A robotics and computer-vision engineer in Cambridge spent the spring of 2026 turning his living room into a working mosquito kill box: a DSLR, a thermal camera, a deep-learning detector, an aluminium gimbal, and a laser that shoots mosquitoes out of the air. The engineering is real, and the safety camera that cuts the beam near people is the cleverest part. The honest conclusion is that for almost every home the answer is still the boring one, not the laser.
On 28 May, a computer-vision and robotics engineer in Cambridge posted a video to X of a machine he had built on his living-room floor. It is the size of a small printer and rides on rollers. Bolted to it are a DSLR with a long zoom lens, a thermal camera, an aluminium gimbal, and a laser. In the video the gimbal swivels, the laser locks on, and a mosquito drops out of the air. A deep-learning model, trained on annotated mosquito images, does the seeing. A second wide-angle camera does the worrying: the moment a person, a pet, or anything flammable crosses the firing arc, it cuts the beam.
Steven Cheng calls it version 2.0 of his "AI laser mosquito defense system". It is real, it is genuinely clever, and for almost every household it is the wrong answer to a question nobody was asking. That gap is the story worth telling.
What the machine actually does
Strip away the drama and the build is a three-step chain. The DSLR scans the room and feeds frames to a model trained to recognise the silhouette and gait of a mosquito. When it flags a candidate, the thermal camera confirms a warm body against the background. The gimbal aims. The laser fires. Version 2.0, the one Cheng announced on 28 May, added harmonic drives, servo motors, and a reinforced aluminium mount, the parts that make the aim steady enough to catch an insect in flight.
The safety layer is the part most write-ups skip, and it is the most interesting engineering in the whole rig. A wide-angle camera watches the room, and a logic gate refuses to let the laser fire whenever its path crosses a person, an animal, or a flammable surface. Cheng is blunt that the system is not for sale and not certified for eye safety. He runs it in his own home, at his own risk, with that camera as his only guarantee.
None of the idea is new. Intellectual Ventures floated a "laser mosquito" concept in the late 2000s, and versions of it have sat on university drawing boards for more than a decade. What Cheng added is not the theory. It is the nerve to build the thing on his carpet and film it running.
The tell is the certificate, not the kill
Look at the nearest thing to a product. A device called Photonmatrix advertises up to 30 mosquitoes killed per second inside a 6-metre radius and sells from around $498 on Indiegogo. It also, by the account in the French technology press, cannot be certified for eye safety, the same wall Cheng hits in his living room. That is the quiet lesson of the whole category. The hard part was never getting a laser to hit a mosquito. The hard part is proving it will never hit a child or a cat.
The results, so far, are a home movie. Cheng reports that every mosquito in his home was "successfully eliminated" after a single night. There is no count, no control group, and no independent verification. Even the detection model, the cleverest part, is described in the coverage only as "rather good". No benchmark has been published.
Cheng is one dot on a much larger map
The reason to read past the video is that Cheng is the amateur edition of a movement that is also running in funded labs. Sterile-male releases built on the Wolbachia technique are scaling across Europe and the Americas, with the largest proposed US releases now under EPA review. Italy logged 133 imported dengue cases in early 2026; Valencia's sterile-insect trial has cut local tiger-mosquito numbers. Gene-drive programmes in Tanzania are advancing under African scientific leadership. XPRIZE-style vector-control funding, which rewards cheap hardware and clever software, has paid for everything from acoustic mosquito sensors to AI species-identification apps. In the field, AI already counts and identifies mosquitoes in trap catches faster than any human could.
The pattern is the same everywhere: a sensor, a model, a software loop, wrapped around an old problem. Cheng's laser is simply the most photogenic entry on a long list.
Three things the prototype teaches
The software is no longer the bottleneck. A model that runs in real time and clears a room in a night has crossed a practical threshold, and the payoff is not lasers. It is surveillance. A trap count that once needed a trained entomologist and a microscope can now run on a Raspberry Pi with a camera and a small model. That is the unglamorous plumbing of public health, and it is where this technology genuinely scales.
The killing hardware is the hard part, and it always will be. Anything that fires in a family home has to be safe, certified, and reliable, which is a multi-year regulatory road, not a weekend project. The history is a graveyard. The bug zapper never worked. The electric racket is a battery fire waiting to happen. The plug-in ultrasonic gadget is a placebo. Software has raced ahead; hardware in this category has not.
Your living room is the wrong unit of measurement. A laser can clear one room. It cannot clear the garden, the courtyard, the park, or the schoolyard. Mosquito control is a numbers problem: you have to suppress the local Aedes population, not pick off the few that drift indoors. The question Mosticare keeps returning to is not "what kills the mosquito in front of me" but "what stops the next generation reaching me at all". Cheng's laser answers the first question beautifully. It will never answer the second.
The boring defence keeps winning
Follow the public-health advice and it lands, every time, in the same unglamorous place. Take away the standing water: tip the saucers under the flowerpots, screen the gutters, empty anything that holds rain. Then put a physical barrier over the two places that matter, where you sleep and where you sit still. This is what French regional health agencies tell residents in the départements marked as "colonised zones" for the tiger mosquito. It is the thread running through Italy's ISS season outlook, the Cyprus "new reality" framing from the ECDC, and the World Health Organization's 2026 insecticide-resistance update. It costs nothing like a £1,500 gimbal, it needs no model, and it works whether or not the next experimental control ever clears a regulator.
That is not a knock on Cheng. His build is exactly what a curious engineer with a thermal camera, a long lens, and a redundant safety chain should be doing with a spring. The trouble is the framing that follows it around. Coverage that calls a living-room laser "the future of mosquito control" quietly undersells the cheap, proven habits that already protect almost everyone.
What to watch next
Three signals will tell you where this goes. The first is a published benchmark: a confusion matrix on a public dataset would turn "rather good" into something the field can trust. The second is regulation, specifically whether any national safety body certifies a Photonmatrix-style laser, which is the line between a hobbyist toy and a real product class. The third is the deployment numbers on the methods that already work at scale: Wolbachia in Singapore, sterile-male releases in Valencia, gene drives in Tanzania, the Italian arbovirus season. None of them come with an AI-gimbal video. All of them are bending the curve.
A laser that clears your living room is a wonderful piece of work. A window screen that keeps the tiger mosquito out in the first place is cheaper, safer, and, framed right, not much less impressive.
What we know
Sources cited
- Futura-Sciences, "Moustiques tigres : cet ingénieur a inventé une arme radicale, un canon laser piloté par IA" (Patrick Ruiz, 2 June 2026). https://www.futura-sciences.com/tech/actualites/technologie-moustiques-tigres-cet-ingenieur-invente-arme-radicale-canon-laser-pilote-ia-134959/
- Numerama, "Il fabrique un canon laser piloté par l'IA pour carboniser les moustiques de son salon." https://www.numerama.com/tech/2264839-il-fabrique-un-canon-laser-pilote-par-lia-pour-carboniser-les-moustiques-de-son-salon.html
- Les Numériques, "Un ingénieur crée le tueur de moustiques ultime à base de laser et d'IA" (Patrick Ruiz, 2 June 2026, 12:36). https://www.lesnumeriques.com/societe-numerique/un-ingenieur-cree-le-tueur-de-moustiques-ultime-a-base-de-laser-et-d-ia-n256912.html
- Developpez.com, "Un ingénieur met au point un système de défense laser à base d'IA pour éliminer tous les moustiques d'un domicile." https://intelligence-artificielle.developpez.com/actu/383750/Un-ingenieur-met-au-point-un-systeme-de-defense-laser-a-base-d-IA-pour-eliminer-tous-les-moustiques-d-un-domicile-IA-ingredient-incontournable-pour-deboucher-sur-le-tueur-de-moustiques-ultime/