RF Safe’s Proposed Federal Research and Policy Agenda for Electromagnetic Exposure
Submitted by John Coates, Founder of RF Safe
Docket HHS-OASH-2026-0397
Responses to Questions 2–18
Executive position
HHS should establish a sustained, independently governed research program that evaluates electromagnetic exposure through three connected dimensions: the physical exposure, the biological receiver, and the receiver’s capacity to recover.
Existing exposure metrics remain essential. The question is whether they adequately predict the outcomes that matter under chronic, intermittent, developmentally timed, and combined exposures.
RF Safe proposes a specific organizing question:
Can a defined electromagnetic exposure alter biological regulation or recovery in ways that change an organism’s response to a subsequent challenge—even after the immediate exposure has ended?
This question connects oxidative stress, calcium signaling, mitochondrial function, tissue susceptibility, and long-term outcomes without assuming that they constitute a single established causal pathway.
HHS should pursue five immediate actions:
- Publish an evidence assessment organized by health outcome and exposure conditions.
- Fund independent replication of consequential animal, cellular, and human findings.
- Test whether biological state and exposure history improve prediction beyond physical dose alone.
- Evaluate practical exposure-reduction technologies using measured performance.
- Publish a research roadmap, responsible agencies, milestones, and criteria for revising recommendations.
A complete mechanism should not be a prerequisite for investigating credible hazard evidence. Equally, a plausible mechanism should not substitute for demonstrating an adverse outcome.
Question 2 — Perspective and basis of this submission
I submit these comments as RF Safe’s founder, a parent, and an engineering and public-health advocate.
RF Safe’s advocacy began in the 1990s. Its work has included openly shared exposure-reduction designs, research education, and development toward optical communications. This advocacy is distinct from separate ownership and commercial use of the RF Safe trademark. Manufacture or sale of products by other entities should not be attributed automatically to the advocacy effort.
This submission draws on published experiments, evidence reviews, federal research, medical-device records, and RF Safe’s proposed biological-fidelity framework. The framework supplies hypotheses and experimental designs; it is not presented as independently validated evidence.
Question 3 — Public experiences and reported effects
Public reports should inform the questions HHS investigates.
A useful reporting system should preserve exposure timing, symptom timing, changes in conditions, functional consequences, and relevant clinical assessment while protecting identifying information. It should distinguish observations from explanations of their cause.
Reports may identify overlooked exposure settings or outcomes that conventional studies have not measured. Their evidentiary value increases when they guide independently testable predictions.
HHS should provide a pathway from reported experience to structured investigation, including consideration of alternative explanations. Respectful evaluation and rigorous causal assessment should operate together.
Question 4 — Research observations that warrant investigation
Three recurring observations deserve coordinated study.
First, oxidative and cellular stress responses appear across experimental literature. Yakymenko and colleagues reported oxidative effects in 93 of 100 studies included in their review. Lai and Levitt subsequently described EMF-associated findings within the broader biology of cellular stress responses. These syntheses identify research priorities; the fraction of positive publications does not establish an effect size or population risk.
Second, chronic animal findings warrant investigation of susceptible cell lineages. NTP reported clear evidence of carcinogenic activity based on malignant heart schwannomas in male rats under its experimental conditions. Research should investigate the affected Schwann cells and their local environment rather than infer susceptibility from characteristics of the surrounding organ.
Third, response may depend on biological state. A randomized study in CACNA1C-genotyped volunteers reported genotype-dependent sleep-EEG findings following 3.6-GHz exposure. Independent replication should establish the reliability, magnitude, and functional significance of that observation.
These observations support a program examining conditional responsiveness: what responds, under which exposure conditions, for how long, and with what consequence?
Question 5 — Standards and exposure classification
HHS should assess the biological coverage of existing standards explicitly.
For each important outcome, the assessment should identify:
- The evidence supporting protection.
- The populations and exposure conditions represented.
- Contrary findings and their methodological strengths and limitations.
- Remaining uncertainties.
- Findings that would trigger reconsideration.
Retain protections against established heating and stimulation effects while evaluating whether additional safeguards are justified for chronic or susceptibility-dependent outcomes.
Exposure classification should preserve frequency, intensity, internal dose, modulation, pulse timing, duty cycle, duration, geometry, and simultaneous sources. Biological classification should preserve cell identity, developmental stage, metabolic condition, and relevant baseline susceptibility.
Standards should be evaluated through their demonstrated protective coverage and revised when evidence warrants.
Question 6 — Evidence relevant to the adequacy of current limits
The relevant question is whether exposures permitted by current limits adequately protect against the outcomes under consideration.
NTP’s chronic rat experiments used whole-body SARs of 1.5, 3, and 6 W/kg. These exposures cannot be treated as equivalent to ordinary human exposure. Their importance lies in the hazard findings and the need for transparent assessment of species differences, dose, duration, and susceptible tissues.
HHS should commission independent dose-response analyses of suitable datasets, compare alternative models, and report how assumptions affect proposed protective levels. Where the evidence cannot support a numerical threshold, identify the missing information and fund its acquisition.
Regulatory uncertainty should result in a defined research obligation, with a timetable.
Question 7 — Exposure measurement
Use calibrated field measurements, personal monitoring, device-operation records, and computational dosimetry together.
External measurements describe the environment. Dosimetry estimates the exposure within tissue. Neither can be replaced by device ownership, distance alone, or an unvalidated estimate of screen time.
Studies should report calibration, detection limits, spatial variability, measurement uncertainty, and the extent to which the monitoring system captures relevant frequencies and temporal patterns.
Question 8 — Improving real-world exposure assessment
Create reproducible reference exposures and shared protocols that laboratories can implement independently.
Record both average exposure and temporal structure. In experiments comparing waveforms, characterize differences in spectrum, peak amplitude, absorbed energy, temperature, and apparatus behavior before attributing an outcome to timing.
An RF modulation envelope should not be assumed to act as an equivalent independently applied low-frequency field. A proposed coupling or demodulation mechanism requires measurement or quantitative validation.
Population studies should improve longitudinal exposure estimation and distinguish RF exposure from behavioral correlates of technology use.
Question 9 — Information disclosure
Consumers and communities should receive understandable information about what compliance establishes and the conditions under which it was tested.
Disclosures should identify test configurations, averaging methods, body-separation assumptions, operating modes, and relevant limitations. Infrastructure information should include source characteristics, measurement or modeling dates, uncertainty, and realistic operating conditions.
Provide machine-readable technical records for independent analysis and plain-language explanations for the public.
Question 10 — Surveillance and reporting
Develop a privacy-protecting surveillance framework linking suitable exposure information with validated health outcomes.
Prioritize longitudinal cohorts, occupational monitoring, disease registries, and structured clinical reporting. Prespecify outcomes and analyses, retain appropriate comparison groups, and account for selection bias.
A symptom registry can identify questions and patterns; it cannot estimate disease incidence without a denominator or establish causation from attribution alone.
Surveillance should support progression from a reported pattern to exposure assessment, controlled testing, and independent replication.
Question 11 — Sensitive populations
Evaluate susceptibility directly rather than assume that an average response characterizes every population.
Research priorities should include developmental stages, pregnancy, childhood, aging, occupational exposure, preexisting conditions, and medical-device interactions.
Where genetic or physiological susceptibility is proposed, define it before examining outcomes and validate it in an independent sample. Avoid clinical labeling based on a single exploratory finding.
Developmental studies should match exposure assessment to the biological window under investigation. A biomarker measured at birth does not establish what occurred during an earlier developmental event.
Question 12 — Evidence below federal limits
Require each below-limit claim to identify the applicable frequency, exposure metric, averaging interval, and restriction.
A cellular response at a low SAR can establish a response under the tested conditions. Additional evidence is needed to establish injury, persistence, or human risk.
Conversely, failure to detect one endpoint under one exposure condition should constrain that endpoint and condition rather than close unrelated questions.
HHS should prioritize independent replication of below-limit findings with strong dosimetry, temperature control, blinded assessment, adequate statistical precision, and meaningful functional outcomes.
Question 13 — Emerging technologies and cumulative exposure
Assess technologies through their actual operating characteristics.
For cellular systems, measure traffic-dependent operation, beamforming, uplink and downlink contributions, frequency bands, and user geometry. For wearables and medical devices, characterize placement and transmission schedules. For satellite systems, distinguish distant sources from nearby transmitting terminals.
Cumulative exposure should retain information about repeated events and recovery intervals. Adding source counts or unlike exposure metrics does not produce a biologically validated cumulative dose.
HHS should determine which summaries of exposure history predict outcomes best.
Question 14 — Environmental effects
Support research using species-relevant exposure conditions and functional outcomes: orientation, reproduction, development, survival, and ecosystem interactions.
Cellular or molecular observations can identify mechanisms to investigate, but ecological conclusions require evidence at relevant biological and environmental scales.
Coordinate exposure methods and study priorities with appropriate environmental and wildlife agencies. Preserve negative findings and replicate consequential positive results.
Question 15 — Infrastructure and public protection
Evaluate infrastructure through measured or validated modeled exposure at occupied locations, including schools, homes, healthcare facilities, workplaces, and maintenance areas.
Account for existing sources, realistic operation, building geometry, device uplink behavior, and uncertainty. Distance is useful for planning but is not a universal measure of dose.
RF Safe supports prospective siting safeguards and comparative evaluation of fiber, wired access, and optical connectivity. Proposed setbacks should be identified as policy measures rather than represented as established biological thresholds.
HHS should provide independent health advice that informs FCC decisions and congressional consideration of local authority. Research funding and public-health assessment should remain distinct from deployment objectives.
Question 16 — Priority research: biological fidelity, susceptibility, and recovery
RF Safe proposes three coordinated research tracks.
Track A: Explain differential responsiveness.
Under matched exposures, compare relevant Schwann, glial, skin, and hematopoietic populations. Measure candidate transducer activity, calcium dynamics, mitochondrial function, redox state, and recovery capacity before exposure.
Test whether these measurements improve prediction in independent samples beyond physical dose and ordinary baseline variables.
This is the operational meaning of RF Safe’s “density gating” hypothesis. Greater abundance of signaling machinery may increase responsiveness, while greater buffering capacity may reduce disruption. The direction must be measured.
Track B: Identify necessary mechanisms.
The S4–Mito–Spin framework organizes candidate pathways involving voltage-sensitive channels, mitochondrial and redox processes, and spin-sensitive chemistry.
For each proposed route, establish the field reaching the target, the predicted molecular response, and the functional consequence. Use complementary interventions to test necessity and account for toxicity and off-target effects.
Kim and colleagues’ engineered EMF-inducible gene-switch work identifies Cyb5b and rhythmic calcium dynamics as experimentally tractable targets. Its relevance to consumer RF exposure requires separate testing under the appropriate exposure conditions.
FDA’s TheraBionic P1 record provides an additional comparator: an intentionally applied, amplitude-modulated RF therapy authorized under a probable-benefit standard, with calcium-channel-blocker contraindications. Its therapeutic context supports investigating defined pathways; it does not establish environmental harm.
Track C: Test persistent changes in susceptibility.
The central ceLLM hypothesis is that repeated exposure may leave a regulatory state that changes the response to a later challenge. RF Safe calls this a proposed “meta-disease” state.
Test sham, single, and repeated exposure groups. Remove the exposure and determine whether acute effects or injury remain. Apply a common subsequent challenge and measure both the regulatory state and functional response.
Evidence for this hypothesis requires more than a molecular difference. The state should predict later function, and an intervention affecting that state should alter the later response.
Prespecify biologically meaningful effect sizes, independent exposure replication, multiplicity controls, and criteria for rejecting the hypothesis. Demonstrate added explanatory value over residual injury, ordinary adaptation, or uncontrolled heating.
A small exploratory planarian study can separately assess whether a defined exposure changes remodeling time after an altered morphology is established. Its results would guide further work without being treated as direct evidence of human RF risk.
Question 17 — Agency coordination and accountability
HHS has an existing statutory foundation for research, exposure evaluation, and development of exposure-minimization techniques under 21 U.S.C. § 360ii, originating in Public Law 90-602. That foundation supports a sustained program; it does not predetermine the result of any experiment.
RF Safe requests:
Within six months: establish an interdisciplinary methods group with balanced expertise, public conflict disclosures, and a published evidence-review protocol.
Within twelve months: publish an outcome-specific evidence map, priority replication protocols, a research budget, and an exposure-reduction pilot plan.
Within twenty-four months: report initial replication findings, methodological improvements, and a reasoned assessment of whether interim recommendations should change.
FDA should lead electronic-product-radiation assessment. NIH and NIEHS should support mechanisms, cohorts, and replication, including renewed NTP-led chronic RF investigation. CDC should contribute surveillance and reporting methods. Coordinate technical measurement with agencies possessing relevant expertise.
Require preregistration, blinded assessment, independent replication, open methods, and publication of informative null results. Apply financial and institutional conflict scrutiny consistently across industry, standards organizations, advocacy groups, and investigators.
Question 18 — Evidence integration and policy direction
Assess the entire evidentiary record without reducing it to a publication count.
RF Safe’s research catalog is a discovery tool. Its classifications are not substitutes for study appraisal, and reviews should not be counted as independent experiments alongside their underlying studies.
Human evidence also constrains conclusions. COSMOS followed 264,574 participants; its highest calibrated cumulative-call-time category had a glioma hazard ratio of 1.07, with a 95% confidence interval of 0.62–1.86. HHS should evaluate that result within its measured exposure, follow-up, and outcomes while considering animal and mechanistic evidence separately.
Similarly, disagreements over oxidative-stress evidence should be resolved through transparent study inclusion, risk-of-bias assessment, assay evaluation, and replication. A 2024 systematic review’s very-low-certainty assessment identifies limitations requiring attention; earlier positive compilations identify findings requiring careful testing.
RF Safe’s proposed Clean Aether Act offers a complementary policy direction: reassessment of exposure protections, prospective siting safeguards, optical-connectivity pilots, and support for interoperable alternatives. These are proposals for evaluation, not claims that legislation has been enacted or that any technology is universally risk-free.
HHS should compare alternatives on measured exposure, performance, reliability, accessibility, cost, and their own safety requirements. Successful pilots should inform procurement and broader deployment.
Requested determination
RF Safe asks HHS to determine that the evidence warrants a sustained federal program capable of connecting exposure to biological response, recovery, and long-term function.
The Department need not endorse ceLLM or any other proposed mechanism to act. It should fund research capable of distinguishing competing explanations and publish how the results affect its recommendations.
The central question is whether an exposure standard protects the biological functions people depend on, across relevant populations and exposure histories.
Answering that question requires a research program with defined experiments, independent replication, transparent decisions, and accountable timelines.
RF Safe respectfully requests that HHS establish that program.

