RF Safe announced an “Ethical Connectivity Pledge” aimed at celebrity- and creator-branded mobile plans, urging them to adopt child-first design standards, improve transparency, and invest in lower-exposure connectivity options such as Li‑Fi where feasible. The organization argues that current microwave-based wireless networks may pose plausible health risks—especially for children—and that business models can externalize long-term health costs onto families and public systems. The pledge is presented as a public signatory framework with tiers of commitment and an intent to enable public scrutiny of follow-through.

RF Safe argues that animal evidence for RF-related carcinogenicity is now strong and should not be dismissed, citing the NTP (2018) and Ramazzini (2018) lifetime rodent studies as showing statistically significant increases in the same rare tumor types (heart schwannomas and brain gliomas). The post further claims that effects occurred at relatively low whole-body SAR levels and references additional mechanistic hypotheses (e.g., VGCC-related models and radical-pair/spin effects) and a reported human ultrasound observation of acute non-thermal changes. These points are presented as supporting a shift away from a “thermal-only” interpretation, but the item is advocacy/commentary and does not provide full methodological details in the excerpt.

RF Safe presents the “S4-Mito-Spin” framework as a hypothesis aiming to unify proposed non-thermal biological effects reported in some EMF studies (e.g., oxidative stress, DNA damage, fertility effects, and tumors in animal models). The article describes a multi-mechanism model involving voltage-gated channel forced oscillation, mitochondrial/NOX amplification to reactive oxygen species bursts, and radical-pair/spin-state effects, with a novel “density-gated” concept to explain tissue-specific and inconsistent findings. It also suggests the framework could connect EMF hazards with therapeutic uses, citing FDA-approved RF devices such as TheraBionic as an example of RF modulation of biology.

This dosimetric study evaluated whether existing EM safety guidelines protect individuals with conductive implants by assessing implant-related local field enhancements. Across 10 kHz to 1 GHz, the authors report large increases in psSAR10mg and local electric fields near implants, particularly below 100 MHz. In human anatomical models with implants exposed to an 85 kHz wireless power transfer coil and a 450 MHz dipole, the study reports guideline exceedances and elevated psSAR10mg, while the modeled temperature rise at 450 MHz remained under 0.4 K after six minutes. The authors conclude current guidelines are insufficient for people with implants and propose regulatory changes.

This exposure-assessment study evaluated magnetic-field and contact-current exposures from modern induction hobs using IEC-based measurement procedures, 3D field scanning, and numerical dosimetry in anatomical models. It reports large between-hob variability in exposure and states that IEC 62233 may substantially underestimate user exposure. The authors argue that design modifications can reduce exposure and that product standards should be revised to better reflect realistic user scenarios.

This conference paper uses numerical simulations to evaluate near-field exposure and thermal effects in a detailed human head model from a realistic 5G mobile phone operating at 26 GHz. The preliminary modeling suggests moderate, localized temperature increases in superficial tissues. The authors emphasize the need for higher-resolution models, refined tissue segmentation, longer exposure durations, and varied phone placements to better characterize potential impacts.

This study uses anatomically detailed computational models of a five-year-old girl, a pregnant woman in the third trimester, and a fetus to simulate mobile phone RF exposure inside an elevator cabin. Simulations at 1000 MHz and 1800 MHz across 48 configurations evaluated SAR10g, whole-body SAR, and maximum temperature. The abstract reports that configuration (positioning and phone orientation) can substantially change absorption and temperature metrics and calls for broader scenario testing to inform safety guidance for vulnerable populations.

This dosimetry study used FDTD simulations at 28 GHz to evaluate how skin stratification and anthropomorphic modeling affect absorbed power density (APD) estimates. APD was higher with stratified skin than with homogeneous skin for a wearable patch antenna (16%–30% higher), while plane-wave differences were smaller (<11%). The authors argue that simplified skin models may underestimate exposure in mmWave wearable scenarios.

This systematic review synthesizes animal studies on low frequency magnetic field exposure in relation to neurodegenerative diseases. It reports no support for a causal induction of Alzheimer’s-type neuropathology in naive animals, while noting that evidence is too limited to draw strong conclusions for motor neuron disease, multiple sclerosis, and Parkinson’s disease regarding induced neuropathology. In models with pre-existing neurodegenerative disease, the review describes possible therapeutic effects on behavioral and neuroanatomical outcomes for dementia-related conditions, and no apparent effect on motor neuron disease progression.

This study tested magnetic fields from tablets, laptops, smartphones, and household/leisure magnets against 13 cardiac implantable electronic device (CIED) models to assess magnet mode activation. It reports that these consumer devices can trigger magnet mode when in close proximity, with median activation distances of 5–6.5 mm for phones/tablets/laptops and mainly contact-level activation for household/leisure magnets. None of the tested devices activated magnet mode at distances of 20 mm or more, and the authors emphasize patient awareness of proximity-related risk.

This occupational epidemiology study used machine learning to evaluate whether workplace exposures (including magnetic and electric fields, vibration, noise, and heat stress) predict male reproductive indicators in 80 workers. The models and explainable AI outputs highlighted magnetic and electric field exposures and age as key predictors linked to lower free testosterone. The authors also report a 10-year forecast identifying electric field exposure as the most important long-term risk factor.

This in vitro study exposed glioblastoma (CT2A), neuroblastoma (N2A), and non-tumor astrocyte (C8D1A) cell models to a 100 μT magnetic field across 20–100 Hz for 24–72 hours. The abstract reports decreased viability and proliferation in the tumor cell models within a frequency window centered at 50 Hz, while astrocyte viability increased at 20 and 40 Hz. The authors conclude that frequency is a key determinant of cell-type-specific responses consistent with a “biological window” model.

This analytical study evaluated machine learning models (SVM and Random Forest) to predict health symptoms in adults living near mobile phone base stations. The SVM model reportedly achieved high predictive performance for headache, sleep disturbance, dizziness, vertigo, and fatigue, and outperformed Random Forest and prior models. The abstract concludes that proximity to base stations is connected with increased prevalence of several symptoms and emphasizes distance, age, and duration of residence as key predictors.

This exposure assessment measured 20 kHz magnetic field leakage from induction cooking heaters across four models and compared results with ICNIRP general public limits. The maximum reported magnetic flux density was 16 µT at a specified measurement point using two S-type pans. Field leakage depended on pan size and configuration, and finite element modeling was reported to align closely with measurements.