New 2026 research turns “non-thermal” wireless effects from a vague dispute into a testable federal research agenda

For decades, the wireless-radiation debate has repeatedly stalled at the same rhetorical roadblock:

Where is the mechanism?

How could radiofrequency electromagnetic fields affect biology at exposure levels that do not produce substantial tissue heating? What part of the cell detects the field? What signaling pathway carries the response? Why might a pulsed commercial signal behave differently from a continuous laboratory carrier? Which tissues would be most susceptible, and which biological changes would actually matter for health?

Those were legitimate scientific questions.

They cannot remain permanent excuses.

A group of studies reviewed in the 2026 edition of ElektrosmogReport does not prove that every phone, router, or wireless network causes human disease. It does something more immediately important for public policy: it identifies specific molecular components, signal characteristics, cellular outcomes, and neural circuits that can now be tested directly.

The debate is moving from an undefined allegation to a falsifiable research program.

That is the real turning point.

A biological effect is not automatically an adverse health effect. A mechanism is not automatically a disease. But once a mechanism becomes specific, measurable, and falsifiable, refusing to test it is no longer scientific caution. It is a policy choice.

The German environmental organization diagnose:funk is right to call attention to this emerging body of work. But the public case must be stated with precision. Some headlines surrounding the studies go further than the underlying experiments support. Correcting those overstatements does not weaken the case for action. It makes the case far more difficult for policymakers to dismiss.

RF Safe’s conclusion is straightforward:

The new studies do not settle the safety of ordinary wireless exposure. They make the unanswered questions sufficiently concrete that HHS, FDA, NIEHS, and the FCC can no longer plausibly dismiss them as mechanistically undefined.

First, keep the electromagnetic categories straight

“Electromagnetic field” is an umbrella term. It includes static fields, extremely low-frequency fields, radiofrequency fields, microwaves, infrared radiation, visible light, and other portions of the electromagnetic spectrum.

That distinction matters because the most important new molecular paper—Kim and colleagues’ 2026 study in Cell—did not expose cells to a cell-phone, Wi-Fi, or 5G carrier frequency. It used a 60-hertz, 2.0-millitesla extremely low-frequency field.

Wireless devices, by contrast, transmit information using much higher-frequency carriers. Their emitted power can also rise and fall according to lower-frequency timing structures, frames, bursts, duty cycles, and modulation patterns.

Those phenomena may be biologically connected. But that connection has not yet been demonstrated merely because both involve low-frequency timing.

The lower-frequency envelope of a radio signal is not automatically equivalent to exposing tissue to a freestanding 60-hertz magnetic field. Establishing whether they activate the same cellular pathway requires bridge experiments designed specifically to answer that question.

This distinction produces a stronger—not weaker—policy argument:

A defined low-frequency electromagnetic exposure can engage a specific molecular pathway.
Commercial wireless signals demonstrably contain lower-frequency temporal structure.
The decisive next experiment is to determine whether realistic modulated RF signals engage that pathway under rigorously controlled exposure conditions.

Government laboratories should be performing those experiments now.

1. The Kim study identifies cellular machinery capable of responding to an electromagnetic field

The most consequential paper in the group is “Electromagnetic field-inducible in vivo gene switch for remote spatiotemporal control of gene expression,” published in Cell in 2026.

The researchers were not trying to prove that wireless technology causes illness. They were developing a way to control gene expression remotely for therapeutic purposes.

Using a 60-hertz, 2.0-millitesla field, they identified an electromagnetic-field-responsive element associated with the Lgr4 gene. A genome-wide CRISPR screen then implicated cytochrome b5 type B, or CYB5B, as an essential mediator and likely sensor in the response. Knocking out CYB5B eliminated activation of the engineered system; restoring it restored the response.

The investigators traced the response further. According to the detailed study review, CYB5B was linked to rhythmic intracellular calcium oscillations involving the voltage-gated calcium channel CACNA1F. Those oscillations activated downstream transcriptional machinery, including the transcription factor SP7, which then controlled expression through the electromagnetic-field-responsive promoter element. The team also demonstrated applications of the engineered gene-switch system in mouse models.

This matters because the researchers did not merely observe a statistical change in an endpoint and speculate afterward about how it happened. They used knockout, rescue, real-time calcium imaging, genetic screening, and in-vivo testing to build a mechanistic chain.

The emerging sequence can be summarized as:

Electromagnetic input → CYB5B-dependent response → voltage-gated calcium-channel activity → rhythmic calcium signaling → transcription-factor activation → gene-expression change

That is the kind of pathway regulators have long said they needed.

What the Kim study establishes

It establishes that cells can possess identifiable molecular machinery through which a defined electromagnetic exposure produces a reproducible signaling response.

It also shows that calcium dynamics matter as a pattern, not merely as a total quantity. The cellular response was associated with rhythmic oscillations rather than an undifferentiated increase in intracellular calcium.

Most importantly, it converts the vague concept of an “electromagnetic sensor” into experimentally accessible targets:

CYB5B can be deleted, restored, inhibited, or measured.
CACNA1F can be blocked or genetically modified.
Calcium oscillations can be imaged in real time.
Downstream transcription can be quantified.
Exposure frequency, intensity, waveform, duration, and duty cycle can be varied independently.

That is a major scientific advance.

What the Kim study does not establish

The exposure was not a mobile-phone or Wi-Fi signal. It was a relatively strong 60-hertz extremely low-frequency field.

The therapeutic outputs depended on an engineered gene-switch system. The study does not show that ordinary human tissues are being remotely reprogrammed by phones.

The detailed review also notes that the researchers did not find evidence of harmful effects from the field in ordinary wild-type mice under the tested conditions.

Therefore, the responsible conclusion is not:

“The Cell paper proves Wi-Fi or 5G causes disease.”

The responsible conclusion is:

The categorical claim that cells cannot possess specific electromagnetic-field-responsive molecular machinery is no longer tenable. The federal government must now determine whether realistic wireless signals can engage the same or related pathways.

The experiment that should follow immediately

Researchers should expose otherwise identical cells to:

A pure 60-hertz field;
An unmodulated RF carrier;
The same RF carrier amplitude-modulated at 60 hertz;
Recorded and replayed commercial wireless waveforms;
Sham exposure; and
Thermally matched controls.

Carrier frequency, average SAR, peak field, duty cycle, exposure geometry, sample temperature, and duration should be documented and controlled.

The test should then be repeated in normal cells and cells lacking CYB5B or CACNA1F.

That experiment could directly answer whether the Kim pathway is relevant to modulated wireless exposure.

The question is no longer impossible to formulate.

2. The Panagopoulos study shows that real wireless signals are temporal patterns—not simply carrier frequencies

The second important paper, by Dimitris Panagopoulos, Roman Litovsky, and Kent Chamberlin, examined the temporal structure of commercially generated wireless signals.

Using a spectrum analyzer in zero-span mode, the researchers recorded emissions from a Wi-Fi 6 router and an iPhone operating in 4G and 5G modes. The reviewed measurements involved a 2.45-gigahertz Wi-Fi signal, an 840-megahertz 4G signal, and a 1,876-megahertz 5G signal.

The reported measurements included:

Wi-Fi power variations around approximately 10 hertz, with observed variation between roughly 5 and 20 hertz;
A hierarchical 4G timing pattern with components around 500, 100, and 3 hertz; and
A comparable 5G pattern with components around 500, 100, and 13 hertz.

The signals also varied in pulse amplitude, length, spacing, and repetition.

These are measurements of how the RF power changes over time. They demonstrate that a commercial wireless exposure is not adequately described by stating only its nominal carrier frequency.

A 2.45-gigahertz Wi-Fi transmission is not necessarily a perfectly continuous 2.45-gigahertz sine wave. It may contain bursts, gaps, packets, beacons, changing amplitudes, and timing patterns created by the communications protocol and current network activity.

Why that matters experimentally

Many biological experiments use a continuous-wave signal generator because it offers control and reproducibility. That can be useful. But it may not reproduce the physical time structure of the signals people actually encounter.

If biological responses depend on peak fields, rise and fall times, pulse intervals, envelope frequencies, duty cycle, or variability, then two exposures with the same carrier frequency and the same average power could produce different results.

That is a testable proposition.

What the Panagopoulos paper does not prove

The paper measures wireless signals. It is not, by itself, an experiment showing biological injury.

The authors connect the measured timing structures to a proposed ion-oscillation and voltage-gated-ion-channel mechanism. But the measurement paper does not directly expose cells, disable a channel, or demonstrate that a particular pulse rate caused a particular biological endpoint.

Nor does a measured 10-hertz Wi-Fi power pattern mean that a separate 10-hertz carrier is radiating into the body as though it were an ordinary 10-hertz magnetic-field exposure. It describes the rate at which the RF signal’s power envelope changes.

The policy implication is nevertheless substantial:

Regulators and health agencies should not assume—without testing—that continuous-wave exposure and real commercial wireless exposure are biologically interchangeable merely because their average SAR or nominal carrier frequency is similar.

3. Average SAR is essential—but the question is whether it is sufficient

This is where the public discussion must become more technically disciplined.

Specific absorption rate, or SAR, measures the rate at which RF energy is absorbed per unit mass. It is indispensable for evaluating heating and comparing energy deposition.

Calling average power or SAR “meaningless” would be scientifically indefensible.

The proper question is whether average SAR, by itself, captures every exposure characteristic relevant to every possible biological interaction.

A six-minute average can distinguish between more and less total absorbed energy. It may not preserve information about whether that energy arrived continuously, in rapid bursts, through changing peaks, or according to a repeating low-frequency timing structure.

Two signals can have equal average power while differing in:

Peak field;
Pulse width;
Pulse spacing;
Duty cycle;
Rise and fall time;
Modulation;
Polarization;
Frequency variability; and
Spatial distribution within tissue.

The new research has not yet proven that each of those variables independently changes human health risk. But it has made the possibility specific enough to test.

The correct regulatory question is therefore not:

“Should SAR be abandoned?”

It is:

“Has the government demonstrated that SAR and time-averaged power are sufficient to predict all health-relevant effects of modern pulsed and modulated signals?”

That is closely aligned with the question the D.C. Circuit said the FCC had failed to address adequately when it discussed long-term exposure, pulsation, modulation, newer technology, and effects unrelated to cancer.

4. The Jangid studies raise legitimate reproductive questions—but they did not study human sperm

Two 2026 papers from the Jangid research group examined radiofrequency exposure in TM3 mouse Leydig cells.

Leydig cells are located in the testes and produce testosterone, making them important to male endocrine and reproductive function. But they are not sperm cells.

That distinction should remain explicit.

The first study, published in Scientific Reports, exposed cultured TM3 cells to:

A 4G Xiaomi smartphone operating at approximately 2,318 megahertz;
A continuous-wave 1,800-megahertz signal; and
A continuous-wave 2,450-megahertz signal.

Exposure periods ranged from 15 to 120 minutes. The phone was positioned one centimeter from the sample, and the researchers estimated a sample SAR of approximately 0.5 watts per kilogram.

The authors reported no detectable temperature change during the phone exposure and less than a 0.1-degree Celsius difference between exposed and control samples in the continuous-wave experiments. “Non-thermal” in this context therefore means that the researchers did not observe a meaningful bulk temperature increase under their measurement conditions. It does not prove that no microscopic temperature gradients or other dosimetric complexities existed anywhere in the samples.

The reported outcomes included:

Changes in cellular morphology;
Reduced DNA synthesis;
Reduced proliferation;
Changes in cell-cycle distribution;
Increased G1-phase arrest under certain conditions; and
Reduced numbers of cells progressing through the DNA-synthesis phase.

The earliest reduction in DNA synthesis was reported in the commercial-phone condition, while other endpoint patterns varied by source, frequency, and exposure duration.

A companion paper in Toxicology and Applied Pharmacology examined redox balance and apoptosis in the same type of cell. It reported source- and frequency-dependent changes in nitric oxide and superoxide, along with increases in apoptotic cells under the tested conditions.

Together, the papers offer a possible sequence:

RF exposure → redox disturbance → altered replication and cell-cycle control → apoptosis or impaired Leydig-cell function

That is biologically relevant because Leydig-cell dysfunction could, in principle, affect testosterone production and the cellular environment that supports spermatogenesis.

Why these papers matter

The work moves beyond a single nonspecific stress marker. It examines morphology, DNA synthesis, cell-cycle distribution, reactive species, and programmed cell death.

It also compares a commercial phone signal with continuous laboratory signals, which helps frame the question of whether a real-world waveform behaves differently from a simplified carrier.

What these studies do not establish

They do not demonstrate sperm damage directly.

They do not establish infertility in men.

They do not show that carrying a phone in a pocket produces the same dose in human testes as exposing cultured cells one centimeter from a phone.

They also do not prove that modulation caused the stronger or earlier response in the phone group. The phone and laboratory exposures differed in carrier frequency, waveform, geometry, field distribution, source, and estimated dosimetry. To attribute the difference specifically to modulation, those other variables must be matched.

The Scientific Reports paper acknowledges important limitations, including the use of one mouse cell line, incomplete three-dimensional field mapping, averaged rather than fully spatially resolved SAR estimates, possible exposure gradients within the samples, and the lack of systematic evaluation of modulation, duty cycle, polarization, and waveform complexity. The authors also acknowledge that comparable studies have sometimes produced little or no measurable effect.

Those limitations do not erase the results.

They define the replication work that must follow.

The next reproductive studies should include

The federal research program should replicate these findings using:

Multiple Leydig-cell models;
Human testicular organoids or other human-relevant systems;
Preregistered protocols;
Blinded biological analysis;
Independent exposure verification;
Thermal clamping;
Three-dimensional dosimetry;
Real commercial waveforms and synthetic replay signals;
Matched average SAR, peak fields, and duty cycles; and
Longer-term animal studies capable of measuring hormones, sperm parameters, fertility, and offspring outcomes.

That is how an in-vitro signal becomes—or fails to become—a credible human risk assessment.

The government should neither proclaim human infertility from one cell-culture experiment nor ignore a reproductive pathway merely because the first experiment was performed in cells.

5. The Cui study maps a neural circuit—but under specialized exposure conditions

A 2026 study by Cui and colleagues investigated conditioned fear-memory retrieval in mice after exposure to static magnetic fields and microwaves.

The study is technically notable because it did not stop at a behavioral observation. Researchers used calcium imaging, neural tracing, histopathology, immunofluorescence, and chemogenetic manipulation to examine a circuit connecting the primary auditory cortex with the basolateral amygdala.

The authors reported reduced conditioned-fear retrieval, structural changes in the examined brain regions, reduced calcium activity, and changes involving downstream cholinergic signaling. Chemogenetic manipulation of the circuit helped test its functional involvement.

This is mechanistically richer than simply reporting that exposed animals behaved differently.

But the exposure must be described accurately.

The mice received a 100-millitesla static magnetic field for one hour and a 9.375-gigahertz microwave exposure at an average power density of 12 milliwatts per square centimeter for 15 minutes. The estimated whole-body average SAR was approximately 2.58 watts per kilogram. The researchers reported no significant increase in rectal temperature during or immediately after microwave exposure.

The authors themselves placed the exposure context in relation to specialized occupational environments involving radar or other electromagnetic equipment and noted that surrounding residential exposure levels are generally much lower.

Therefore, this study does not prove that routine cell-phone use causes anxiety, memory loss, or damage to the human amygdala.

Its significance is more specific:

Under a defined, relatively intense, combined electromagnetic exposure, researchers identified a circuit-level pathway associated with a measurable behavioral effect and tested that pathway using modern neuroscience tools.

That provides a model for how future RF studies should be conducted.

Instead of measuring one behavioral endpoint and speculating broadly, researchers can ask which neurons respond, which calcium patterns change, which pathways carry the effect, and whether manipulating those pathways prevents or reverses the outcome.

6. The Cantiello microtubule paper is mechanistic background—not direct wireless evidence

The 2025 study by Cantiello and colleagues examined spontaneous electrical oscillations in tissue from the adult rat hippocampus and neocortex.

The authors reported electrical activity with prominent frequency components around 38 and 93 hertz, regional differences in the distribution of oscillatory energy, and inhibition by the microtubule-stabilizing drug paclitaxel. The work suggests that microtubular structures could contribute to endogenous electrical organization within neural tissue.

But the study is a bioRxiv preprint and had not completed conventional journal peer review at the time cited.

More importantly, it did not expose the tissue to a wireless signal.

It is therefore not evidence that cell phones alter brain microtubules.

Its relevance is hypothesis-generating. If neural cytoskeletal structures support endogenous oscillations, researchers can investigate whether externally applied fields alter those oscillations, under what conditions, and with what functional consequence.

Until that experiment is done, connecting the paper directly to wireless injury remains speculative.

RF Safe’s position is that potentially important mechanistic evidence should not be exaggerated. A claim that cannot survive close reading will not survive regulatory review.

7. Melnick and Moskowitz challenge the limits through risk modeling—not a newly observed human threshold

A separate 2026 analysis by Ronald Melnick, Joel Moskowitz, and the International Commission on the Biological Effects of Electromagnetic Fields applied benchmark-dose and risk-assessment methods to experimental-animal data.

The researchers used linear low-dose extrapolation from benchmark-dose values for cancer and traditional uncertainty factors for reproductive toxicity. They estimated model-based whole-body SAR values below the current general-public whole-body limit of 0.08 watts per kilogram. The paper recommended an independent re-evaluation of existing exposure limits.

The analysis is significant because it asks a question conventional exposure-limit discussions often avoid:

If animal evidence is treated through the risk-assessment methods commonly applied to environmental hazards, what exposure values result?

But its numerical values are not directly observed human thresholds.

They depend on:

The selected animal datasets;
Benchmark-dose modeling;
Linear low-dose extrapolation;
The chosen acceptable-risk level;
Exposure-duration assumptions;
Animal-to-human extrapolation; and
Applied uncertainty factors.

Reasonable experts can debate those assumptions.

That debate should occur openly. It should not be avoided by pretending the animal data have no risk-assessment implications.

The paper does not, by itself, establish the replacement limit regulators must adopt. It does demonstrate that different health-protective methodologies can yield materially different answers from a framework centered on preventing established acute heating effects.

These studies form a convergence map—not one completed causal chain

Taken together, the papers suggest a possible research framework:

Wireless signal timing
↓
Candidate cellular sensing or transduction
↓
Voltage-gated calcium-channel and redox responses
↓
Calcium oscillations and gene regulation
↓
Cell-cycle, endocrine, apoptotic, or neural-circuit outcomes
↓
Potential tissue or health consequences

That framework is scientifically useful.

But the arrows should be treated as hypotheses requiring direct verification.

No single study has demonstrated the entire sequence from an ordinary phone signal to CYB5B activation, calcium oscillation, reproductive injury, neural-circuit disruption, and human disease.

The studies used different:

Frequencies;
Field types;
Intensities;
Exposure durations;
Biological models;
Endpoints; and
Experimental objectives.

Convergence does not mean interchangeability.

The correct scientific conclusion is not that every paper proves the same disease mechanism.

It is that the papers have made a coordinated series of bridge experiments possible.

The mechanism question has not disappeared. It has become experimentally manageable.

Five conclusions the evidence supports

First, a defined electromagnetic exposure can activate a specific molecular pathway involving CYB5B, voltage-gated calcium signaling, rhythmic calcium dynamics, and gene expression. The demonstrated field was 60-hertz ELF, not wireless RF.

Second, commercial wireless emissions contain lower-frequency timing and amplitude structures that continuous-wave laboratory carriers may not reproduce.

Third, selected cellular and animal experiments have reported biological changes under conditions where researchers did not detect meaningful bulk temperature increases. That does not automatically establish adverse human health effects or exclude every possible thermal contribution.

Fourth, reproductive and neurological endpoints now have sufficiently specific candidate pathways to justify targeted, independently replicated investigation.

Fifth, the sufficiency of a safety framework based primarily on absorbed energy and thermal protection is a legitimate research and regulatory question—not a subject that agencies may resolve through conclusory assurances.

Five conclusions the evidence does not yet support

The Kim study does not prove that Wi-Fi or 5G activates CYB5B in ordinary human tissue.

The Jangid studies do not prove that smartphones cause human infertility.

The Panagopoulos paper does not prove that a particular wireless pulse rate causes cellular damage.

The Cui study does not prove that routine consumer wireless exposure damages human memory or anxiety circuits.

The Melnick–Moskowitz analysis does not establish a directly measured human harm threshold.

Those distinctions are not concessions to industry.

They are what separate a durable scientific case from an advocacy claim that can be dismissed by identifying one exaggeration.

Why this now becomes a federal legal-responsibility question

The new research arrives in a legal environment that already contains an explicit federal assignment.

Public Law 90-602, now codified principally at 21 U.S.C. §§ 360hh–360ss, defines electronic-product radiation to include ionizing or non-ionizing electromagnetic radiation emitted from an electronic product. That language encompasses radiofrequency emissions from wireless electronic products.

The law states:

“The Secretary shall establish and carry out an electronic product radiation control program…”

The Secretary referenced in this statutory framework is the Secretary of Health and Human Services.

The required program includes planning, conducting, coordinating, and supporting research; studying emissions and exposure conditions; developing and evaluating methods of minimizing unnecessary exposure; coordinating with other agencies; administering performance standards; and making research information available.

Robert F. Kennedy Jr. is the current HHS Secretary. That makes performance of the statute an official responsibility of his office. HHS may assign work to FDA, NIH, NIEHS, NTP, or another operating division, but internal delegation does not move Congress’s assignment outside HHS.

The significance of the new papers is therefore not that Congress must first enact a law authorizing research.

Congress already did.

The central question is:

What current HHS program is testing whether the newly identified molecular pathways, pulse characteristics, redox effects, and tissue-specific outcomes occur under contemporary wireless exposure conditions?

HHS should be able to answer that question with office names, projects, budgets, protocols, grants, deliverables, and publication dates.

FDA supplies health expertise; the FCC owns the final communications decision

FDA’s own website explains that it shares regulatory responsibility for cell phones with the FCC.

FDA says it provides scientific input and expertise, consults on testing and evaluation, and collects and analyzes information concerning electronic-product-radiation hazards. The FCC sets RF-emission limits for phones and similar wireless products. FDA’s public page also states that the weight of scientific evidence has not linked cell-phone RF exposure with health problems—but the page identifies its content as current as of May 2021.

That does not mean FDA’s conclusion must be presumed wrong.

It means the Department should publish the analytical work supporting it and update that work when significant new evidence appears.

A legally and scientifically useful FDA assessment should contain:

A documented literature-search strategy;
Inclusion and exclusion criteria;
Evidence tables;
Study-quality and risk-of-bias assessments;
Exposure and dosimetry comparisons;
Separate treatment of human, animal, cellular, and mechanistic evidence;
Analysis of conflicting findings;
Evaluation of children and developmental exposure;
Evaluation of modulation, pulse structure, peaks, duty cycle, and chronicity;
Discussion of remaining uncertainties; and
Identification of the experiments needed to resolve those uncertainties.

The FCC cannot perform all of that work simply by applying communications-engineering compliance rules.

But the FCC also cannot escape its own responsibility by pointing to an unexplained FDA conclusion.

The D.C. Circuit already rejected interagency handwaving

In August 2021, the D.C. Circuit held that the FCC had failed to provide a reasoned explanation for its conclusion that its RF guidelines adequately protected against harmful effects unrelated to cancer.

The court found the Commission’s treatment of evidence concerning children, device-testing procedures, long-term exposure, pulsation and modulation, technological developments, ubiquitous wireless use, and environmental effects legally inadequate.

The court did not rule that RF exposure causes a particular disease.

It did not order the FCC to adopt lower limits.

It found the FCC’s explanation insufficient and expressly stated that it was taking no position in the underlying scientific debate.

Most importantly, the court addressed the relationship between the FCC and the health agencies.

The FCC could seek a reasoned scientific explanation from FDA. But when FDA supplied only conclusory statements, the FCC had to obtain competent analysis elsewhere or produce its own explanation. The silence of other expert agencies was not evidence that the scientific questions had been resolved.

That holding creates a clear institutional chain:

HHS must build the health-science record. The FCC must act on and publicly explain a reasoned record. Neither agency may use the other agency’s silence as an excuse.

The new mechanistic papers fall directly within subjects the court said required reasoned treatment—especially modulation, non-cancer effects, reproductive evidence, neurological outcomes, children, and technological change.

The decision will reach its fifth anniversary on August 13, 2026.

The court did not impose a five-year statutory deadline. But after nearly five years, the FCC should identify the final agency action it believes answers every element of the remand—or publish a timetable for completing that work.

HHS has retired its old RF research system without identifying a modern successor

The National Toxicology Program’s earlier two-year studies reported a link between the tested 2G/3G radiofrequency exposures and malignant heart tumors in male rats, along with other findings that varied by sex, species, and endpoint. NTP cautions that the exposure conditions differed significantly from ordinary human phone use and should not be directly extrapolated to human risk.

NIEHS subsequently developed a smaller follow-up exposure system. Its current public page says the work was technically challenging and resource-intensive, that the system was not representative of modern 4G or 5G technologies, and that NIEHS has no plans to conduct additional RF-exposure studies with that system—or additional RFR exposure studies at this time.

Retiring an inadequate experimental platform can be scientifically responsible.

Retiring it without creating a visible modern successor is a different matter.

Public Law 90-602 does not require HHS to preserve one obsolete chamber or repeat one experiment forever. It does require a functioning program capable of researching electronic-product radiation, evaluating exposure conditions, and developing exposure-minimization methods.

The appropriate demand is therefore:

Restart or replace the federal RF research capability.

Obsolete equipment may be retired.

The statutory research duty may not.

Protecting children requires research—not panic and not denial

None of the new studies directly establishes injury in children.

That statement must remain clear.

But children should be placed at the center of the federal research agenda because their exposure can begin early, continue for decades, and occur in environments they do not control—including schools, childcare facilities, transportation systems, and homes.

The D.C. Circuit specifically found the FCC’s treatment of children inadequate. The court emphasized that showing a compliance test was “conservative” did not answer the underlying question of whether exposures allowed by the limits could produce effects unrelated to cancer.

A serious children’s-health program would ask:

Whether developmental tissues respond differently from mature tissues;
Whether prolonged exposure produces outcomes not seen after short tests;
Whether sleep-period exposure merits separate study;
Whether real commercial modulation differs biologically from continuous-wave exposure;
Whether simultaneous exposure from multiple sources changes total dose or response;
Whether brain, cardiac, reproductive, or endocrine tissues require specialized dosimetry; and
Whether practical design choices can reduce unnecessary exposure while preserving connectivity.

This does not require officials to declare wireless technology unsafe before the research is complete.

It requires them to stop treating incomplete research as proof that no research is needed.

The public does not need agencies to prejudge harm. It needs them to stop prejudging irrelevance.

The federal research program that should begin now

1. Conduct signal-matched bridge experiments

Commercial and laboratory exposures should be compared while matching carrier frequency, average SAR, peak field, duty cycle, geometry, duration, and temperature.

Researchers should compare:

Continuous-wave RF;
Regularly modulated RF;
Irregularly modulated RF;
Recorded commercial-device signals;
Synthetic replay of those signals; and
Pure low-frequency fields matching the RF envelope rate.

This would directly test whether modulation or temporal variability changes biological response independently of total absorbed energy.

2. Test the CYB5B–calcium pathway under realistic RF exposure

Human-relevant cell models should be exposed to contemporary Wi-Fi, 4G, and 5G waveforms while researchers measure:

CYB5B activity;
CACNA1F and other calcium-channel activity;
Intracellular calcium oscillations;
Mitochondrial redox state;
Gene expression;
Cell-cycle regulation; and
Apoptosis.

The work should be repeated after genetic deletion, rescue, and pharmacological inhibition of the candidate pathway.

A response that disappears when CYB5B is deleted would be mechanistically informative. A response that persists would indicate that another pathway is involved.

Either result advances science.

3. Use rigorous dosimetry and thermal controls

Bulk temperature monitoring alone is not sufficient to characterize every exposure.

Studies should include:

Three-dimensional computational dosimetry;
Direct field mapping;
Spatial SAR estimates;
Peak and average measurements;
Temperature probes with appropriate temporal and spatial resolution;
Thermally matched controls;
Sham exposures indistinguishable to laboratory personnel; and
Complete publication of waveform files and device settings.

4. Move from isolated cells to human-relevant systems

Cell lines are useful for discovering mechanisms. They cannot establish whole-body risk by themselves.

Findings should progress through:

Multiple normal cell types;
Human organoids;
Co-culture systems;
Animal models;
Developmental and chronic-exposure studies; and
Carefully designed human observational or experimental research where ethically appropriate.

5. Require independent replication

High-impact findings should be replicated by laboratories that did not design the original hypothesis or exposure apparatus.

Protocols should be preregistered where feasible. Biological analysts should be blinded. Raw data, code, dosimetry, waveforms, and negative findings should be published.

6. Study the technology people actually use

The federal program must examine contemporary exposure rather than assuming that 2G and 3G experiments answer every question about modern systems.

That includes representative combinations of:

Current cellular protocols;
Wi-Fi generations;
Wearables;
Bluetooth devices;
Connected school equipment;
Near-body devices;
Adaptive power control;
Beamforming where relevant; and
Mixed-source environments.

The purpose is not to test every device individually. It is to identify which physical exposure variables predict biological response.

What policymakers should demand

HHS: publish a Public Law 90-602 compliance matrix

HHS should map every mandatory function in 21 U.S.C. § 360ii to:

The responsible operating division;
Current wireless-RF projects;
Staffing and funding;
Exposure conditions being evaluated;
Relevant grants and contracts;
Planned scientific products;
Publication dates; and
The accountable official.

If HHS believes its present activities fully satisfy the statute, it should show the program.

If gaps exist, it should publish a corrective plan.

No court has yet issued a ruling that HHS is presently violating Public Law 90-602. RF Safe does not need to invent a judicial finding that does not exist.

The stronger demand is evidentiary:

Show how every statutory function is being performed for contemporary wireless exposure.

If HHS cannot do that, the failure becomes documentable rather than rhetorical.

NIEHS and NTP: create a modern successor program

The federal government should establish an exposure platform capable of studying contemporary waveforms, long-term exposure, developmental periods, non-cancer endpoints, and the variables identified by the D.C. Circuit.

FDA: publish a formal technical assessment

The assessment should be suitable for incorporation into an FCC administrative record. It should reveal the analytical path from studies to conclusions rather than relying on an undated or outdated assurance concerning the “weight of evidence.”

FCC: publish a remand-compliance map

The FCC should identify where it has addressed:

Children;
Long-term exposure;
Pulsation and modulation;
Device-testing procedures;
Ubiquitous wireless use;
New technologies;
Non-cancer evidence; and
Environmental effects.

If the Commission has not completed that work, it should open or identify a public proceeding with deadlines.

Congress: hold joint oversight hearings

Congress should call HHS, FDA, NIEHS/NTP, FCC, EPA, and other relevant agencies before the appropriate committees.

The hearing should request projects, budgets, protocols, interagency communications, and timelines—not generalized expressions of confidence.

Congress should also request a GAO or inspector-general review of HHS performance under the electronic-product-radiation statutes and the status of the FCC’s court remand.

Local officials may be constrained—but they are not powerless

Federal law limits the ability of state and local governments to regulate the placement, construction, or modification of compliant personal-wireless facilities based on the environmental effects of RF emissions.

That makes federal accountability more important, not less.

Local officials do not need authority to rewrite the FCC’s numerical limit before they can:

Hold a public health hearing;
Pass a resolution requesting HHS and FCC action;
Ask their congressional delegation to investigate Public Law 90-602 compliance;
Request participation from local physicians, scientists, engineers, parents, and educators;
Review wireless procurement and network-management policies in public buildings;
Evaluate wired, fiber, and optical alternatives where practical;
Avoid unnecessary always-on transmitters where connectivity is not needed;
Require transparent documentation of device and facility compliance; and
Place wireless exposure within broader children’s environmental-health discussions.

A city council resolution will not complete the missing federal research.

It can make continued federal inaction politically visible.

A school board cannot decide the national toxicological question.

It can ask why children’s exposure policies continue to rely on a federal review the court found inadequately explained.

Three questions every official should be required to answer

1. Where is the HHS program?

Which current offices, projects, budgets, and deliverables perform the wireless-RF research and exposure-evaluation functions required by 21 U.S.C. § 360ii?

2. What replaced the federal NTP research capability?

If the earlier exposure system was obsolete, what modern program now evaluates contemporary signals, developmental exposure, non-cancer endpoints, modulation, and long-term effects?

3. Where is the FCC’s completed response to the 2021 remand?

Which final agency action addresses children, long-term exposure, device testing, modulation, technological change, ubiquitous use, non-cancer effects, and environmental impacts?

The public should ask for document titles, docket numbers, responsible officials, budgets, and dates.

Not talking points.

Not institutional silence.

Not another assurance that a different agency is handling it.

A message constituents can send today

Subject: Put wireless-radiation health research and Public Law 90-602 on the public agenda

Public Law 90-602, now codified at 21 U.S.C. §§ 360hh–360ss, assigns the electronic-product-radiation health program to the Secretary of Health and Human Services. The law expressly includes non-ionizing electromagnetic radiation and requires research, exposure evaluation, exposure-minimization work, and interagency coordination.

New 2026 studies have identified specific electromagnetic-field-responsive molecular machinery, measured the pulse structure of commercial wireless signals, and reported reproductive-cell and neural-circuit effects that warrant targeted independent investigation. These studies do not predetermine the outcome of a health-risk assessment, but they make the necessary experiments clear.

In 2021, the D.C. Circuit ordered the FCC to provide a reasoned explanation addressing children, long-term exposure, device testing, technological change, non-cancer effects, and environmental impacts. The fifth anniversary of that decision is approaching.

Please place this issue on the public agenda and formally request:

An HHS compliance report mapping every duty in 21 U.S.C. § 360ii to current programs, budgets, and deliverables;

A modern successor to discontinued or inadequate federal RF-exposure research;

A transparent FDA scientific assessment suitable for the FCC’s administrative record; and

A section-by-section FCC response to the 2021 court remand.

This request does not presume that every wireless exposure is harmful. It asks the responsible agencies to perform the research, analysis, and public explanation required by law—especially where children’s health is concerned.

Send it to city councils, county commissions, school boards, public-health departments, state legislators, governors, representatives, senators, and relevant congressional committees.

Ask that it be entered into the public record.

Ask for a written response.

Ask for the issue to be placed on a meeting agenda.

The turning point is accountability

The latest research has not delivered a final verdict on wireless health risk.

Science rarely advances through a single final paper.

What has changed is the level of specificity.

Researchers now have:

A candidate electromagnetic-field-responsive protein;
A calcium-channel pathway;
A measurable calcium-oscillation signature;
Real-device temporal waveforms;
Reproductive-cell redox and cell-cycle endpoints;
A mapped neural circuit;
Animal evidence suitable for formal risk modeling; and
A list of decisive bridge experiments.

The federal government does not need to assume those pieces form one completed disease pathway.

It needs to test whether they do.

Public Law 90-602 places the electronic-product-radiation health program under the HHS Secretary.

FDA supplies health expertise.

NIEHS and NTP possess toxicological and experimental capabilities.

The FCC owns the communications rules and the final regulatory explanation.

Congress owns the oversight responsibility.

Local officials and the public own the responsibility to make continued silence politically impossible.

HHS must build the health record.
The FCC must act on a reasoned record.
Neither agency may use the other’s silence as an excuse.

The mechanism excuse is running out.

The research duty is not.

And children should not have to wait another generation for the government to study the technology surrounding them every day.

Source

Facebook X Reddit Pinterest Google+LinkedIn