Age Verification and Online Pornography under the DEA 2017: whose fine is it anyway?

May 21, 2017

The Digital Economy Act 2017 was passed by
Parliament at the end of April 2017, introducing a range of measures likely to
be of interest to IT lawyers, including rules around communications service
provision, IP, direct marketing, and digital government. Part 3 of the Act
contains the framework relating to access to online pornography by under 18s in
the UK..

Online pornography and age verification

The primary obligation relating to online
pornography is contained in s 14(1):

‘A person contravenes this subsection if
the person makes pornographic material available on the internet to persons in
the United Kingdom on a commercial basis other than in a way that secures that,
at any given time, the material is not normally accessible by persons under the
age of 18.’

The enforcer of Part 3, the newly-formed ‘age-verification
regulator’, can impose a financial penalty on such a person — administrative/civil,
rather than criminal, in nature — of up to the greater of £250,000 or 5% of
that person’s qualifying turnover.

Unfortunately, the Act does not clarify
what is meant by ‘makes … available on the internet’, leaving the reach of the
primary obligation unclear, both in terms of what it covers, and who it covers.

What is ‘making available’? Does it apply
to those who upload pornographic videos? Or those who live-stream them? What
about ‘porn sites’ — sites which are dedicated to the dissemination of
pornography? Social networks and online forums, with video upload capabilities?
What about internet service providers?

The scope of ‘makes … available’

A copyright perspective

‘Making available’ is one of the rights
reserved to authors under the WIPO copyright treaty, under Article 8. In 2011,
in the TV Catch Up case ([2010] EWHC 3063 (Ch)), the High Court considered the
right to control the ‘making available’ of a work to be a subset of the broader
right of ‘communication to the public’. This position was adopted by the CJEU
in the C More case (C-279/13) in 2015. The CJEU went on to state that:

‘in order to be classified as an act of
‘making available to the public’ within the meaning of [Article 3(2) of the
InfoSoc directive], an act must meet, cumulatively, both conditions set out in
that provision, namely that members of the public may access the protected work
from a place and at a time individually chosen by them.’

This right is, according to the CJEU, ‘intended
to refer to ‘interactive on-demand transmissions’. … That is not the case of
transmissions broadcast live on [the] internet…’.

If the scope of ‘making available’ under
the DEA 2017 was interpreted in this way, s 14 would cover only videos, audio
or photographs which were streamed on-demand to each individual user, and would
not appear to cover live transmissions of pornography, such as live-streaming
or webcam transmissions.

The more likely approach

It seems unlikely that Parliament intended
the Act to be limited in this manner or that a court would consider itself
constrained to follow the jurisprudence arising from the copyright use of the
term. Given that the overall intent of Part 3, according to the initial
explanatory notes, is to ‘restrict access to harmful sexualised content online’,
the type of transmission would seem immaterial.

Moreover, even if a court did adopt a
copyright-based approach, while it may give some clarity to what is meant by ‘making
available’, it does not assist in determining who it is that makes the content
available.

Who is a ‘person’?

In terms of who is in scope, let’s start
with the obvious. As well as a living individual — what most normal people
would consider to be a ‘person’ — by virtue of the Interpretation Act 1978, ‘person’
includes a ‘body of persons corporate or unincorporate’.

Individuals, companies, unincorporated
associations, all of these could fall within the prohibition.

Intermediary liability shields

Section 14(3) provides that ‘Regulations 17
to 20 and 22 of the Electronic Commerce (EC Directive) Regulations 2002 (S.I.
2002/2013) [the intermediary liability shields] apply in relation to this Part,
despite regulation 3(2) of those Regulations’.

Broadly speaking, these regulations shield
service providers from liability arising from content transmitted or uploaded
by a third party, where the service provider’s role is limited to ‘activity …
of a mere technical, automatic and passive nature, which implies that the
information society service provider has neither knowledge of nor control over
the information which is transmitted or stored’ (Recital 42, 2000/31/EC).

Internet access providers

An access provider is shielded from
liability for what it transmits, provided that it does not initiate the
transmission, select the receiver of the transmission, or select or modify the
information contained in the transmission.

As such, since the shield has been
expressly afforded to access providers under the DEA 2017, it seems safe to
assume that, even if their transmission of someone’s content means that they ‘make
available’ online pornography within the meaning of s 14(1), ISPs are excepted
from liability.

This does not mean that access providers
are home and dry though, as s 23 permits the regulator to compel ISPs to block
access to ‘offending material’: a statutory framework for an administrative
blocking order. An ISP served with such an order has a statutory duty to comply
with it.

Although Parliamentary debate focussed on
blocking access at retail ISP level — providers of connectivity to subscribers
in the UK attempting to prevent their customers from accessing the sites
notified to them — it is conceivable that the age-verification regulator might
attempt to impose a notice on the provider of connectivity to a pornography
provider’s server, stopping the server from connecting to the Internet, rather
than requiring each retail ISP to implement a block.

However, in such a case, considerable care
would need to be taken to avoid ‘overblocking’. While s 23(3) explicitly
permits an administrative blocking notice to ‘have the effect of preventing
persons in the United Kingdom from being able to access material other than the
offending material using the service provided by the internet service provider’,
the age-verification regulator’s actions must be consistent with the general
requirement of proportionality.

Website operators and social media
providers

In terms of website operators, and hosting
service providers, such as social network sites and operators of online forums,
the position is less clear.

As the Act expressly states that the
intermediary liability shields apply, Regulation 19 — the hosting shield —
offers initial protection to some service providers.

For sites disseminating pornography (as
defined by s 15) which the sites’ operators have selected and uploaded, the
protection of the regulations will not be available: these operators are not
merely technical intermediaries. Where the service in question is operated ‘on
a commercial basis’, the s 14 requirement will be made out, and age
verification will be required.

For services which consist of the storage
of information provided by a recipient of the service — for example, Twitter —
the hosting shield will be available.

However, the hosting shield is more porous
than that offered to access providers, and Regulation 19 excludes a service
provider’s liability only until the point at which the service provider has
knowledge of the problematic information and where it acts expeditiously to ‘remove
or disable access to the information’.

Ancillary service providers and the notice
regime

Considerable Parliamentary time was spent
on which service providers should be required to do what. Twitter, Tumblr,
Facebook and Instagram were all individually named.

On the sixth sitting of the House of
Commons on the Bill, on 20th October 2016, Matt Hancock MP (Minister of State
for Digital and Culture at the Department for Culture, Media and Sport) said
that ‘social media sites can be classified by the regulator as ancillary
service providers for facilitating or enabling the making of available
pornographic material’.

In his view, there was a ‘difference
between somebody who is actively putting up adult material and choosing not to
have age verification, and a platform where others put up adult material, where
it is not necessarily impossible but much harder to have a control over the
material. There is an important distinction here. If we try to pretend that
everybody putting material onto a platform … should be treated the same way as
a porn-providing website, we will be led into very dangerous territory and it
makes it harder to police this rather than easier’.

Lord Ashton of Hyde summarised the
government’s position in respect of such services in the second reading of the
Bill in the Lords, namely that ‘services, including Twitter, can be classified
by regulators as ancillary service providers where they are enabling or
facilitating the making available of pornographic or prohibited material’.

Who is an ‘ancillary service provider’

Defined in s 21(5), an ‘ancillary service
provider’ includes a person who appears to the age-verification regulator to ‘provide,
in the course of a business, services which enable or facilitate the making
available of pornographic material or extreme pornographic material on the
internet by’ the person who breaches s 14.

As described above, the definition is
intended to include social media service providers. It is likely also to cover
other intermediaries, such as web hosting providers or ‘web performance’
companies such as CloudFlare. Amendment 91, which did not make it into the Act,
proposed that the term ‘includes, but is not limited to, domain name
registrars, social media platforms, internet service providers, and search
engines’. The DCMS’s  ‘Draft Guidance to
the Regulator’ adds to the list ‘Online advertisers … Discussion fora and
communities in which users post links … Cyberlockers’ [sic] and cloud storage
services … App marketplaces that enable users to download Apps’.

In October 2016, Matt Hancock commented
that ‘it would be surprising if ISPs were not designated as ancillary service
providers’. The DCMS guidance of March 2017 takes a different stance,
commenting that as ‘a specific provision is included in respect of internet
service providers … it is not expected that the Regulator will notify internet
service providers’ under the ancillary service provider framework. This
guidance is not binding, but is at least somewhat persuasive: the
age-verification regulator must, according to s 27, ‘have regard’ to it.

The age-verification regulator is compelled
by s 25 of the Act to publish guidance ‘about the circumstances in which it
will treat services provided in the course of a business as enabling or
facilitating the making available of pornographic material’ so, in time,
clarity will (I hope) be brought to this issue.

The impact of being an ancillary service
provider

The age-verification regulator is empowered
to issue notices to ancillary service providers, where it considers that a
person is contravening s 14(1).

Unlike the imposition of a statutory duty
to block on an ISP given an administrative blocking notice, the Act does not
impose any positive obligation on an ancillary service provider. There is no
new statutory framework compelling such a provider to remove ‘offending
material’.

However, a notice of this nature is likely
to give a hosting services provider ‘actual knowledge’ for the purposes of the
eCommerce Regulations. Faced with such knowledge, the provider must act ‘expeditiously’
to remove or disable access to that information, to retain its protective
shield.

But against what liability is that shield
protecting the ancillary service provider? The impact of losing the shield, in
respect of this particular content, is unclear. Are they liable for the primary
obligation of ensuring age-verification is in place, failing which they could
be fined?

This seems a stretch. If we go back to the
language in s 14, the contravention of not having age verification in place
applies to a person who ‘makes pornographic material available’ on the
internet. However, to be an ‘ancillary service provider’, a person must merely ‘enable
or facilitate’ making porn available. The definitions do not appear to overlap,
and there is no reason why, simply by being served with a notice, a service
provider moves from ‘enabling or facilitating’ to ‘making available’.

As such, it seems highly questionable that,
in failing to take action once in receipt of a notice under s 21, an ancillary
service provider breaches the s 14 prohibition.

There was no debate in Parliament as to
whether an ancillary services provider would, in failing to take some form of
action in response to a notification, commit an inchoate offence, such as ‘assisting’
or ‘encouraging’ under the Serious Crime Act 2007. As with direct liability
under s 14, liability under the 2007 Act seems unlikely too.

In the draft Impact Assessment for the Bill,
the government states that it ‘will work closely with social media companies to
make sure they are committed to protecting children who use their platforms’. The
DCMS factsheet suggests that, in the context of payment service providers, the
age-verification regulator will ‘work with [those providers] to enable them to
withdraw their services from infringing sites’.

The approach, it seems, is one of
encouraging them to take action under their own terms of service, to withdraw
the service in question. In the context of payment service providers, the
government has assumed in its impact assessment that each major provider will
need to employ a dedicated, full-time member of staff, to engage with the
age-verification regulator, but there is no similar assumption in respect of
non-payment ancillary service providers.

It is, of course, entirely conceivable
that, if sufficient support is not forthcoming on a ‘voluntary’ basis, more
direct obligations or liabilities might be imposed by future regulation.

Compelling an ISP to block an ancillary
service provider

Could an ISP be compelled to block Twitter
or other social media provider if they did not remove content following a
notice? The regulator’s power to order blocking applies only in respect of a
person who contravenes s 14(1) (or who makes extreme porn available, but that’s
another topic) and, as above, it seems unlikely that a social media provider
does that. As such, without overstepping their mark legally, the
age-verification regulator would not appear to have that power.

(Even if it did, exercising it in a manner
which blocked a service like Twitter would seem disproportionate. As the Earl
of Erroll put it during a debate in the House of Lords, ‘[i]t is probably
unrealistic to block the whole of Twitter—it would make us look like idiots’.)

‘Porn sites’ and apps

Where a site falls outside the hosting
shield — for example, because it is distributing its own content — and the
material in question is pornographic within the meaning of s 15 and is
available to persons in the UK, that person would be within the scope of the s 14
requirement if the provision is on a ‘commercial basis’.

‘Commercial basis’

Unsurprisingly, ‘commercial basis’ is to be
interpreted more broadly than merely in situations where payments exchange
hands. The draft Online Pornography (Commercial Basis) Regulations 2017
provide, at reg 3, that a commercial basis is found where payment, reward, or
benefit is made or received ‘in connection with’ the making available of the
pornography, including where received by ‘other companies owned or controlled’
by the person making the pornography available.

The government’s impact assessment notes
that:

‘if [providers of pornography] charge a
fee, by whatever means and for whatever duration for access to the material,
sell advertising directed at users of the material or otherwise receive a
commercial benefit (for example, through the usage of user data), even if that
commercial benefit is not significant or is offset by losses whether from that
business or associated businesses, we want them in principle to be caught by
the provision.’

Whether this goes as far as the CJEU’s
interpretation of ‘within the course of its economic activity’ in McFadden (C
484/14) is unclear, but the scope is clearly intended to be broad.

Encyclopaedias and similar resources

Clause 6 provides a limited exception for
sites where ‘it is reasonable to assume that the overwhelming majority of
persons who visit [that site] do not view content of a pornographic nature on
[that site] or if it is reasonable to assume that the overwhelming majority of
the overall content [on that site] is not content of a pornographic nature’.

It is not the clearest piece of drafting,
but it would appear to mean that, for example, a general encyclopaedia, or a
site offering sexual education or sexual health information, would not be
treated as making pornography available on a commercial basis, even if they did
include pornography within their service offering. According to Lord Ashton of
Hyde, speaking in the Lords in March 2017, such sites could still be ‘enabling
and facilitating the availability of commercial pornography’ and subject to an
ancillary service provider notification although, as above, it is unclear what,
if anything, this means in practice.

A popularity-based approach

For those running sites which fall squarely
within the s 14 requirement, it appears that the age-verification regulator is
likely to take a systematic, popularity-based approach. David Austin of the
British Board of Film Classification, the body which is to act as the
age-verification regulator, was invited as a witness to the Public Bill
Committee. He commented that:

‘[The BBFC] would devise a proportionality
test and work out what the targets are in order to achieve the greatest
possible level of child protection. We would focus on the most popular websites
and apps accessed by children —those data do exist. We would have the greatest
possible impact by going after those big ones to start with and then moving
down the list.’

In the Lords in early 2017, Baroness Howe
stated that ‘the Government have made clear that they are expecting a ‘proportionate
enforcement’ targeting the biggest pornography sites … likely to be the top 50
to start with’.

This type of ‘top down’ approach is borne
out in s 26, which permits the age-verification regulator to exercise its powers
as against persons who make pornography available to a ‘large number of persons’
or who ‘generate a large amount of turnover’ by doing so.

Not just websites

Although websites received the most
discussion in Parliament, and the most attention outside Parliament, the Act is
not limited to websites. Someone making pornography available online other than
through a web interface would also be in scope.

For example, the DCMS factsheet on the Bill
described it as covering ‘all websites and “apps” containing pornographic
material’. It would seem likely that any online interactive system, including
peer-to-peer distribution, would count as ‘on the Internet’.

What about an individual who uploads
pornography?
 

There has been some discussion as to
whether an individual who uploads pornography — for example, a pornographic
photograph which they have taken — would breach the requirement of s 14 if they
failed to ensure that the material was not accessible to persons under 18.

Relatively little Parliamentary time was
spent on this issue, although this is not surprising given the early
acknowledgement that the strategy is one of focussing on the ‘top 50’ providers
initially. In his witness evidence to the Joint Bill Committee, Alan Wardle, of
the NSPCC, commented that ‘[w]e know that there are porn stars with Twitter
accounts who have lots of people following them and lots of content, so it is
important that that is covered’, but there was no further Parliamentary debate
on this point.

‘commercial basis’

In terms of individual user liability, as a
starting point, we can exclude users who are not acting ‘on a commercial basis’.
The consultation document underpinning the legislation provided that there was
no intent that the proposal ‘should impact on individuals who engage in
consensual sharing of private sexual photographs and films’, and the draft
Online Pornography (Commercial Basis) Regulations 2017 do not appear to intrude
onto this territory.

Where a user has a connection to a
commercial pornography site, there is likely to be a reasonable argument that
any posting by them of pornography, perhaps otherwise than in a truly anonymous
manner, confers a benefit on them, such as traffic to their commercial site.

‘makes … available’

There is, perhaps, some scope for argument
around the interpretation of ‘makes … available’, and whether the act of
uploading, rather than just the act of hosting, is sufficient. On the one hand,
as it is the act of hosting the content which, in a literal sense, is what
makes the content available to others over the Internet, a restrictive approach
might hold that this, and only this, falls within the scope of the framework.

On the other hand, uploading is an
unavoidable pre-requisite of the overall ‘making available’ of that content,
and it seems unlikely that the age-verification regulator (or a subsequent
appellate body) would draw a meaningful distinction between the two acts where
the intention of the uploader is to make the content available.

Likely approach

Given that the intended approach of the
age-verification regulator is one of tackling the biggest sites first, it seems
unlikely that individuals uploading pornography to ancillary service providers’
services will receive particular attention, at least initially.

If there is to be any movement against
individual uploaders, one would have thought that it would be in the context of
asking ancillary service providers to enforce whatever rules they might have in
place regarding the acceptability of content on their platforms.

Twitter’s rules, for example, state that ‘You
may not use our service for any unlawful purposes or in furtherance of illegal
activities.’ The age-verification regulator might expect Twitter to disable a
user’s account in response to a notice to Twitter that it considers that a user
is breaching s 14(1).

The responses of each social media site,
and other ancillary service providers, will be interesting to see.

What next?

Part 3 of the Act is not currently in
force. Implementation requires an appropriate statutory instrument (s 118(6))
which, presumably, will be laid before Parliament after the general election.
Similarly, although the government indicated in October 2016 its intent to
appoint the British Board of Film Classification as the age-verification
regulator, the formality of this appointment will also need to be conducted.

We will also need to await guidance on who
the BBFC considers to be an ‘ancillary service provider’, and the types of
arrangements for making pornographic material available that it  will treat as complying with s 14(1). This
guidance must be prepared in draft form and submitted to the Secretary of
State, who must lay it before Parliament. There is then a 40-day period in
which Parliament is afforded an opportunity to object to the guidance. If no
resolution to reject the guidance is passed, the BBFC must publish that
guidance as the finalised version.

Conclusion

The precise scope of s 14 of the Act
remains unclear.

Through a process of elimination,
predominantly reliant on the eCommerce Regulations’ intermediary shields, other
provisions of the Part, and governmental guidance, it appears that ISPs are
unlikely to fall within the s 14(1) requirement. It is less clear in the
context of ancillary service providers, but it seems that a ‘notice and strong
expectation that they will do something’ is the approach at this time.

For individuals connected with the
commercial provision of pornography, their personal uploading of pornographic
material is likely to fall within s 14. However, it seems more likely, in the
short term at least, that the regulatory focus will be on major pornography
sites, and that such uploading will not receive regulatory attention. Failing
this, the most likely course of action is for the regulator to notify the
ancillary service provider in question, to take some form of action under their
terms of service, such as removing the content in question or even suspending
or deleting the user’s account.

 Neil Brown is an experienced Internet, telecoms
and technology lawyer and managing director of law firm decoded:Legal (
https://decodedlegal.com). For many years,
he was responsible for legal aspects of Vodafone’s ‘content control’ filtering
system, and for liaison with the BBFC on content classification matters.