A robotics engineer’s bedroom experiment may have just produced the most sophisticated mosquito-elimination system in history and its implications for West Africa, where malaria kills hundreds of thousands every year, could be nothing short of transformative.
By isqil Najim
Steven Cheng, an engineer specialising in computer vision and robotics, based in Changzhou, China, spent four months constructing what he calls the ultimate mosquito killer an AI-guided laser system capable of spotting, tracking, and eliminating a mosquito in roughly three milliseconds. He unveiled the project on social media in late May 2026, and it has since captured the attention of both the engineering and public health communities.
Inside the Machine
Cheng built a custom image dataset by pairing a DSLR camera with a high-magnification zoom lens and photographing mosquitoes at close range a process he said left him covered in bites. Those annotated images trained a deep learning model that, after putting his graphics card through extended strain, reached detection accuracy.
Once the AI identifies a target, a precision laser mounted on an industrial rotary stage and gimbal takes over, redirecting itself in real time based on continuous data from the model. It is a closed-loop system: detect, confirm, redirect, eliminate.
A second version, released shortly after the first, adds thermal imaging to the sensor array and upgrades the hardware with harmonic drives, servo motors, and a reinforced aluminium gimbal all of which improve tracking speed and mechanical stability.
A laser strong enough to vaporise an insect in milliseconds presents obvious risks to eyes and flammable materials. Cheng addressed this with a dedicated wide-angle camera running in parallel, continuously scanning the environment for humans, pets, and fire hazards. If any are detected within the targeting zone, the laser is immediately disabled.
After running the system overnight in his own home, Cheng reported the mosquito population had been entirely cleared by morning.
A Growing Field
Cheng’s system is not alone. A commercial product called Photon Matrix developed by Jim Wong, also from Changzhou uses LiDAR sensor technology to detect and eliminate mosquitoes at up to 30 per second within a six-metre radius, and is currently progressing from crowdfunding toward production. Earlier still, Intellectual Ventures developed the Photonic Fence, a perimeter-based laser system conceived explicitly for disease control in developing regions and projected to cost as little as $50 per unit to manufacture. None of these have yet reached the communities that need them most.
What sets Cheng’s prototype apart is its accessibility: built from consumer-grade components using open-source computer vision tools, it demonstrates that this class of technology no longer requires a corporate research budget or laboratory setting.
What This Means for Africa and Why West Africa Cannot Wait
The timing of this breakthrough matters enormously on this side of the world. According to the WHO’s World Malaria Report 2025, Africa recorded an estimated 265 million malaria cases and 579,000 deaths in 2024 representing 95% of the global disease burden. Three in four of those deaths were children under the age of five. West Africa sits at the centre of this catastrophe. Nigeria alone accounted for 24.3% of all global malaria cases and 30.3% of all malaria deaths in 2024, with an estimated 184,933 fatalities in a single year.
Existing control strategies insecticide-treated bed nets, indoor residual spraying, preventive medication remain valuable but insufficient. Critically, partial resistance to artemisinin, the foundation of modern malaria treatment, has now been confirmed or suspected in at least eight African countries, per the same WHO report. The toolbox is shrinking precisely when it needs to grow.
AI-guided laser systems offer something qualitatively different: targeted, chemical-free, resistance-proof elimination. They do not rely on behaviour change or consistent correct usage. They do not contaminate water, soil, or food. They actively pursue individual insects rather than waiting passively for contact. For communities where mosquito populations are dense and persistent, that active posture is a significant departure from anything currently deployed.
Practical Barriers Are Real
Translation from a private home in China to a rural clinic in Ghana, a school in Mali, or a household in the Niger Delta is not automatic. Reliable electricity is a foundational requirement for AI inference and precision laser hardware and remains inconsistent across much of West Africa. Current commercial analogues are also priced for consumer markets in high-income countries, not community health budgets. And at individual-device scale, the impact is limited; for population-level effect on malaria transmission, the technology would need coordinated, wide-area deployment across homes, schools, and health facilities simultaneously.
These are solvable problems, but they require international public health investment, not just individual innovation.
A Proof of Concept That Belongs in the Right Hands
What Cheng’s project ultimately demonstrates is that the technical barrier has been crossed. An individual engineer with consumer hardware and four months of effort built a functioning, AI-driven mosquito elimination system. The cost curve for such hardware has historically dropped sharply with scale and time.
If West African governments, global health institutions, and technology partners treat this not as a curiosity but as a viable complement to existing interventions alongside the expansion of rural electricity infrastructure laser-guided mosquito control could become part of the region’s public health arsenal within this decade.
Malaria has devastated West Africa for generations. The technology to fight it differently now exists in a bedroom prototype. Getting it out of the lab and into the communities that need it most is the challenge that remains.
Sources: TechSpot, Interesting Engineering, Tom’s Hardware, WHO World Malaria Report 2025, Severe Malaria Observatory (Nigeria Data), Statbase Global Health Dataset.