It is important to emphasise that it is not ESB Networks who ultimately decide on safe levels of exposure from smart meters.  That assessment is made by completely independent international experts, and ESB Networks follow their conclusions.  This is as it should be.  We at ESB Networks are the company who are installing the smart meters.  We have to satisfy ourselves that they are safe and we have indeed satisfied ourselves of that before starting to install them.  But the public should expect assurances of safety from bodies independent of us. 
That comes in the form of exposure limits.  Exposure limits for the protection of the public are set by authoritative independent expert review groups.  We at ESB Networks have no influence whatsoever over the setting of those exposure limits.  We simply guarantee that every smart meter we install (and, for that matter, all our other equipment and operations) complies with those exposure limits.

The main international group setting exposure limits is the International Commission on Non-Ionizing Radiation Protection (ICNIRP).  They are made up of experts from various disciplines, both academics and representatives of national protection bodies, with strict rules to prevent any industry involvement at all.
ICNIRP work closely with the World Health Organization (WHO).  There is also a USA-based body called the International Commission on Electromagnetic Safety (ICES) who publish exposure limits.  As explained in the next paragraph, in Ireland it is the ICNIRP rather than the ICES limits that apply, but as they are fairly similar anyway, this doesn’t make much difference.
As already stated, ICNIRP publish exposure limits, and they have authority by virtue of coming from the International Commission.  The European Union (EU) have then created a type of document called a Recommendation, which takes the limits from ICNIRP and says that they should apply in EU member states, thus giving them a legal status in the EU.  The Irish government has confirmed that it accepts this Recommendation and that the limits set by ICNIRP do indeed apply in Ireland.
All this happens completely independently of ESB Networks.  We simply guarantee that all our smart meters will comply with these exposure limits (in fact, they comply by quite a large margin).

The relevant exposure limits were published by ICNIRP in 1998.  They were published in a scientific journal called Health Physics, but are also available from ICNIRP’s website.
ICNIRP GUIDELINES FORLIMITING EXPOSURE TO TIME-VARYING ELECTRIC, MAGNETIC AND ELECTROMAGNETIC FIELDS(UP TO 300 GHZ) PUBLISHED IN: HEALTH PHYSICS 74 (4):494-522; 1998
The EU Recommendation that adopted these exposure limits was published in the Official Journal of the EU.
COUNCIL RECOMMENDATION of12 July 1999 on the limitation of exposure of the general public to electromagnetic fields (0 Hz to 300 GHz), 1999/519/EC 
The Department responsible for policy in this area in Ireland is the Department of Communications, Climate Action and Environment.  See their statement on their website.

The exposure limits come in two parts.
Firstly, there is the “basic restriction”.  That is the limit on the quantity of direct biological relevance, the power deposited in tissue by the radiofrequencies, known as the Specific Absorption Rate (SAR). That is measured in watts per kilogram (W/kg). The limit is:

• 0.08 W/kg for the average over the whole body
• 2 W/kg for localised exposure to any 10 g of tissue in the head and trunk
• 4 W/kg for localised exposure to any 10 g of tissue in the limbs

Each of these is averaged over 6 minutes.
Secondly, there is the “reference level”. This is measured in watts per square metre (W/m²) and gives the power in the radiofrequency wave that would be needed in order to produce the Specific Absorption Rate specified in the basic restriction. So, if you know that the power in the radiofrequency transmission is below the reference level, you know that you are also below the basic restriction. The reference level varies with frequency, but for the two commonest radiofrequencies, it is:

• 4.5 W/m² at 900 MHz
• 9 W/m² at 1800 MHz

(The reference levels can also be expressed as an electric field in V/m, but for simplicity, we stick to W/m² here.)
The significance of the reference level is that it is often a lot easier to measure than the basic restriction, so we often assess compliance using the reference level. If we are very close to a source of exposure, the power density (as given in the reference level) is not very helpful, so we would probably have to use the SAR and the basic restriction. But once we are a reasonable distance away from the source, we would usually use the reference level to assess compliance.
It is important to note that these exposure limits are set so as to include a large safety margin. And these are the limits for the general public. There are higher limits for workers – the limits for the public are lower partly to take account of the greater variability in health status among the general population.

 

ESB Network’s smart meters fully comply with international exposure limits.
Even in the worst case, if your head was very close to the meter while it was transmitting, smart meters comply with the exposure limits. That is, of course unlikely, but it is theoretically possible. That would be just like holding a mobile phone against your head. The exact exposure would depend on the exact location and orientation of the head relative to the smart meter, and can only be assessed by quite detailed numerical modelling or by laboratory tests. But that modelling has been done for mobile phones. The relevant limit is the limit for localised exposure in the head and trunk, which is a Specific Absorption Rate (SAR) of 2 W/kg. Mobile phones generally produce maximum SARs from around 0.5 – 1.5 W/kg, depending on the model in question. All mobile phones comply with the limits, and hence the smart meters will also comply, even in that unlikely scenario of putting the head right against the meter.

In practice, smart meters comply by a much larger margin, because they are usually further away, and because the power levels are often less than the maximum.

The exposure from a smart meter drops rapidly the further away from it you are. If we take 1m as a typical distance from a smart meter, that reduces the exposure by a factor of 20 or more. And when we take into account that the meters rarely need to transmit at full power, that reduces the exposure further.

So, overall, smart meters will typically comply with the exposure limits by a considerable margin. Combined with the margin of safety already built into the exposure limits, this provides a large measure of confidence in their safety.

The current exposure limits originate from the independent body ICNIRP. Following extensive consultations about revising their exposure limits over several years, ICNIRP issued new limits in 2020. 
The new limits take advantage of new scientific understanding to make quite a few detailed changes. But these are, mainly, changes to the details. The core numerical values of the limits relevant to smart meters have not changed. Smart meters were compliant with the 1998 limits and are still compliant with the 2020 limits.
Our commitment at ESB Networks is that we will comply with whatever the current exposure limits are. If the exposure limits ever changed again, we would make sure smart meters complied with the new limits.   

There are groups that lobby for much lower exposure limits.  If that were ever to happen, it would cause problems mainly for mobile phones, wifi, etc.   Equally, there are scientists who are very dismissive of any possible risks from radiofrequencies at all.  Everyone is of course entitled to their opinion.  But what is relevant here is not the opinions of any individuals, but the considered views of authoritative experts, such as the International Commission on Non-Ionizing Radiation Protection (ICNIRP).