While the WHO’s EMF review has been mired in controversy, France has been quietly doing its own thing. ANSES — the French Agency for Food, Environmental and Occupational Health & Safety — just published a detailed paper outlining how they evaluated 5G health risks across all three frequency bands.
Their approach was methodical, their conclusions were measured, and their findings tell us something important about where the science actually stands.
Who Is ANSES and Why Should You Care?
ANSES is France’s equivalent of a combined FDA + EPA for health and environmental safety. They’re the agency that decides whether something poses a risk to the French public — from pesticides in food to electromagnetic fields from wireless networks.
Unlike advocacy groups on either side of the EMF debate, ANSES has no financial stake in 5G succeeding or failing. Their mandate is public health protection, and they’ve been studying RF-EMF health effects for over a decade, publishing major reports on radiofrequencies and health (2013), RF and children (2016), electromagnetic hypersensitivity (2018), and phones worn close to the body (2019).
This new paper, published in Annals of Telecommunications (2026), describes their methodology for assessing 5G specifically.
The Three-Band Approach
Here’s what makes ANSES’s approach smart: they didn’t treat “5G” as one thing. They broke it into three separate assessments based on frequency band, because the physics and biology are genuinely different:
Band 1: 700–2100 MHz — “Nothing New Here”
ANSES conclusion: Existing knowledge from previous mobile technologies is relevant. No new health risk identified.
This is the low-band 5G that most people encounter day-to-day. These are the same frequencies used by 3G and 4G for years — 5G just uses them more efficiently with better encoding and scheduling.
ANSES’s logic is straightforward: we have decades of research on these frequencies from earlier mobile generations. 5G at 700-2100 MHz doesn’t introduce fundamentally different exposure. The existing body of evidence — including the massive Interphone study, the Danish cohort study, and numerous animal experiments — applies directly.
What this means for you: If you’re using a 5G phone on low-band spectrum, your exposure profile is essentially identical to what 4G users experienced. The research base is substantial and, according to ANSES, sufficient.
Band 2: 3.5 GHz — “Probably Fine, But We’re Extrapolating”
ANSES conclusion: Limited specific data, but extrapolation from lower frequencies suggests no new health risk.
This is mid-band 5G — the “sweet spot” frequency that most 5G networks rely on for the balance of speed and coverage. It’s also relatively new: 3.5 GHz wasn’t widely used for mobile communications before 5G.
ANSES acknowledges a data gap here. There are far fewer studies specifically at 3.5 GHz compared to the 900 MHz or 1800 MHz frequencies researchers have studied for decades. However, the physics of RF absorption doesn’t change dramatically between 2.1 GHz and 3.5 GHz — the penetration depth decreases somewhat, but the fundamental interaction with tissue is similar.
Their assessment: the existing evidence can be extrapolated to 3.5 GHz with reasonable confidence. No evidence suggests this band behaves differently from a health perspective.
What this means for you: Your mid-band 5G exposure is likely safe based on what we know about similar frequencies, but dedicated research at exactly 3.5 GHz is still catching up. Studies like the INERIS controlled human exposure trial and Bektas fertility study are starting to fill this gap.
Band 3: 26 GHz (Millimeter Wave) — “We Don’t Know Enough”
ANSES conclusion: Effects are mainly localized to superficial tissues. Current data remain insufficient for definitive conclusions.
This is the honest and important part. Millimeter wave 5G at 26 GHz (and the related US band at 28 GHz) is genuinely new territory. These frequencies barely penetrate the body — energy is absorbed in the outer layers of skin and the surface of the eyes.
ANSES found that while the physics limits exposure to superficial tissues (a good thing — no deep organ exposure), the research base for these specific frequencies is thin. We simply don’t have enough studies on long-term skin and eye exposure to millimeter waves at the power levels used by 5G.
The agency explicitly called for:
- Enhanced exposure characterization — better measurement of how much mmWave energy people actually encounter
- Targeted experimental research — specifically for skin and eye effects at 26 GHz
What this means for you: If you’re in the US or parts of Asia where mmWave 5G is deployed (Verizon’s Ultra Wideband, for example), the exposure is shallow but not yet fully characterized. ANSES isn’t saying it’s dangerous — they’re saying the jury is still out. Our coverage of the 26 GHz human exposure study showing no stress biomarker changes is one of the early data points here.
Check your EMF exposure
See cell towers, power lines, and substations near any US address.
Search Your AddressHow This Compares to Other Agencies
ANSES’s conclusions land in a similar place to other national health agencies, with some noteworthy variations:
| Agency | Low-Band 5G | Mid-Band 5G | mmWave 5G |
|---|---|---|---|
| ANSES (France) | No new risk | No new risk (extrapolated) | Insufficient data |
| ARPANSA (Australia) | No evidence of harm | No evidence of harm | Limited data above 6 GHz |
| Health Canada | No health risk at compliant levels | No health risk at compliant levels | Same position |
| Sweden (SSM) | No established health effects | No established health effects | Research needed |
| Netherlands Health Council | No additional health risks expected | No proven health risks | Insufficient basis for conclusions |
The consensus pattern: low and mid-band 5G gets a pass from every major health agency. Millimeter wave gets a cautious “we need more research.”
Note what’s not in this table: the WHO, which is still working on its controversial RF-EMF health risk assessment with no publication date in sight.
What ANSES Gets Right
1. Frequency-specific assessment. Treating all 5G frequencies as equivalent would be scientifically lazy. A 700 MHz signal and a 26 GHz signal interact with human tissue in fundamentally different ways. ANSES’s three-band approach reflects reality.
2. Honesty about data gaps. It would be easy to say “5G is safe” and leave it at that. Instead, ANSES explicitly identifies where evidence is strong (low-band), where they’re extrapolating (mid-band), and where they genuinely don’t know enough (millimeter wave). That’s scientific integrity.
3. Separation from industry. ANSES’s funding comes from the French government, not telecommunications companies. Their conflict-of-interest statement is clean. That matters when evaluating an agency’s credibility.
4. Actionable recommendations. Rather than vague calls for “more research,” ANSES specifically identifies what’s needed: exposure characterization and targeted experimental research on skin and eye effects at higher frequencies.
What’s Missing
No mention of non-thermal effects. ANSES’s assessment primarily addresses whether 5G stays within existing thermal safety limits. The broader debate about whether current safety limits are adequate — particularly regarding non-thermal biological effects — isn’t engaged.
No meta-analysis of specific health outcomes. Unlike the WHO’s approach (which attempted outcome-specific systematic reviews), ANSES’s paper describes their methodology more than their findings. It’s a framework paper, not a comprehensive evidence review.
No long-term data. 5G has only been deployed commercially since 2019-2020. No agency can assess long-term effects of something that hasn’t existed long enough. ANSES acknowledges this implicitly by calling for ongoing research.
The Bottom Line
France’s health agency delivers a nuanced, frequency-specific verdict on 5G:
- Low-band (700-2100 MHz): Safe, based on extensive existing evidence ✅
- Mid-band (3.5 GHz): Likely safe, based on reasonable extrapolation ✅
- Millimeter wave (26 GHz): Insufficient data for definitive conclusions ⚠️
This is about as honest and measured as a regulatory agency can be. They’re not dismissing concerns, and they’re not raising alarms. They’re telling you exactly where the evidence is strong and where it’s still developing.
For most people using 5G phones today — which overwhelmingly connect on low and mid-band frequencies — ANSES’s assessment is reassuring. For the smaller number of people with regular mmWave exposure, the answer is: we’re still studying it, and the early data is cautiously reassuring, but the research isn’t complete.
This is part of EMF Radar’s Study Spotlight series, where we break down the latest EMF research and policy developments in plain language. Want to check the RF environment near you? Try our free EMF map tool.
Paper: Attia D, Merckel O. “Health risk assessment of 5G technologies: the ANSES approach.” Annals of Telecommunications (2026). Read the paper
Related Reading
- Study Spotlight: Scientists Blasted People with 26 GHz 5G Millimeter Waves — Their Stress Hormones Didn’t Budge
- How Much 5G Radiation Are You Actually Exposed To? Greece’s Largest RF Measurement Study Answers
- Study Spotlight: Living Near Cell Towers Changed These People’s Blood — Here’s What Researchers Found
- Study Spotlight: Italian Researchers Blasted Brain Cancer Cells with 4G LTE and WiFi Simultaneously — Nothing Happened
Concerned about EMF? Check your address on EMF Radar to see nearby towers and power lines, or find a certified EMF consultant for professional testing.