Knives out for knives

As part of a broader campaign targeting knife crime the Home Office has published its consultation response on a new procedure for authorised police officers to issue takedown notices to online platforms (also now to include search engines). These would require 48-hour removal of specified illegal weapons content items, on pain of civil penalty sanctions.

The government has also tabled implementing amendments to the Criminal Law and Policing Bill. These merit close attention. A takedown regime of this kind inevitably faces some similar issues to those that confronted the Online Safety Act, particularly in how to go about distinguishing illegal from legal content online. The Online Safety Act eventually included some fairly tortuous provisions that attempt (whether successfully or not) to meet those challenges. In contrast, the Policing Bill amendments maintain a judicious silence on some of the thorniest issues.

Parenthetically, as a policy matter the idea of a system for giving authoritative illegal content removal notices to platforms is not necessarily a bad one — so long as the decision to issue a notice is independent and accompanied by robust prior due process safeguards.  Previously, back in 2019, I suggested a system of specialist independent tribunals that could be empowered to issue such notices to platforms, as (along with other measures) a preferable alternative to a ‘regulation by discretionary regulator’ scheme. That idea went nowhere.

But back to the Bill amendments. The most critical aspects of an official content removal notice regime are how illegality is to be determined, independence of the notice-giver, prior due process and safeguards. How do the government’s proposals measure up?

What is unlawful weapons content?

As the Online Safety Act has reminded us, the notion of illegal content is not as simple a concept as might be thought; nor is making determinations of illegality.

First off, there is the conceptual problem. Online content as such cannot be illegal: persons, not content, commit offences. It is only what someone does with, or by means of, content that can be illegal.

Of course, in everyday parlance we say that zombie knives are illegal, or that extreme pornography is illegal, and we know what we mean. Statutory drafting has to be more rigorous: it has to reflect the fact that the offence is constituted by what is done with the item or the content, with what intent, and subject to any available defences. It is legally incoherent to say that content constitutes an offence, without seeking to bridge that gap.

The Online Safety Act attempted to grapple with the conceptual difficulty of equating content with an offence. The Policing Bill amendments do not.

For England and Wales new clause NC79 in the Bill amendments asserts that content is “unlawful weapons content” if it is:

“content that constitutes…  an offence under section 1(1) of the Restriction of Offensive Weapons Act 1959 (offering to sell, hire, loan or give away etc a dangerous weapon)”

NC79 provides the same for offences under section 1 or 2 of the Knives Act 1997 (marketing of knives as suitable for combat etc and related publications), and under section 141(1) of the Criminal Justice Act 1988 (offering to sell, hire, loan or give away etc an offensive weapon).

That is all. The Online Safety Act (Section 59(2)) does kick off in a similar way, by stipulating that:

“ “Illegal content” means content that amounts to a relevant offence.”

But (unlike the Policing Bill amendments) section 59(3) goes on to try to bridge the gap between content and conduct:

“Content consisting of certain words, images, speech or sounds amounts to a relevant offence if—

(a) the use of the words, images, speech or sounds amounts to a relevant offence,

(b) the possession, viewing or accessing of the content constitutes a relevant offence, or

(c) the publication or dissemination of the content constitutes a relevant offence.”

The Bill amendments contain no equivalent clause.

Determining illegality

Even if the conceptual gap were to be bridged by a similar amendment clause, that does not mean that illegality is necessarily obvious just by looking at the online content. Each offence has its own conduct elements, mental element and any defences that the legislation may stipulate. Ofcom’s Illegal Content Judgements Guidance under the Online Safety Act devotes three pages to section 1(1) of the Restriction of Offensive Weapons Act 1959 alone.

Two issues arise with determining illegality: what information does the authorised police officer need to have in order to be able to make a determination? How sure does the officer have to be that an offence has been committed?

The Online Safety Act, recognising that illegality may have to be considered in a broader context than the online content alone, stipulates that a service provider’s determination of illegality has to be made in the light of all relevant information that is reasonably available to the service provider.

That has some parallels with the duties of investigating police officers under the Criminal Procedure and Investigations Act 1996: that all reasonable steps are taken for the purposes of the investigation and, in particular, that all reasonable lines of inquiry are pursued.

The 1996 Act duty applies to a police investigation conducted with a view to ascertaining whether a person should be charged with an offence, or whether a person charged with an offence is guilty of it. However, ascertaining whether an offence has been committed for the purpose of a content removal notice is not the same as doing so with a view to making a charging decision. In order to issue a content removal notice the officer would not need to identify who had committed the offence – only determine that someone had done so.

Assuming, therefore, that the 1996 Act duty would not apply if a police officer were considering only whether to issue a content removal notice, how far would the police have to go in gathering relevant information before deciding whether an offence had been committed?

There will of course be cases, perhaps even most cases, in which the illegality may be obvious – for instance from the kind of knife involved and what has been said online – and the possibility of a defence remote. But it will not necessarily always be simple, or even possible, to make an illegality determination simply by looking at the online content alone.

The Online Safety Act (and Ofcom’s guidance on making illegality judgements) attempts to indicate what information the service provider should consider in making judgements about illegality. The Bill amendments are silent on this.

Indeed, the Ofcom Online Safety Act guidance (which regards law enforcement as a potential ‘trusted flagger’ for this kind of offence) anticipates that the flagger may provide contextual information: “Reasonably available information for providers of user-to-user and search services” is:

• The content suspected to be illegal content.

• Supporting information provided by any complainant, including that which is provided by any person the provider considers to be a trusted flagger.

The silence of the Bill amendments on this topic is all the more eloquent when we consider that nowhere in the procedures – from content removal notice through to appeal against a civil penalty notice – is there any provision for the person whose content is to be removed to be notified or given the opportunity to make representations.

Comparison with the Online Safety Act

The government emphasises, in its Consultation Response para 6.7, that:

“The proposed measure sits alongside, and does not conflict with, the structures established through the Online Safety Act 2023.”

Strictly speaking that is right: a notice from a police officer under the Bill amendments could have three separate functions or effects:

-        Constitute a notice requiring 48-hour takedown under the new provisions.

-        Fix the service provider with awareness of illegality for the purpose of the OSA reactive duty under S.10(3)(b).

-        Fix the service provider with knowledge of illegality for the purpose of the hosting liability shield derived from the eCommerce Directive. 

Since these are three separate, parallel structures, it is correct that they do not conflict[1]. Nevertheless, they are significantly different from each other. As well as the differences from the Online Safety Act already outlined, the role of law enforcement under the Bill amendments is significantly different.

In particular, although under the Online Safety Act law enforcement may be considered to be a trusted flagger, Ofcom cautions that:

“A provider is not required to accept the opinions of a third party as to whether content is illegal content. Only a judgment of a UK court is binding on it in making this determination. In all other cases, it will need to take its own view on the evidence, information and any opinions provided.”

Therein lies the biggest difference between the Online Safety Act and the Bill amendments. Under the Bill amendments, subject to the review procedure outlined below, a service provider is required to act on the opinion of the police.

The government plans that the content removal system will be operated by a new policing unit, which will be responsible for issuing removal notices. That is presumably reflected (in part) by the Bill amendment provision that a content removal notice has to be given by an officer authorised by the Director General of the National Crime Agency or the chief officer of the relevant police force.

How sure that an offence has been committed?

A related aspect of determining illegality is how sure the person making the decision has to be that the content is illegal.  The Online Safety Act stipulates that the provider has to treat the content as illegal if it has ‘reasonable grounds to infer’ that the content is illegal. ‘Reasonable grounds to infer’ is a relatively low threshold, which has given rise to concerns that legitimate content will inevitably be removed with consequent risk of European Convention on Human Rights incompatibility.

The Bill amendments take a different approach: the police officer making the decision must be ‘satisfied’ that the content is unlawful weapons content. ‘Satisfied’ presumably is not intended to be a wholly subjective assessment. But if not, what degree of confidence is implicit in ‘satisfied’? If the police officer has residual doubts, or has insufficient information to make up his or her mind, could the officer be ‘satisfied’ that the content amounts to an offence? Equally, it probably does not mean ‘satisfied beyond all reasonable doubt’.

The Online Safety Act provides that a service provider does not have to take into account the possibility of a defence unless it has reasonable grounds to infer that a defence may be successfully relied upon. By contrast, under the Bill amendments it seems likely that the police officer would always have to be satisfied that no defence was available.

Safeguards

The government has sought to address the risk of ill-founded notices by means of a review mechanism. The content removal notice has to explain the police officer’s reasons for considering that the content is unlawful weapons content.  The service provider can request review of a notice by a more senior officer. The reviewing officer must then give a decision notice, setting out the outcome of the review and giving reasons. The government has said that it:

“…believes that the review process designed within the proposal adequately addresses online companies concerns with cases where it would be difficult to determine the illegality of content.” (Consultation Response, [6.8])

The review process, however, sheds no light on how much contextual information gathering by police officers is contemplated, nor on the degree of confidence implicit in being ‘satisfied’. It contains no element of independent third party review, nor any opportunity for the person whose content is to be removed to make representations.

That said, the procedure could perhaps be fleshed out by guidance to law enforcement that the Secretary of State may (but is not required to) issue under NC84.

Underlying all these considerations is the matter of ECHR compatibility. The lower or more subjective the threshold for issuing a notice, the less the predictability of the process or outcome, and the fewer or weaker the safeguards against arbitrary or erroneous decision-making, then the greater the likelihood of ECHR incompatibility.

It might be said against all of this that of course the police would only issue a content removal notice if was obvious from the online content itself that an offence was being committed. If that were the intention, might it be preferable to make that explicit and write a “manifest illegality” standard into the legislation?

Does it matter?

It could well be questioned why any of this matters. Who really cares if a few less knives appear online because content is wrongly taken down? That kind of argument is depressingly easy to make where impingements on freedom of expression are concerned. Thus in a different context, what does it really matter if, in our quest to root out the evils in society, we sacrifice due process and foreseeability to flexibility and remove a few too many tasteless jokes, insulting tweets, offensive posts, shocking comments, wounding parodies, disrespectful jibes about religion or anything else that thrives in the toxic online hinterland of the nearly illegal?

Opinions on that will differ. For me, it matters because the rule of law matters. Due process provides the opportunity to be heard. It matters that you should be able to predict in advance, with reasonable certainty, whether something that you are contemplating posting online is liable to be taken down as the result of official action (or, for that matter, the action of a platform seeking to comply with a legal or regulatory duty).

If you cannot do that, you are at the mercy of arbitrary exercise of state power. It is knives today, but who knows what tomorrow (we can, however, be sure that once one 48-hour takedown regime is enacted others will follow).  Abandon the rule of law to ad hoc power and, as Robert Bolt had Sir Thomas More declaim to William Roper in A Man For All Seasons:

“…do you really think you could stand upright in the winds that would blow then? Yes, I'd give the Devil benefit of law, for my own safety's sake!”.

However skilled a dedicated police unit may be, expertise is no substitute for due process, safeguards and independent adjudication. Otherwise, why would we bother with courts at all? The fact that content, rather than a person, is condemned is not, I would suggest, a good reason to skimp on rule of law principles.

It may be said that the Bill amendments provide for recourse to the courts. They do, but only once matters have got as far as a civil penalty notice imposing a fine for non-compliance; and they concern only the platform, not the person who posted the content. That is not the same as due process, safeguards or independent review at the outset of the decision-making process.

Extraterritoriality

To finish with a more technical matter: extraterritoriality. The Online Safety Act, although fairly aggressive in its assertion of jurisdiction, did recognise the need to establish some connection with the UK in order for a U2U or search service to fall within its territorial scope. Thus Section 4 of the OSA sets out a series of criteria to determine whether a service is UK-linked.

The Bill amendments contain no such provision. On the face of it a police officer could serve notices under the Act by email and (in the event of non-compliance) impose civil penalties on any service provider anywhere in the world, regardless of whether they have any connection with the UK at all. If that is what is intended, it would be an extraordinary piece of jurisdictional overreach.

That would also (presumably) bring into play delicate judgements by authorised police officers, when considering whether to serve a content removal notice, as to whether an activity on a platform that had no connection with the UK amounted to an offence within the UK. That is a matter of the territorial scope of the underlying UK offence. The Online Safety Act circumvents questions of that kind by, for the purpose of service provider duties, instructing the service provider to disregard territorial considerations:

“For the purposes of determining whether content amounts to an offence, no account is to be taken of whether or not anything done in relation to the content takes place in any part of the United Kingdom.”

The Bill amendments are silent on these difficult jurisdictional issues.

---

[1] This is on the basis that the notice regime would fall within the eCommerce Directive exception for specific court or administrative authority orders to terminate an infringement. That would depend on whether the police are properly regarded as an administrative authority.  If not, it could be argued that the Policing Bill amendments in substance are inconsistent with the eCommerce Directive hosting liability shield to which, as a matter of policy, the government ostensibly continues to adhere: "The government is committed to upholding the liability protections now that the transition period has ended." (The eCommerce Directive and the UK, last updated 18 January 2021).

The computer is always right - or is it?

This is my submission to the Ministry of Justice Call for Evidence on computer evidence in criminal proceedings. 

Some takeaways from 24 pages of rather dense legal analysis:

• The evidential presumption of reliability (properly so called) is a different animal from informal assumptions about the reliability of computers. They are related, and the latter may influence attitudes to e.g. the threshold for disclosure applications, or to the basic understanding of what the prosecution has to do to prove a case ‘beyond reasonable doubt’ (as opposed to theoretical or fanciful doubt). But the evidential presumption is legally distinct and has a specific, limited, function. The review should cast its net wider than the evidential presumption properly so-called, but equally not confuse it with informal assumptions.

• There is evidently a perception (possibly fostered by the Law Commission 1995 and 1997 recommendations) that the presumption of reliability applies automatically to all computing devices. I don’t think the caselaw supports that. As I read the cases, the court can decide whether or not to apply the presumption.

• If the prosecution deploys expert evidence on reliability the presumption is irrelevant: it’s a matter of deciding between experts. So for general purpose computers and software (i.e. excluding breathalysers, speed guns etc, where the presumption is routinely relied upon) how often does the prosecution actually rely on (and then the court apply) the presumption? This is a question for criminal practitioners (which I am not).  I have found no reported criminal cases (other than where accuracy was not questioned), but will readily stand to be corrected if there are any.

• Reliability of computer evidence always has to be considered in the context of what specifically is sought to be proved by it (which can vary widely). For instance, to make a general point, adducing computer records to evidence presence of a transaction is different from proving absence. The latter requires the computer records to be not just accurate, but complete.

• It should not be forgotten that a defendant may wish to adduce computer evidence (e.g. a video taken by them on their mobile phone).

• There may be a distinction to be made between the output of general purpose computing systems and dedicated forensic tools.

• It may also be pertinent to consider whether the computer evidence sought to be relied upon is central to the prosecution’s case, and whether it is corroborated or uncorroborated.

• A computer evidence regime that is predicated on whether a document is ‘produced by a computer’ is potentially problematic, for two reasons. First, there is hardly a document now that has not been touched by a computer at some point in its history. Are those all ‘produced by a computer’? Second (as presaged in the caselaw on S.69 PACE 1984), a bright line definition of that kind is liable to give rise to satellite disputes about what does and does not fall on either side of the line. It may be more fruitful to view matters through the lens of a regime for documentary evidence generally.

• The proposed distinction between generated and captured or recorded evidence is difficult to apply conceptually. The practical examples given in the Call for Evidence throw up many questions.

• Any proposed computer evidence regime should be tested against concrete hypotheticals.  I have suggested a list of fourteen, drawn from reported cases.

[Updated 16 April 2025 with a list of takeaways]

The Online Safety Act grumbles on

Policymakers sometimes comfort themselves that if no-one is completely satisfied, they have probably got it about right. 

On that basis, Ofcom’s implementation of the Online Safety Act’s illegality duties must be near-perfection: the Secretary of State (DSIT) administering a sharp nudge with his draft Statement of Strategic Priorities, while simultaneously under fire for accepting Ofcom’s advice on categorisation of services; volunteer-led community forums threatening to close down in the face of perceived compliance burdens; and many of the Act’s cheerleaders complaining that Ofcom’s implementation has so far served up less substantial fare than they envisaged. 

As of now, an estimated 25,000 UK user-to-user and search providers (plus another 75,000 around the world) are meant to be busily engaged in getting their Illegal Harms risk assessments finished by 16 March. 

Today is Safer Internet Day. So perhaps spare a thought for those who are getting to grips with core and enhanced inputs, puzzling over what amounts to a ‘significant’ number of users, learning that a few risk factors may constitute ‘many’ (footnote 74 to Ofcom’s General Risk Level Table), or wondering whether their service can be ‘low risk’ if they allow users to post hyperlinks.  (Ofcom has determined that hyperlinks are a risk factor for six of the 17 kinds of priority offence designated by the Act: terrorism, CSEA, fraud and financial services, drugs and psychoactive substances, encouraging or assisting suicide and foreign interference offences). 

Grumbles from whichever quarter will come as no great surprise to those (this author included) who have argued from the start that the legislation is an ill-conceived, unworkable mess which was always destined to end in tears. Even so, and making due allowance for the well-nigh impossible task with which Ofcom has been landed, there is an abiding impression that Ofcom’s efforts to flesh out the service provider duties - risk assessment in particular – could have been made easier to understand. 

The original illegal harms consultation drew flak for its sheer bulk: a tad over 1,700 pages. The final round of illegal harms documents is even weightier: over 2,400 pages in all. It is in two parts. The first is a Statement. In accordance with Ofcom’s standing consultation principles, it aims to explain what Ofcom is going to do and why, showing how respondents’ views helped to shape Ofcom’s decisions. That amounts to 1,175 pages, including two summaries. 

The remaining 1,248 pages consist of statutory documents: those that the Act itself requires Ofcom to produce. These are a Register of Risks, Risk Assessment Guidance, Risk Profiles, Record Keeping and Review Guidance, a User to User Illegal Content Code of Practice, a Search Service Illegal Content Code of Practice, Illegal Content Judgements Guidance, Enforcement Guidance, and Guidance on Content Communicated Publicly and Privately. Drafts of the two Codes of Practice were laid before Parliament on 16 December 2024. Ofcom can issue them in final form upon completion of that procedure.

When it comes to ease of understanding, it is tempting to go on at length about the terminological tangles to be found in the documents, particularly around ‘harm’, ‘illegal harm’ and ‘kinds of illegal harm’. But really, what more is worth saying? Ofcom’s documents are, to all intents and purposes, set in stone. Does it help anyone to pen another few thousand words bemoaning opaque language? Other than in giving comfort that they are not alone to those struggling to understand the documents, probably not. Everyone has to get on and make the best of it.

So one illustration will have to suffice. ‘Illegal harm’ is not a term defined or used in the Act. In the original consultation documents Ofcom’s use of ‘illegal harm’ veered back and forth between the underlying offence, the harm caused by an offence, and a general catch-all for the illegality duties; often leaving the reader to guess in which sense it was being used. 

The final documents are improved in some places, but introduce new conundrums in others. One of the most striking examples is paragraph 2.35 and Table 6 of the Risk Assessment Guidance (emphasis added to all quotations below). 

Paragraph 2.35 says: 

“When evaluating the likelihood of a kind of illegal content occurring on your service and the chance of your service being used to commit or facilitate an offence, you should ask yourself the questions set out in Table 6.”

Table 6 is headed: 

“What to consider when assessing the likelihood of illegal content”

The table then switches from ‘illegal content’ to ‘illegal harm’. The first suggested question in the table is whether risk factors indicate that: 

“this kind of illegal harm is likely to occur on your service?” 

‘Illegal harm’ is footnoted with a reference to a definition in the Introduction: 

“the physical or psychological harm which can occur from a user encountering any kind of illegal content…”. 

So what is the reader supposed to be evaluating: the likelihood of occurrence of illegal content, or the likelihood of physical or psychological harm arising from such content? 

If ‘Illegal Harm’ had been nothing more than a title that Ofcom gave to its illegality workstream, then what the term actually meant might not have mattered very much. But the various duties that the Act places on service providers, and even Ofcom’s own duties, rest on carefully crafted distinctions between illegal content, underlying criminal offences and harm (meaning physical or psychological harm) arising from such illegality. 

That can be seen in this visualisation. It illustrates the U2U service provider illegality duties - both risk assessment and substantive - together with the Ofcom duty to prepare an illegality Risks Register and Risk Profiles.  The visualisation divides the duties into four zones (A, B, C and D), explained below. 

A: The duties in this zone require U2U providers to assess certain risks related to illegal content (priority and non-priority). These risks are independent of and unrelated to harm. The risks to be assessed have no direct counterpart in any of the substantive safety duties in Section 10. Their relevance to those safety duties probably lies in the proportionality assessment of measures to fulfil the Section 10 duties. 

Although the service provider’s risk assessment has to take account of the Ofcom Risk Profile that relates to its particular kind of service, Ofcom’s Risk Profiles are narrower in scope than the service provider risk assessment. Under the Act Ofcom’s Risks Register and Risk Profiles are limited to the risk of harm (meaning physical or psychological harm) to individuals in the UK presented by illegal content present on U2U services and by the use of such services for the commission or facilitation of priority offences. 

B:  This zone contains harm-related duties (identified in yellow): Ofcom Risk Profiles, several service provider risk assessment duties framed by reference to harm, plus the one substantive Section 10 duty framed by reference to harm (fed by the results of the harm-related risk assessment duties). Harm has its standard meaning in the Act: physical or psychological harm. 

C: This zone contains two service provider risk assessment duties which are independent of and unrelated to risk of harm, but which feed directly into a corresponding substantive Section 10 duty. 

D: This zone contains the substantive Section 10 duties: one based on harm and three which stand alone. Those three are not directly coupled to the service provider’s risk assessment.

This web of duties is undeniably complex. One can sympathise with the challenge of rendering it into a practical and readily understandable risk assessment process capable of feeding the substantive duties.  Nevertheless, a plainer and more consistently applied approach to terminology in Ofcom's documents would have paid dividends.