2012 - 2019

The Silkworm Project is an installation series consisting of machine objects, experimentation and artifacts that explore the possibilities of hybrid bio-machine worlds that can generate self-organizing silk structures. The series explores possibilities of creating automated production machine systems with the intervention of insects.

The silk machines create a closed feedback loop between the silkworm and the machine to form one fluid eco-system that provides an arena for autonomous interaction between worm and machine. The goal of the series is to develop a system where silkworm input drives the output of the machine in generating silk structures that are not mere cocoons. The result of the spun silk represents both the silkworm and machine’s influence on each other. The series consists of three chapters in three machine artifacts: Machine I: Flat SpinningMachine II: Spatial Spinning, and Machine III: Levitation. The design of these machines begins from a human-centric cartesian perspective and gradually evolve towards a silkworm-centric insect perspective. This transformation reflects the artist's evolving understanding of the silkworm’s spinning and building behavior as she rears and studies silkworms every spring for the duration of the project. By understanding how the worm perceives and measures space with its body while spinning, after trial and error, Machine III: Levitation was able to hack that spatial perception to a small degree in creating self-generated silk sculptures.

In 2022, this piece was selected for the Lumen Prize HUA Awards Longlist. You can find more details about each specific chapter in this series in the below sections.

The Silkworm Project is supported in part by Art Laboratory Berlin in the production of Machine III: Levitation. You can find a video of my presentation on the series along with a panel discussion that was part of the Nonhuman Subjectivities conference held by Art Laboratory Berlin in winter of 2017. Click here to see video. 

You can also find a video about the series produced by Kapeilica Gallery in 2019 in conjunction with my solo exhibition of the series.

Research on a new artist book to further contextualize this series has been supported by the Max Planck Institute for the History of Science. The artist book is currently under development.


Images above are taken by Kapelica Gallery of my solo exhibition

The Silkworm Project in 2019.




While the start of information technology was largely influenced by weaving, for the creation of Machine I, I wanted to reverse this historical cross-over between technologies, and instead have electrical information influence the organization of silk, and have the silk be disperse directly from the silkworm's mouth. In the early design of the machine system, exemplified in this first prototype, my approach was more mathematical and cartesian in its essence, referencing historical and present day artifacts such as the magnetic core memory, and different sericulture environments and equipment. These designs placed the silkworm metaphorically and literally within a human-centered, industry-oriented frame of reference, where the silkworm is a commodity. 

The specific design of Machine I references both the magnetic core memory, a type of computation memory poplar between the 1950s and 1970s, as well as the racks with stacked baskets for used in southeast Aisa for silkworm rearing in the larva stage of the insect. The machine is modularly designed, where the strip of wood at the very front of the machine acts as a removable lock that holds the control circuits in place. This lock can be removed, allowing the plates of circuits to be repaired and replaced if needed. The top the machine has a removable square loop. You can stretch a piece of fabric or silk over the look to provide a surface for the silkworm to spin on. The loop is then placed with the fabric on top of the stimulation electrode matrix, where the worm is stimulated via electric stimulation to move it around within the restraints of the loop. The control circuits below provide individualistic control to the electrodes above. The input for such a system can be twofold, either by humans via control knobs, or via camera capture, utilizing computer vision to capture silkworm movement and trigger electrode stimulation. This prototype is designed like an automatic embroidery machine, where ideally the input and output is both the silkworm. The generated result anticipated is organic and unpredictable.

This experiment was not successful, mainly due to the fact that I failed to consider the behavioral aspects of the spinning behavior of the silkworm. Silkworms tend to climb and wonder when they are ready to spin silk, in which case the environment designed for them does not really hold them. Silkworms also empty the contents of their stomachs as they enter into their spinning phase, which creates dampness for the machine environment, that doesn't bode well for electronics. Finally, the stimulation didn't work, as I couldn't provide a damp contact interface between machine and worm. The dampness is not good for the silkworms nor the machine.


Silkworm Project, Machine I: Flat Spinning




In 2015, Neri Oxman and the Mediated Matter Group at MIT Media Lab published their work Silk Pavilion. With this project, they proposed a new way of looking at digital fabrication through the lens of biomimicry, abandoning the additive method of DIY 3D printers, and directly printing an outward, encompassing structure in 3-dimensional space. This model was based on the silkworm’s spinning behavior, where the team conducted numerous spatial experiments and tests on spinning silkworms in attempt to understand their method and how spatial restraints and variations can affect the final spun outcome. This provided me with a new way of looking at the spinning behavior. I adjusted my approach from designing with the

silkworm to designing for the silkworm. Different from Silk Pavilion in the final goal, I was interested in presenting the spatial perspective of the worm through the machine design, to try and probe into inner workings of the silkworm’s worldview. This became the new design premise for Machine II: Spatial Spinning.

I started conducting my own spatial spinning experiments with silkworms, both looking at how the insects navigate through space individually and collectively. For the collective experiments, I cultivated silkworms that produced multi-colored silk using both the Singaporean method of

feeding colored feed to the worms, and the Japanese approach of

genetically engineered silkworms. Through color tracking methods, I was able to observe the negotiations of two worms spinning in a common space, and how they built upon each other. It was surprising to find that there was very little mistake, back and forth or overlap in collaborative silk spinning, where the spatial territory of worms were clearly marked through the differentiation of silk color, as if it were following a predetermined blueprint.

Further research into areas such as sensory ecology and animal

architecture provided a clearer explanation for what I was observing. In the case of insects that build, as the complexity level of their brain is limited, they instead utilize basic aspects of their body morphology, such as size, shape, and function, almost like a measuring stick to help predict and understand their surrounding. Over thousands of years of evolution, this relationship between form and function is fine-tuned to such a degree, that

the behavioral process that emerges can stand to be simple and

mechanical, while ensuring complex, guaranteed results.


The goal I set for Machine II was to disrupt this equilibrium of insect perception, to introduce confusion, and to track that blindspot of the organisms through spun silk structures. The design of the vertical spinning mechanism starts from the basic form and size of the glass spinning

chamber, each slightly beyond the reach of a fully developed healthy pupa.


The curvature surface of the glass prevents the worm from identify corners and angles, which it utilizes to set up a 3-dimensional framework for its silk construction. The vertical spinning motion of the chamber is based on the silkworm’s sense of gravitational pull, where the slow spinning provides a constant change of gravitational direction, confusing the insect’s spatial orientation. The results, though not always guaranteed, when successful, illustrate the negative space in worm perception, and displays the insect’s process of confusion and adjustment.


Silkworm Project, Machine II: Spatial Spinning, Prototype I

This first prototype was a vertical machine that didn't count for the gravitational influence on silkworm perception. The machine didn't in the end work too well towards hacking their behavior and was abandoned in the end.


Silkworm spinning in machine as vertical machine spun the jar from left to right and back again. Image taken in 2015.


Colored silkworms I raised between 2014-2015. I realized that to study silkworm weaving behavior in a collective, I need to differentiate between the work of different silkworms. Color-tracking, a method borrowed from visual computing, inspired me to seek for ways of rearing silkworms that produced natural colored silk. The image on the left was taken of silkworms I raised using the Singaporean method of feeding colored foods to silkworms during their 4th and 5th instar, in a way dying their silk organs as they developed. This method did not work as well for me as the silkworms were unhealthy and the silk color faint. The image on the right was taken of genetically modified silkworms, first developed by the Japanese. These worms have florescent protein injected into their DNA, which modifies the color of silk they spin.


Silkworm spinning methodology breakdown video, created in 2019 from footage taken of silkworms reared in New York in 2012.

Silkworm Project, Machine II: Spatial Spinning, Prototype 2. Exhibited in Hangzhou.

Experiments of silkworm spinning results in jar.




Machine III: Levitation opens a new chapter in The Silkworm Project series. In this final chapter, I am interested in speculating on a new bionic system logic, where worm and machine form a levitated rotatory ecosystem. A circular capsule that provides a spinning chamber for the worm,

ideally, levitates via artificial magnetic fields controlled and altered by electronic circuits, rotates and spins in arbitrary directions in three dimensional space, breaking the two dimensional spinning logic of previous machines. The current prototype, Machine III-1, supported by Art

Laboratory Berlin, though not yet levitating, is a first step towards this final design. This final installment harkens back to an essential research question that directs my practice: how do we design bio machines? The Silkworm Project answers this question with Levitation, illustrating an

insect / machine utopia, where a balance can be reached between the artificial and the natural,  where fabrication can be digital and biological, where we do not use machines, but they open us up to new perspectives and possibilities.

The current prototype of Machine III: Levitation is a glass capsule suspended in a wooden structure which, driven by magnetics and motors. My ultimate goal it to create a suspended spinning chamber that can spin around on multiple axises. The current prototype has two spinning modes. The first spins slowly continuously in one direction, speed can be adjusted within a certain range. The second spins slowly in one direction for 30 seconds and then stops for 30mins, this cycle continues on and on. Sped can also be adjusted within a certain range.

Silkworm IV - Side 30 Degree - 2.jpg
Silkworm IV - Back.jpg
Silkworm IV - Front.jpg