Evolution has been a billion years drama that has lead to the grandest creation of life: human intelligence. However, we are now not that far from the creation of Artificial life (ALife). The concept of ALife is already being explored in the context of computational creativity (CC) as aspects of life like simulation and synthesis in the context of software, hardware and wetware. 

One remarkable quality of human intelligence which delineates it from the Large language models (LLMs) approach of AI is the human brains capacity to appear complex in parts, because of our competing internal values and goals. Human beings tend to have diverse values and goals that create meaning for them. This in turn creates their reflective or reactive emotions to the goals and that often conflicts each other. These are unavoidable by-product of the levels of abstractions the human brain deals with as a result of values and emotions. On the other hand, machines do not have a sense of humor or conflicting opinions-yet. 

In order to build Alife that is capable of competing the human life in complexity- a grand challenge to explore would be open endedness (OE) as a characteristic trait of life and natural evolution. Open - endedness is a fairly human heuristic and is seen in scenarios of creativity where greatness has been exemplified to be achieved without an objective. The three pillars that will help in addressing the OE are:

The closest we have come to objective-free thinking in terms of machine theory is the novelty search paradigm, a search process developed for evolutionary robotics . Novelty search differentiates from random search,as a principled search strategy that orders the search space from the simple to the complex.  However,  all living systems exhibit various forms of cognitive capacities. Therefore, together with the attempts of constructing generative models of cognitive systems,  it's equally important to explore generative models of living systems, as insightful and important ALife targets.

ALife has long utilized hardware approaches, exemplified by robotics. Robots embody artificial systems, engaging directly with physical environments. Unlike software simulations constrained by predefined rules, robots must contend with real-world forces like gravity, energy constraints, and material limitations. Yet, even these interactions are bounded by human design, as sensor input and material properties are selected by creators. Robotics highlights the challenges of creating adaptive systems and remain restricted in applications. Optimus and other robotic endeavors may change that. 

Software ALife models rely on algorithms to simulate life-like behaviors within virtual environments. While software enables experimentation with complex systems, it inherently lacks the unpredictability of physical embodiment. Simulations generate results confined to the designer’s input parameters, making truly novel discoveries rare. This deterministic nature of software underscores a critical limitation: though it mirrors biological phenomena, it fails to replicate the open-ended adaptability and meaning-making intrinsic to life itself.

Wetware, emerging through synthetic biology or even synthetic chemistry, represents ALifes most profound contribution. Wetware models, being physically embodied and thermodynamically constrained, can engage in unbounded free-energy-driven interactions with the environment. This "informational openness" allows them to adapt autonomously to perturbations, potentially overcoming fundamental issues in AI like the symbol-grounding problem. Synthetic biology and systems chemistry are at the forefront of wetware ALife, constructing artificial cells and protocells that serve as simplified models of living systems.

Examples from synthetic chemistry and biology illustrate the power of wetware ALife:

The convergence of synthetic biology, chemistry, robotics, and software in ALife research has led to beginning of innovation. The development of "Xenobots" or Neuralink's brain-computer interfaces exemplify the integration of biological systems with advanced robotics and software.  We are at the beginning of an era of transformative impact of interdisciplinary collaboration in ALife, paving the way for a deeper philosophical question: Are computers thinking or just calculating? Conversel, are humans thinking or just calculating?

I therefore consequencially ask: Are creativity and OE entirely different concepts? or do their synthesis and measurement represent closely related research endeavors?

And if so, how are they connected?

Picture from: Chemical Systems for Wetware Artificial Life: Selected Perspectives in Synthetic Cell Research

Pri

https://world.hey.com/priyata

Ci
tations:

Luisi PL, Ferri F, Stano P. Approaches to semi-synthetic minimal cells: a review. Naturwissenschaften. 2006 Jan;93(1):1-13. doi: 10.1007/s00114-005-0056-z. PMID: 16292523.

Noireaux V, Libchaber A. A vesicle bioreactor as a step toward an artificial cell assembly. Proc Natl Acad Sci U S A. 2004 Dec 21;101(51):17669-74. doi: 10.1073/pnas.0408236101. Epub 2004 Dec 10. PMID: 15591347; PMCID: PMC539773.

Glass JI, Merryman C, Wise KS, Hutchison CA 3rd, Smith HO. Minimal Cells-Real and Imagined. Cold Spring Harb Perspect Biol. 2017 Dec 1;9(12):a023861. doi: 10.1101/cshperspect.a023861. PMID: 28348033; PMCID: PMC5710109.

Woo, S., Saka, S.K., Xuan, F. et al. Molecular robotic agents that survey molecular landscapes for information retrieval. Nat Commun 15, 3293 (2024). https://doi.org/10.1038/s41467-024-46978-2