How to reduce EMF exposure from your phone

Two different things get called EMF. Ionizing radiation, the X-ray kind, can break chemical bonds. A phone does not emit that. A phone, a Wi-Fi router, and a Bluetooth earbud emit radiofrequency fields. At high enough power those fields heat tissue. Exposure limits from the International Commission on Non-Ionizing Radiation Protection are built to stay under that heating. Below it, the same body of research has one finding that keeps showing up: a small change in brain electrical activity on EEG. ICNIRP’s mobile-phone statement says that change has not been tied to worse sleep or to another adverse effect, and that phone exposure under the heating threshold is unlikely to cause harm. The World Health Organization’s February 2025 note says the low levels typical of the environment have not been shown to be a health risk, and that higher levels are what the standards are for.

That is the agency position. It is not a proof that every study is empty. In 2011 the International Agency for Research on Cancer put radiofrequency fields in Group 2B, possibly carcinogenic, on limited human evidence about certain brain tumors in heavy long-term users. A set of systematic reviews meant to feed a new WHO monograph was still being argued over in 2025, including an animal-cancer review whose method the journal’s own introduction said had left the protocol. Treat 2B as a label for limited evidence. It is not a finding that a handset gives you cancer, and it is not a reason to ignore a transmitter you keep on your skin for years.

Where the dose actually is

A tower down the street is a weak field by the time it reaches you. The phone at your ear is the strong one, because the antenna is centimeters from tissue and the handset raises its own power when the signal is poor. Specific absorption rate, the energy absorbed per kilogram, is the number regulators use for that close exposure. Moving the handset off the body drops it faster than any argument about 5G as a mood. Speaker mode, a wired headset, and a call you do not make from a basement stairwell are the same move: less power, farther from the head.

Sleep is the long dose. A phone transmitting under a pillow, and a router on the nightstand aimed at the bed, are hours of close exposure you can end without giving up the network. Airplane mode, or the phone in another room, removes the transmitter. A case or a sticker that ruins the signal does the opposite. The handset compensates by transmitting harder.

Routers are the same geometry. One router for the dwelling, placed out of the bedroom, beats a second extender beside the mattress. The name of the generation, 4G or 5G or Wi-Fi 6, is not the dose. Power, distance, and hours are the dose.

What Aires Tech claims, and what the papers are

Aires sells a passive resonator. The company says it does not block, absorb, or reduce signal strength. It says the device changes the coherence of the field. The homepage then talks as if the field had been handled the way a filter handles water: 50-plus studies, 25-plus peer-reviewed papers, 20-plus patents, about $20 million in research, almost a million users, and customers who report that children are safe and that headaches stopped. Those are different claims. A patent is not a health result. A customer sentence is not a trial.

In September 2019, before a Canadian listing, an underwriter commissioned a review. Aires published its own account of that review in 2026. The quoted conclusion is that the technology changed applied radio energy in measurements, affected some physical parameters of water, showed a protective effect on rat cells, and showed very limited evidence of protection in humans. The reviewer asked for randomized, sham-controlled, double-blind human trials, and said the best cardiovascular protocol should be repeated on a group large enough for statistics and then published in a health journal.

The human work they put first now is not that journal paper. A 2024 report from the Kirov Military Medical Academy, as summarized on the Aires site, used 24 healthy men aged 20 to 24, a Prestigio Android phone on GSM, EEG and ECG, and the Lifetune One. The writeup says the phone shifted brain activity toward activation, and that the resonator was followed by a return toward baseline. A 2025 summary under the names Sysoev and Rybina is described as 14 men on the clinical index and as 24 subjects in the archive headline. Both live on the company site. I did not find either as an independent medical-journal paper.

The Trent University material is two case studies by Magda Havas from 2015. One subject was 52 and identified as electrically hypersensitive. The other was 46. A cordless-phone base raised the room to about 120 milliwatts per square meter. With the Aires Defender Infinity, some pulse-wave and blood-pressure readings moved; with a placebo unit they moved back. Live blood and, in the second case, a GDV camera were part of the session. The writeup also says the shield did not change the radiofrequency level in the room, nor the low-frequency electric and magnetic fields. “FDA Class II” on the MaxPulse is a classification of that measuring gadget. It is not an FDA trial of the resonator. One blinded subject is a protocol sketch. It is not a population result. Provocation studies of people who report electromagnetic intolerance, reviewed as a set, have not produced a reliable exposure effect. The better-controlled ones point to expectation as the stronger driver of symptoms.

The study that is a real paper is Dyuzhikova and colleagues, Ecological Genetics, 2019. Male rats, a Linksys router at 2.4 gigahertz and 16.5 dBm, about 40 centimeters away, six hours a day for four days, inside a Faraday cage. In Wistar rats the rate of chromosome disturbances in bone-marrow cells rose about threefold to fourfold against controls. Six Aires Defender Pro units, fixed in the center of each face of the cage, cut that rate, and the animals were still above the controls. A high-excitability strain looked more sensitive than a low-excitability strain. Igor Serov of the Aires Human Genome Foundation is an author. The device in that experiment is not a pendant on a person, and the animal is not a person. A related 2018 engineering-journal paper includes an Aires engineer among the authors.

Heart-rate papers they list are on the order of 13 to 15 subjects. The peer review they cite for one of them is PACE, the Planetary Association for Clean Energy. The president of that association is named on Aires’s own endorsement block. That is not the same object as a clinical trial in a medical journal. The UFC material is a testimonial from Duncan French of the Performance Institute, plus demonstration EEGs. No methods paper of that athlete series turned up in the search.

The clinical index says the human studies use double-blind, placebo-controlled, or triple-blind protocols. The word “or” does a lot of work. A one-person case study can be blinded and still be one person. The physics papers, including a Springer conference simulation and thermal images of the resonator surface, describe the object. They do not show a disease avoided in a human being.

What is worth doing

Cut the close, long exposures. Phone off the body for calls. No transmitter under the pillow. Router out of the bedroom. Poor signal is a reason to wait or to move, because the handset transmits harder there. Those steps reduce absorbed power. They are the whole of the practical advice that matches the physics.

If a product does not reduce that power, it is not doing this job. Aires says it does not reduce it, and the Havas measurement they publish agrees. Whether a coherence change matters to health is the claim. The record they offer is a rat cage lined with six company resonators, a handful of young men in company-posted EEG reports, and two single-subject sessions from 2015. That record is too small for the sentence on the homepage.