI developments with human values and societal goals, leading to a harmonious coexistence. We must continually reassess our interactions to ensure they remain mutualistic, keeping in mind that we evolve at very different paces, which puts us at a disadvantage. It's crucial to respond appropriately if harmful trends are detected.

Above all, we must avoid creating a competitor or, worse, a parasite that could one day destroy our society, particularly for the fleeting glory of outcompeting another company or being the first to achieve a significant milestone in AI.

We don’t just need "responsible AI," as many discuss, but also responsible AI developers.

A Pyrrhic victory is the last thing we need.