Is 5G Dangerous? The Burning Towers, the Rat Tumours, and the Real Answer

Why this fight is louder than ever

No technology in living memory has been argued about the way 5G has. People did not burn down 4G towers. Nobody set fire to a Wi-Fi router and called it a health protest. But in April 2020, in the middle of a pandemic, people in Britain set fire to mobile phone masts — at least 20 of them in a single week — because they believed the new network was spreading the coronavirus, or suppressing immune systems, or both. Telecoms engineers were spat at and abused in the street. The British government called the theory "dangerous nonsense". The towers that burned were mostly 3G and 4G. One of the first, a BT tower in Birmingham, did not even have 5G capability. The fire was real. The reason was not.

That is the strange shape of the 5G argument. It is the only health debate in which the fear has produced arson, and the only one in which the fear is routinely answered with the observation that the fear itself did more damage than the technology ever has. And yet the argument will not die, because each side has a piece of real evidence the other side would rather not talk about. The defenders have the measurements: real-world exposure is a tiny fraction of the limit, and sixty years of radio-wave research has produced no established mechanism by which these waves could cause cancer. The critics have the rats: two large, expensive, government-funded animal studies found the same rare heart tumour in male rats, and one of them found it at exposure levels meant to mimic living near a cell tower. Both sides are holding genuine evidence. Neither side is holding the whole story.

The question is also one of the most searched health questions on the internet. "Is 5G dangerous?" and "does 5G cause cancer?" are typed into Google millions of times a year, and the answers people find are mostly useless: content-farm pages that repeat the same three sentences, activist sites that quote the rat studies as proof of a cover-up, and telecom industry pages that declare the matter settled. None of them does the obvious thing, which is to lay the whole record out — the studies, the limits, the measurements, the conspiracy, and the gaps — and let the evidence speak. That is what this article does.

What 5G actually is — and what it is not

5G is the fifth generation of mobile network technology. It is not a new kind of energy. It is the same radiofrequency electromagnetic field — radio waves — that mobile phones have used since the 1980s, carrying more data, faster, with less delay. What changed is the engineering: more base stations, smaller ones, and antennas that steer their signal toward the phone using it instead of broadcasting in all directions.

Most 5G today runs on frequencies between about 3.4 and 3.8 gigahertz, just above the frequencies 4G uses. That is the sub-6 GHz band, and it behaves much like 4G: it travels through walls, it covers kilometres, and it has been studied for decades because nearby frequencies have been in use for years. The newer, faster flavour of 5G uses millimetre waves, roughly 24 to 40 gigahertz. These are the frequencies that frighten people, because they are higher, and higher sounds more dangerous. In practice millimetre waves are weaker in the ways that matter: they travel only a few hundred metres, they are blocked by walls, windows and even rain, and — critically — they are absorbed in the outermost layers of skin. They do not reach internal organs. That is why the international safety guidelines switch, above 6 gigahertz, from measuring energy absorbed deep in the body to measuring energy absorbed at the surface.

The word that causes all the trouble is "radiation". In physics, radiation simply means energy travelling as waves, and the electromagnetic spectrum runs from radio waves at one end to gamma rays at the other. The dividing line that matters for health sits in the middle. Ionising radiation — X-rays and gamma rays — carries enough energy to knock electrons off atoms, damage DNA, and raise cancer risk. Non-ionising radiation — radio waves, visible light, and microwaves — does not carry enough energy to do that. It can heat tissue, and enough heat can burn you, but it cannot alter DNA and it cannot make anything radioactive. 5G sits on the non-ionising side of that line, next to FM radio and Wi-Fi. The sun's ultraviolet light, which genuinely does damage DNA and cause skin cancer, is more energetic than any 5G signal will ever be.

None of that settles the argument, because the argument was never really about physics. It is about trust: who wrote the safety limits, what they were designed to protect against, and whether a technology rolled out by the same companies that profit from it was tested for long enough before it arrived on every street corner. Those are fair questions, and they have real answers — some reassuring, some not.

The cancer question: what "possibly carcinogenic" actually means

The single most quoted fact in the 5G debate is that the World Health Organization's cancer agency has classified radiofrequency fields as "possibly carcinogenic to humans". It is true, and it is also the most misunderstood sentence in the whole argument. In May 2011, the International Agency for Research on Cancer (IARC) convened 31 scientists from 14 countries in Lyon, France, and classified radiofrequency electromagnetic fields as Group 2B — "possibly carcinogenic to humans" — based on limited evidence of an increased risk of glioma, a malignant type of brain cancer, associated with wireless phone use.

Here is what Group 2B means, and why it matters. IARC's scale runs from Group 1 (carcinogenic to humans: tobacco, asbestos, sunlight) down through Group 2A (probably carcinogenic), Group 2B (possibly carcinogenic), Group 3 (not classifiable) and Group 4 (probably not carcinogenic). Group 2B is the weakest positive classification on the scale. It means the evidence is limited, the working group could not rule out a link, and it could not establish one either. The other members of Group 2B include pickled vegetables, aloe vera extract, and gasoline engine exhaust. Coffee sat in Group 2B until 2016, when it was moved down to Group 3. Being in Group 2B is not a verdict that something causes cancer. It is a verdict that the question is still open.

The evidence that put radio waves in Group 2B came mostly from two sources. The first was the INTERPHONE study, a 13-country case-control study published in 2010, which found that the heaviest mobile phone users — people reporting an average of 30 minutes a day over ten years — had a 40% higher risk of glioma than people who had never used a phone. The second was a pooled analysis of Swedish studies, which found a 30% higher risk of glioma among people who had used a mobile phone for more than a year. Both studies were vulnerable to the same flaw: they relied on people remembering, years later, how much they had used a phone, and how they held it — and people with brain tumours remember their phone use differently from people without them. The IARC working group itself noted that the findings could reflect bias, but concluded they could not be dismissed as bias alone. Its chairman, Dr Jonathan Samet, said the conclusion meant "there could be some risk, and therefore we need to keep a close watch for a link between cell phones and cancer risk." That was 2011. The watch has been running for fifteen years.

What has the watch found? The strongest counter-evidence is not a single study but a population trend. Mobile phone use went from rare to near-universal in the space of two decades — from about 5 billion subscriptions worldwide in 2011 to far more today, with most people carrying a phone against their head for hours a day. If phones caused brain tumours at anything like the rate the INTERPHONE heavy-user figure suggested, brain tumour rates should have climbed steeply. They have not. National cancer registries in several countries show no rise in glioma rates tracking the explosion in phone use, and a Danish cohort study that followed mobile phone subscribers from 1982 to 1995 found no increased risk of glioma or other brain tumours at all.

Then came the most comprehensive review yet. In 2024, a WHO-commissioned systematic review published in Environment International examined 63 studies on mobile phone use and cancer, and found no association between mobile phone use and brain cancer — not for glioma, not for meningioma, not for acoustic neuroma, and not for other head and neck cancers. The review covered studies published between 1994 and 2022, including the INTERPHONE data that drove the original 2B classification. It is the single largest synthesis of the evidence ever done, and its answer is that the signal the IARC saw in 2011 has not survived fifteen years of further research.

That does not make the 2011 classification wrong — it was an honest reading of the evidence available at the time — but it does mean the classification is now stale. The IARC has not re-evaluated radiofrequency fields since 2011, and the evidence base it judged has been substantially overtaken. The critics are right that the 2B label exists and right that it has never been formally withdrawn. The defenders are right that the evidence behind it has weakened rather than strengthened since it was issued. Both facts belong in the same paragraph, and most articles about 5G contain only one of them.

The rat studies: the critics' strongest evidence, honestly stated

If the 5G debate has a piece of evidence that should make the defenders uncomfortable, it is not a blog post or a petition. It is two large, peer-reviewed animal studies, run by government-funded research institutes, that found the same rare tumour in the same sex of the same species. Dismissing them requires work, and the people who dismiss them usually do not do it.

The first was the US National Toxicology Program (NTP) study, published as two technical reports in November 2018 after more than a decade of work reported to have cost around $25–30 million. Male and female rats and mice were exposed to radiofrequency radiation like that used by 2G and 3G phones — 900 megahertz for rats, 1,900 megahertz for mice — in special chambers, ten minutes on and ten minutes off, for about nine hours a day, for two years. The exposure levels ranged from 1.5 to 6 watts per kilogram in rats, and the lowest level was equal to the highest level permitted for local tissue exposure from a phone today; the highest was four times that. The NTP's conclusions were unambiguous: clear evidence of an association with malignant schwannomas — tumours of the nerve tissue — in the hearts of male rats, some evidence of malignant gliomas in the brains of male rats, and some evidence of tumours in the adrenal glands of male rats. Female rats and mice showed no clear pattern. About 2% of male rats developed heart tumours at the lower exposure level, and 5–6% at the highest.

The second study came from Italy. The Ramazzini Institute in Bologna exposed 2,448 Sprague-Dawley rats — the largest lifetime study of its kind — to a 1.8 gigahertz GSM signal at field strengths of 0, 5, 25 and 50 volts per metre, for 19 hours a day, from before birth until natural death. The design was deliberately different from the NTP's: instead of a phone held to the head, it simulated the far-field exposure of living near a base station. The result, published in Environmental Research in 2018, was a statistically significant increase in heart schwannomas in male rats at the highest dose — 8 of 207 animals (3.9%), against 3 of 412 (0.7%) in the unexposed controls. The authors' conclusion was direct: their findings were "consistent with and reinforce the results of the NTP study", because both reported increases in tumours of the brain and heart in the same strain of rat. And the dose matters: 50 volts per metre works out to about 6.6 watts per square metre, which is below the international reference level for that frequency — 9 watts per square metre. The Ramazzini rats were not being cooked. They were exposed at levels the guidelines permit.

That is the strongest case the critics have, and it deserves to be stated in full before it is answered. Here is the answer, also in full.

First, the NTP study itself warned against the extrapolation its critics make. John Bucher, the NTP's senior scientist, said at the time that "the levels and duration of exposure were much greater than what people experience with even the highest level of cell phone use, and exposed the rodents' whole bodies. So, these findings should not be directly extrapolated to human cell phone usage." The rats were bathed in radiation for nine hours a day, whole-body, for their entire lives, at levels up to four times the maximum a phone is allowed to emit — and a phone almost never emits at its maximum, which it only approaches when the signal is weak. The Ramazzini exposure was lower, but it was still 19 hours a day, every day, for a lifetime, starting before birth.

Second, the US Food and Drug Administration — the agency that asked for the NTP study in the first place — reviewed the final reports and disagreed with the NTP's own conclusion. Its statement was blunt: "we disagree with the conclusions of their final report regarding 'clear evidence' of carcinogenic activity", and "the totality of the available scientific evidence continues to not support adverse health effects in humans caused by exposures at or under the current radiofrequency energy exposure limits." The FDA's reasoning was that the tumour increases appeared only at the highest doses, there was no clean dose-response pattern, the findings did not reproduce in female rats or in mice, and the study was not designed to test the safety of phone use in humans.

Third, there is the awkward fact that the exposed rats lived longer. In the NTP study, the male rats exposed to the highest levels of radiation outlived the unexposed controls — a result the researchers attributed to a lower rate of chronic kidney disease, the usual cause of death in ageing laboratory rats. Longer life means more time to develop tumours, which complicates the reading of the tumour counts. It does not erase the finding, but it is the kind of confound that makes a "clear evidence" label harder to defend than it looks in a headline.

Fourth, and most importantly, the animal findings have not been matched by human evidence. Two independent studies finding the same rare tumour in male rats is a real signal, and it is the reason research continues. But fifteen years of human studies since the IARC classification — including the 63-study WHO review of 2024 — have not produced a matching signal in people. A rat is not a small human, and a whole-body, nine-hour-a-day exposure is not a phone call. The honest position is that the animal studies raise a question that has not been fully answered, not that they answer it.

The safety limits: who wrote them, and what they actually protect against

The most common criticism of the safety limits is also the most reasonable one: they were written to stop phones and base stations from heating you up, and heating is not the only thing people worry about. The critics are right about the design of the limits. The question is whether that matters.

The limits most of the world uses come from the International Commission on Non-Ionizing Radiation Protection (ICNIRP), an independent body formally recognised by the WHO. Its current radiofrequency guidelines were published in March 2020 in the journal Health Physics, replacing the 1998 guidelines after seven years of work, and they cover 100 kilohertz to 300 gigahertz — which explicitly includes 5G. For the general public, the basic restrictions are a whole-body specific absorption rate (SAR) of 0.08 watts per kilogram averaged over 30 minutes, and a head-and-torso SAR of 2 watts per kilogram over 10 grams averaged over 6 minutes. Above 6 gigahertz — the millimetre-wave band — the guidelines switch to absorbed power density, limited to 20 watts per square metre over 4 square centimetres, because the energy is absorbed close to the skin surface. The limits carry safety margins: a reduction factor of 50 for whole-body exposure and 10 for local exposure, meaning the permitted level is set far below the level at which harmful heating has actually been observed.

The United States uses limits set by the Federal Communications Commission in 1996, which cap phone SAR at 1.6 watts per kilogram over 1 gram of tissue. Those limits are older than the ICNIRP 2020 guidelines, and the critics are right to point that out — they predate 5G, predate smartphones, and have not been revised since the year the first clamshell phones appeared. The FCC reviewed them in 2019 and decided to keep them, on the basis that the evidence did not justify a change. Whether that was caution or complacency is a matter of judgement, but the age of the American limits is a fact, and it is a legitimate complaint.

What the limits do not do is set out to protect against long-term, low-level, non-heating effects, because the scientific bodies that write them have concluded — after reviewing thousands of studies — that no such effects have been substantiated. That is the circularity the critics attack: the limits only cover heating because only heating has been proven, and only heating has been proven because the limits only cover heating. The honest answer to that attack is that it is partly true as a description of the logic, and that the way out of the circle is more research, not more certainty. The WHO's own position reflects exactly that: no adverse effect has been causally linked to wireless technologies to date, while research continues, including a WHO-led systematic review programme on radiofrequency health effects.

The other side of the limits question is what they look like in practice, and here the defenders have the strongest numbers in the whole debate. The UK regulator Ofcom has been measuring electromagnetic field levels around mobile base stations for years, and it began measuring 5G sites specifically in 2020. In its first round, at 16 sites in 10 cities chosen for the highest usage, the maximum measured level from all mobile signals combined was about 1.5% of the ICNIRP reference level. The highest level from 5G signals specifically was 0.039% of the limit — roughly 2,500 times below it. A later round covering 33 sites found the highest total level at any site was 7.1% of the reference level, with the next highest at 1.5%, and the 5G contribution still a small fraction. The UK Health Security Agency's assessment is that overall exposure is expected to remain low relative to the guidelines and "there should be no consequences for public health". The numbers are not close to the limit. They are not in the same neighbourhood as the limit.

What was measured	Where and when	Result vs the safety limit
All mobile signals at 5G sites	Ofcom, 16 sites, 10 UK cities, Feb 2020	Maximum 1.5% of the ICNIRP reference level
5G signals specifically	Ofcom, same 16 sites, Feb 2020	Maximum 0.039% of the ICNIRP reference level
All mobile signals at 5G sites	Ofcom, 33 sites, 18 towns and cities, 2020–21	Highest 7.1%, next highest 1.5% of the reference level
5G contribution	Ofcom, 33 sites, 2020–21	0.04% of the ICNIRP reference level at most
ICNIRP 2020 public limit, head and torso	International, 100 kHz–6 GHz	2 W/kg SAR over 10 g, 6-minute average
ICNIRP 2020 public limit, above 6 GHz	International, millimetre-wave band	20 W/m² absorbed power density over 4 cm²
FCC public limit, phones	United States, set 1996	1.6 W/kg SAR over 1 g of tissue

The gap between the measured levels and the limits is so large that it survives even the critics' best argument. Suppose the limits are wrong, and suppose the real threshold for harm is a hundred times lower than ICNIRP thinks. The measured 5G levels would still be below it. That is not a proof of safety — no measurement can prove a negative — but it is a reason the burden of proof sits where it does. The people claiming harm have to explain how an effect appears at levels thousands of times below the threshold at which any effect has ever been observed.

The symptoms people report — and what the blind tests show

Some people are convinced that 5G, Wi-Fi and mobile signals make them ill: headaches, fatigue, skin tingling, insomnia, difficulty concentrating. They call it electromagnetic hypersensitivity, or EHS, and their suffering is real. The question is whether the cause is what they think it is, and on that question the evidence is unusually clear.

The World Health Organization's position is that there is no scientific basis to link EHS symptoms to electromagnetic field exposure. That conclusion rests on a particular kind of experiment: the double-blind provocation study. Volunteers who report EHS are exposed to real electromagnetic fields and to sham exposure — identical in every way except that the field is switched off — without knowing which is which. Again and again, the results are the same: participants cannot tell the real exposure from the sham at a rate better than chance, and their symptoms appear just as often under sham exposure as under real exposure. If the fields were causing the symptoms, the people who are most sensitive to them should be able to detect them. They cannot.

That does not mean the symptoms are imaginary. The WHO's fact sheet on the subject is careful on this point: the symptoms are real and can be disabling, and people who experience them need medical care. What the evidence points to is a different cause — and the leading explanation is the nocebo effect, the mirror image of the placebo effect, in which the belief that something is harmful produces real physical symptoms. In one well-known line of experiments, people who believed they were being exposed to Wi-Fi signals developed headaches and skin sensations even when the router in the room was switched off. The belief did the work, not the signal. That is not a dismissal of the sufferers. It is an explanation that fits the data, and it is the reason health agencies treat EHS as a real condition with a cause that is not the field itself.

There is a second, simpler reason the symptom reports do not track the technology. People started reporting these symptoms decades before 5G existed — with 2G phones, with Wi-Fi routers, with power lines, with the old cathode-ray television. The symptoms follow the fear, not the frequency. When a new technology arrives, the same cluster of symptoms attaches itself to the new name. That is a pattern of human psychology, and it is worth knowing about, because it explains why the argument feels so personal: people who feel ill are told by experts that the thing they blame is not the cause, and they hear that as being told they are not ill at all. They are ill. The field is simply not what is making them ill.

The COVID conspiracy and the burning towers

No honest account of the 5G debate can skip April 2020, because it is the month the argument left the internet and started setting things on fire.

The theory had two versions. The first was that 5G radiation suppressed the immune system, making people more vulnerable to the coronavirus. The second, stranger one was that the virus itself could somehow travel, or "communicate", through 5G radio waves. Both were debunked in detail by the independent fact-checking organisation Full Fact, which pointed out that viruses are biological particles that spread through droplets and contact, not through radio waves, and that there is no mechanism by which radio waves could suppress immunity at the levels involved. The UK government said there was "absolutely no credible evidence" of a link. Cabinet Office minister Michael Gove called the theory "dangerous nonsense". NHS England's national medical director, Stephen Powis, called it fake news with no scientific backing that risked damaging the emergency response.

It did not matter. In the first week of April 2020, at least 20 mobile phone masts across Britain were set on fire or vandalised, in Birmingham, Liverpool, Belfast and Greater Manchester. One of the first, a BT tower in Birmingham, provided 2G, 3G and 4G services to thousands of people and had no 5G capability at all. Telecoms engineers were abused in the street while trying to keep the network running during a pandemic. The UK's four mobile operators — EE, O2, Three and Vodafone — issued a joint statement: "There is no scientific evidence of any link between 5G and coronavirus. Fact." The attacks continued anyway, and they were not confined to Britain: towers were attacked in the Netherlands, Ireland, Australia, New Zealand and elsewhere, and the theory spread in every language.

Where did it come from? The trail, traced by Wired, leads to an interview in the Belgian newspaper Het Laatste Nieuws with a general practitioner, Kris Van Kerckhoven, who suggested a link between 5G and the virus. The newspaper later withdrew the article. By then it was far too late: the claim had been clipped, translated, reposted and remixed across Facebook, YouTube and WhatsApp, picking up new details at every step, until it reached people who had never heard of 5G and were suddenly certain it was killing them.

The tower burnings are the reason the 5G debate is treated differently from other health debates, and it is worth being honest about why. The conspiracy theories were false, and they did real damage: they destroyed infrastructure people depended on during a pandemic, they endangered engineers, and they diverted attention from a virus that was actually killing people. But the burnings also did something else. They gave the defenders of 5G a reason to stop listening. Once the other side is burning towers, it is easy to file the entire sceptical position under "conspiracy theory" and stop engaging with the parts of it that are not conspiratorial at all — the rat studies, the IARC classification, the age of the FCC limits, the thin long-term data on millimetre waves. Those are not conspiracy theories. They are evidence and open questions, and they survive the debunking of the COVID claim completely untouched.

Brussels, and the city that said no

The critics' other favourite piece of evidence is a real political event: in April 2019, the government of Brussels halted its 5G pilot programme. The city's environment minister, Céline Fremault, said the people of Brussels would not be "guinea pigs" and that 5G plans would remain on hold as long as radiation from the new antennas could not be accurately measured. Brussels has some of the strictest telecom radiation rules in the world — a limit of 6 volts per metre, against the roughly 58 volts per metre the international guidelines allow at mobile frequencies — and the Belgian telecom regulator had concluded that no new radio frequency could be brought into service unless the city relaxed its rules. Ministers had approved raising the limit to 9 volts per metre indoors and 14.5 outdoors; Fremault halted the pilot anyway, because the massive MIMO antennas 5G uses steer their beams dynamically, and she was not satisfied that their emissions could be measured reliably enough to guarantee compliance.

The Brussels story is routinely quoted as proof that a European capital banned 5G because it was dangerous. That is not what happened. Science Feedback, reviewing the claim in 2020, found it inaccurate: the halt was about measurement and compliance with the city's own unusually strict rules, not about any demonstrated health effect, and the city was expected to welcome 5G once the measurement question was resolved. But the story is still worth telling, because it captures something real about the debate: a democratic government looked at a new technology, looked at the uncertainty around measuring it, and decided to wait. That is a reasonable response to uncertainty. It is not the same as proof of harm, but it is also not the same as proof of safety, and the difference between those three things is where most of the argument lives.

Brussels was not alone in hesitating. In 2017, a group of 170 scientists appealed to the European Union to halt the 5G rollout pending further research, and several Swiss cantons imposed temporary moratoria. None of those pauses was based on a demonstrated health effect. All of them were based on the same judgement: that a technology being rolled out at this scale, with this little long-term data on its newest frequencies, deserved a slower pace. Reasonable people can disagree about whether that judgement was right. What they cannot do is pretend it never happened.

What the critics get right — the honest gaps

A fair verdict has to give the critics their due, because the case for "5G is safe" has real gaps in it, and pretending otherwise is what feeds the conspiracy theories.

First, the long-term data on millimetre waves is thin. Sub-6 gigahertz 5G uses frequencies close to ones that have been studied for decades, but millimetre-wave 5G — 24 to 40 gigahertz — is genuinely new at population scale, and the longest human exposures to it are only a few years old. The WHO's reviews cover radiofrequency fields broadly, but no study can tell you what thirty years of millimetre-wave exposure does, because thirty years have not happened yet. The defenders' answer — that the physics gives no reason to expect a new effect, because the energy is absorbed in the skin and cannot reach deeper tissue — is a good argument from principle. It is not the same as data, and the critics are right to say so.

Second, the limits are built around heating. ICNIRP's guidelines are designed to prevent tissue heating, and the safety margins are calculated against heating thresholds. If there are non-thermal biological effects at low levels — changes in cell signalling, oxidative stress, effects on the nervous system — the limits were not designed to catch them. Hundreds of studies have reported such effects, and the mainstream reviews have generally concluded that they are inconsistent, unreplicated, or too small to matter. But the critics are right that the reviews and the limits share an assumption, and that assumption has never been tested by a study designed to falsify it at population scale.

Third, the American limits are old. The FCC's phone exposure limits date to 1996, and they were written for a world of voice calls, not of phones pressed to the body for hours a day streaming video. The FCC reviewed them in 2019 and kept them. That may be the right call — the ICNIRP 2020 guidelines, which are newer and cover 5G explicitly, set a head-and-torso limit of 2 watts per kilogram, and the American 1.6 watts per kilogram over a smaller volume is, if anything, stricter — but the optics of a limit older than the smartphone are bad, and they feed the suspicion that nobody is minding the store.

Fourth, the rat studies are real. They have not been explained away. The FDA's disagreement with the NTP's "clear evidence" label is a judgement call, not a refutation, and the Ramazzini study found its signal at levels below the international reference limit. The honest mainstream position is that the animal findings are a genuine unresolved signal that has not been reproduced in human studies. That is a thinner claim than "the rat studies prove nothing", and it is the claim the evidence actually supports.

Against those four gaps, set the four strongest facts on the other side. The energy cannot damage DNA, because it is non-ionising. The measured exposure from live 5G base stations is thousands of times below the limit. Sixty-three studies, reviewed for the WHO in 2024, found no link between phone use and brain cancer. And brain tumour rates have not risen with the explosion in phone use. The gaps are real, and so is the reassurance. The argument is not between evidence and ignorance. It is between a small, unresolved risk and a large, well-measured absence of harm — and the size of each is what people actually disagree about.

At a glance: what 5G does, and what it does not

Claim	What the evidence says	Verdict
5G radiation can damage DNA and cause cancer	5G is non-ionising: it cannot break chemical bonds or damage DNA. Cancer Research UK states radio waves do not have enough energy to damage DNA, and there is no established mechanism by which they could cause cancer	False
5G caused or spread COVID-19	Viruses spread through droplets and contact, not radio waves. Full Fact debunked both versions of the theory; the UK government said there was "absolutely no credible evidence" of a link	False
5G is classified as possibly carcinogenic	True: IARC classified radiofrequency fields as Group 2B in 2011, based on limited evidence. But 2B is the weakest positive classification, and the WHO-commissioned 2024 review of 63 studies found no association between phone use and brain cancer	True — and weaker than it sounds
Animal studies found tumours from radiofrequency exposure	True: the NTP (2018) found heart schwannomas in male rats with "clear evidence"; the Ramazzini Institute (2018) found the same tumour at lower, tower-like exposure. The FDA disagreed with the NTP's conclusion, and human studies have not reproduced the finding	True — a real unresolved signal
People living near 5G masts are exposed to dangerous levels	Ofcom measured 5G signals at a maximum of 0.039% of the international safety limit, and all mobile signals combined at a maximum of 1.5% (16 sites) to 7.1% (33 sites)	False
5G millimetre waves penetrate deep into the body	Above 6 GHz, the energy is absorbed in the outermost layers of skin and does not reach internal organs — which is why the safety guidelines switch to a surface measure above that frequency	False
5G causes headaches, fatigue and tingling (EHS)	Double-blind provocation studies repeatedly show people who report EHS cannot detect real exposure better than chance, and symptoms appear just as often under sham exposure. The WHO finds no scientific basis linking the symptoms to electromagnetic fields; the nocebo effect is the leading explanation	No evidence
Cities banned 5G because it was proven dangerous	Brussels halted its 2019 pilot over measurement and compliance with its own unusually strict 6 V/m limit — not because any health effect had been demonstrated	False
There are no long-term studies of 5G at all	Partly true: millimetre-wave 5G has only a few years of population exposure, so thirty-year data cannot exist yet. But the sub-6 GHz frequencies have decades of nearby-frequency research behind them, and the WHO review programme continues	Partly true
The safety limits are old and only cover heating	True: the US FCC limits date to 1996, and the ICNIRP guidelines are built around preventing tissue heating, with safety factors of 50 (whole body) and 10 (local). The ICNIRP guidelines were updated in 2020 and explicitly cover 5G	True

Claim vs evidence: the arguments, scored

Argument	Who makes it	Strength of the evidence
"5G cannot damage DNA"	WHO, Cancer Research UK, ICNIRP	Strong — basic physics of non-ionising radiation, uncontested
"Measured exposure is thousands of times below the limit"	Ofcom, UK Health Security Agency	Strong — direct measurement at 16–33 live sites
"No link between phones and brain cancer"	WHO-commissioned review, 2024, 63 studies	Strong — largest synthesis to date; INTERPHONE signal not reproduced
"The rat studies found tumours"	NTP 2018, Ramazzini 2018	Moderate — real, peer-reviewed, same tumour twice; but whole-body lifetime exposure in rats, not humans
"IARC says possibly carcinogenic"	IARC, 2011	Moderate — real classification, but 15 years old and based on evidence that has since weakened
"Limits only cover heating"	Critics of ICNIRP/FCC	Moderate — true of the design, but no non-thermal effect has been substantiated at low levels
"5G suppresses immunity / spreads COVID"	Conspiracy threads, 2020	None — debunked by Full Fact, WHO, UK government; no mechanism, no evidence
"EHS proves the fields make people ill"	EHS advocacy groups	Weak — symptoms real, but blind tests show no link to actual exposure

What people actually say: how divided the argument really is

The 5G debate is not a polite disagreement between experts. It is one of the most furious arguments on the internet, and the popular-opinions section is where that fury lives. What follows is paraphrased from the recurring positions in real threads — on Reddit, Quora and the comment sections of news articles — not invented quotes. No usernames, no vote counts, no fabricated comments: just the arguments people actually make, in the shape they actually make them.

The engineers. The largest and loudest camp in the technical threads is the one that treats the whole debate as settled physics. Their argument runs: radio waves are non-ionising, the energy levels are trivial, your microwave oven puts out more power, the sun is more dangerous, and anyone who thinks otherwise failed high-school physics. They post the Ofcom numbers, they post the electromagnetic spectrum, and they are genuinely baffled that the argument continues. Their weakness is tone: they answer a question about trust with a lecture about physics, and they treat the rat studies as a joke rather than as evidence. The people they are arguing with hear contempt, not reassurance, and the contempt becomes the story.

The tower-watchers. The second camp is the one that lives in the conspiracy threads, and it is not one camp at all but a federation: people who believe 5G is a surveillance grid, people who believe it is a weapon, people who believe it caused the pandemic, and people who believe all three at once. Their arguments are internally consistent and externally uncheckable — every debunking is proof of the cover-up, every expert is bought, every measurement is rigged. The tower burnings came from this camp's edges. The rest of the camp disowns the arson while sharing the posts that inspired it. Their weakness is that the theory has no end point: there is no evidence that could falsify it, which means it is not a hypothesis at all. It is a worldview.

The careful sceptics. The third camp is the smallest and the most interesting, because it is the one the other two camps both ignore. These are the people who quote the NTP study accurately, who know what Group 2B means and what it does not, who can tell you the difference between sub-6 and millimetre wave, and who conclude that the technology is probably safe but was rolled out faster than the evidence justified. They do not believe 5G caused COVID. They do not burn towers. They want longer studies, lower limits, and a slower rollout — and they are furious that the existence of the conspiracy camp lets everyone else dismiss them as conspiracy theorists. Their weakness is that their position is boring: "probably fine, but let's keep checking" does not trend, does not get shared, and does not win arguments against people who want a yes or a no.

The sufferers. The fourth camp is the people who feel ill and blame the signals. They describe headaches that start when a new mast goes up, sleep that breaks when the router is on, a tingling in the hands that follows them from room to room. They are not performing. They are not grifting. They are frightened, and they have found a community that believes them and an explanation that fits their experience. The blind-test evidence says the cause is not the field — but the blind-test evidence has never once made a sufferer feel better, because the suffering is not about the field. It is about being unheard. This is the camp the debate fails most completely, because both sides use it: the sceptics cite it as proof, the defenders cite the blind tests as dismissal, and nobody in the argument is actually helping the people who are ill.

The pragmatists. And then there is the largest camp of all, which never posts: the people who bought a 5G phone because it was the phone on sale, who have never thought about the question, and who will only ever hear about it when a mast goes up on their street. They are the audience this article is written for. They do not need a lecture, and they do not need a conspiracy. They need the record, stated plainly, with the gaps left in.

The verdict

So: is 5G dangerous? On the evidence available in 2026, the answer is no — not at the levels anyone is actually exposed to, and not by any mechanism anyone has been able to demonstrate. The physics is on the defenders' side: non-ionising radiation cannot damage DNA. The measurements are on the defenders' side: live 5G signals run thousands of times below the safety limit. The epidemiology is on the defenders' side: the largest review ever conducted found no link between mobile phone use and brain cancer, and brain tumour rates have not risen with phone use. And the conspiracy theories are false in the most complete sense a claim can be false: 5G did not cause COVID-19, it does not suppress immunity, and the towers that were burned were mostly not even 5G.

But the verdict has a second sentence, and the second sentence is the reason this article is six thousand words long instead of six. The evidence is not complete, and the people who say it is are overstating their case. Two large animal studies found the same rare tumour. The WHO's own cancer agency has never withdrawn its "possibly carcinogenic" classification. The newest frequencies have only a few years of population exposure behind them. The American safety limits are thirty years old. None of that makes 5G dangerous. All of it makes the claim "the science is settled" a little less true than it sounds — and it is the gap between "settled" and "mostly settled" that the entire argument lives in.

The practical advice follows from that. There is no reason to avoid 5G, to move house because of a mast, or to throw away a 5G phone. The exposure from a base station is a tiny fraction of the limit, and the exposure from the phone in your hand is the largest part of your total exposure — which is why the simplest, cheapest precaution, if you want one, is the one the health agencies actually suggest: use hands-free or speakerphone for long calls, and do not sleep with the phone under your pillow. Not because the evidence says you must, but because it costs nothing, and because it is the one exposure you control. Beyond that, the honest position is to keep watching: the WHO's review programme is ongoing, the IARC classification is overdue for re-examination, and the millimetre-wave data will keep accumulating for decades. The science is not finished. It is, on the present evidence, reassuring — and reassurance that admits its limits is worth more than reassurance that pretends there are none.

Frequently asked questions

Is 5G dangerous to live near?

No, on the current evidence. The UK regulator Ofcom has measured the signals around live 5G base stations and found the highest level from 5G itself was 0.039% of the international safety limit — roughly 2,500 times below it. All mobile signals combined at those sites reached at most 1.5% of the limit in the first round of testing, and 7.1% at one site in a later round. The exposure from a mast falls rapidly with distance, and the phone in your hand is a far larger source of exposure than any mast.

Does 5G cause cancer?

There is no established evidence that it does. 5G uses non-ionising radio waves, which cannot damage DNA. The WHO's cancer agency classified radiofrequency fields as "possibly carcinogenic" (Group 2B) in 2011, but that is the weakest positive classification, and a WHO-commissioned review of 63 studies published in 2024 found no association between mobile phone use and brain cancer. Two animal studies did find tumours in rats, but human studies have not reproduced the finding.

Is 5G worse than 4G?

Not in any way the evidence supports. Most 5G runs on frequencies just above 4G's, and the newer millimetre-wave band (24–40 GHz) is absorbed in the outermost layers of skin and does not reach internal organs. 5G base stations are smaller and lower-powered than the big 4G masts, and the measured exposure from 5G is a tiny fraction of the safety limit — smaller, in Ofcom's measurements, than the exposure from the older networks at the same sites.

Did 5G cause COVID-19?

No. COVID-19 is caused by the SARS-CoV-2 virus, which spreads through droplets and contact. Radio waves cannot carry, create or spread a virus. The theory was debunked by Full Fact, the WHO and the UK government, and the towers attacked in Britain in April 2020 were mostly 3G and 4G — one of the first burned, a BT tower in Birmingham, had no 5G capability at all.

Why do some people get headaches near 5G and Wi-Fi?

The symptoms are real, but the cause is almost certainly not the signal. In double-blind provocation studies, people who report electromagnetic hypersensitivity cannot tell real exposure from sham exposure at a rate better than chance, and their symptoms appear just as often when the field is switched off. The WHO finds no scientific basis linking the symptoms to electromagnetic fields, and the leading explanation is the nocebo effect — the belief that something is harmful producing real physical symptoms.

Should I worry about the rat studies?

They are worth knowing about, but they are not a reason to change how you live. The NTP study found heart tumours in male rats exposed whole-body for nine hours a day at up to four times the maximum phone emission level; the Ramazzini study found the same tumour at lower, tower-like exposure. The FDA reviewed the NTP findings and disagreed with its "clear evidence" conclusion, and fifteen years of human studies have not found a matching effect in people. The honest position is that the animal studies are a genuine unresolved question — not proof of harm, and not proof of nothing.

Sources

• World Health Organization — "To date, no adverse health effect has been causally linked with exposure to wireless technologies"; WHO Q&A on 5G mobile networks and health (position summarised at factscheck.co.uk and Ericsson's radio waves and health Q&A)
• International Commission on Non-Ionizing Radiation Protection (ICNIRP) — Guidelines for limiting exposure to electromagnetic fields (100 kHz to 300 GHz), Health Physics, March 2020 (limits and 5G coverage summarised at airwaves.cold.fyi and ARPANSA)
• Karipidis et al. — WHO-commissioned systematic review of 63 studies on mobile phone use and cancer, Environment International, 2024: no association between mobile phone use and brain cancer
• International Agency for Research on Cancer (IARC) — Press Release No. 208, 31 May 2011: radiofrequency electromagnetic fields classified as possibly carcinogenic to humans (Group 2B) (IARC press release; summary at EurekAlert)
• National Toxicology Program — Technical Reports TR-595 and TR-596, November 2018: clear evidence of heart schwannomas in male rats, some evidence of brain and adrenal tumours (NIH news release; NTP programme page)
• US Food and Drug Administration — statement disagreeing with the NTP's "clear evidence" conclusion, November 2018 (Medscape report; Healio report)
• Falcioni L et al. — Report of final results regarding brain and heart tumors in Sprague-Dawley rats exposed to mobile phone radiofrequency field representative of a 1.8 GHz GSM base station environmental emission, Environmental Research 165 (2018) 496–503 (PubMed 29530389)
• Ofcom — EMF measurements near 5G mobile phone base stations, technical report, February 2020 and March 2021 update: 5G at 0.039% of the ICNIRP reference level (Ofcom summary of results; Mobile UK summary; The Register)
• UK Health Security Agency — 5G technologies: radio waves and health: overall exposure expected to remain low relative to guidelines
• Full Fact and UK government — debunking of the 5G–COVID-19 theories; DCMS: "absolutely no credible evidence" of a link; Michael Gove: "dangerous nonsense" (Gadgets 360 / Reuters; 5G.co.uk; Tom's Guide)
• Brussels 5G pilot halt, April 2019 — environment minister Céline Fremault; measurement and compliance with the city's 6 V/m limit, not a demonstrated health effect (Mobile Europe; Fierce Network; Science Feedback review)
• World Health Organization — Fact Sheet on electromagnetic hypersensitivity: no scientific basis linking EHS symptoms to electromagnetic field exposure; symptoms real, cause not the field
• Cancer Research UK — radio waves do not have enough energy to damage DNA; mobile phone use does not increase cancer risk

More from the Unanswered Questions series: If Poison Expires, Does It Become Safe? · Is MSG Bad for You? · Is Organic Food Actually Healthier? · Are Seed Oils Toxic? · Is Fluoride in Tap Water Safe? · Does Red Meat Really Cause Cancer? · Does Sugar Really Make Kids Hyperactive? · Is Aspartame Safe? · Is Vaping Really Safer Than Smoking? · Are Eggs Bad for Your Cholesterol? · Are GMOs Safe to Eat? · Is Bottled Water Safer Than Tap Water? · Is Breakfast Really the Most Important Meal of the Day? · Do Vaccines Cause Autism? · Is Intermittent Fasting Actually Dangerous? · Is Microwaving Food Bad for You?

Disclaimer: this article is general information, not medical advice. It summarises published research and the positions of public health agencies as of November 2026. If you have a health concern, speak to a qualified clinician.