My Lords, I will speak to Amendments 3 and 5 and in support of the other two amendments in this group. All these amendments refer to Clause 1 and seek to add some specificity to its general nature. The first amendment in my name and that of my noble friend Lord Clement-Jones is Amendment 3. This inserts a new paragraph (c) into Clause 1(1), adding the text
“children where they are not primary users of products but are subjects of product use”.
Why is this necessary? Here I am indebted to a report on cybersecurity, the UK Code of Practice for Consumer IoT Security produced by the PETRAS National Centre of Excellence for IoT Systems Cybersecurity. Noble Lords may be aware of this group; it has a very strong record in this area. It is a consortium of leading UK universities dedicated to understanding the critical issues of the privacy, ethics, trust, reliability, acceptability and security of IoT. I commend this organisation to the small number of noble Lords in this Chamber interested in this area.
This report highlighted, among other things, the importance of children’s connected toys receiving the necessary scrutiny, due to the implications of embedded cameras and microphones, with the aim of ensuring the child’s and the parents’ protection and right to privacy. Such devices include a wide range of everyday artefacts with internet connectivity intended for use by children or in caring for them, such as interactive toys, learning development devices and baby or child monitors.
These connected toys and tools have the potential for misuse and unauthorised contact with vulnerable minors. The British Toy & Hobby Association has responded by offering a range of guidance notes and by interpreting the code of practice, but with SMEs manufacturing most of these devices, there is much more to be done to ensure that those organisations are sufficiently informed and equipped to produce and market toys that are secure.
Security is not straightforward, as the Minister has already pointed out. While these devices offer a range of advantages through their connectivity, they also potentially expose children and their families to risks that have not yet been fully articulated to many of the consumers who are buying these toys.
A real-life example is that the toy giant Mattel launched Hello Barbie. The Minister may be familiar with it—I do not know. This was as far back as 2015. It was a very innovative toy which it launched with a start-up business called ToyTalk. The principle of this toy was that it could converse using internet connectivity with speech recognition, so as well as talking it could listen. Hello Barbie also allowed parents to log in later and eavesdrop on their children’s conversations with their toys. I will leave your Lordships to decide the ethics of that.
But this connectivity raised some concerns, primarily around who could listen in and record these devices and store conversations and behavioural and location data, and for what purpose this data could be used. Toys like these are now prevalent and they raise significant questions about the appropriate support and guidance for the toy manufacturers, which understand an awful lot about conventional safety—they know how to make physically safe toys—but do not have a track record on developing informationally and data-safe toys because they have never been asked to do that before. This is a new venture for them, and it requires a totally new set of skills and standards, as the Minister might say.
As technology evolves hacking is increasing in sophistication, so it is necessary to keep moving forward. The challenge for cybersecurity in remaining ahead of
the risks is inevitably a technological one, and the Minister may remember that the Hello Barbie toy, having been launched and lauded for its security, was ultimately found at some point to have serious security issues. Even that toy, from a very large manufacturer, fell foul of the progress of information crime.
Nevertheless, it is clear that today some toy manufacturers are releasing connected toys without adequate safety and security features. This is a competitive and dynamic marketplace—a lot of it is to do with price—and first movers are rewarded. In addition, the skillset and knowledge base, as I have just said, for conventional toy safety is mismatched with these new toys and we need to find a way of addressing that divergence. This is going to require investment and new learning and will not happen unless the toy manufacturers are required to do it.
Secure software development and cybersecurity are novel demands on this sector. However, the fact remains that these toy manufacturers are potentially placing consumer safety and privacy at risk. It does not matter whether this occurs due to the immaturity of the sector, market pressures or the lack of sectoral attention to the problem.
In the view of the Petras report,
“there are no indications that this will be addressed through market forces. Instead, the certainty of legislation to maintain standards would level the playing field and make clear for SMEs where they need to invest to make their toys market ready.”
Thus, more than the technological challenge of staying ahead of hackers, what is salient here are the challenges to the implementation of basic security features in manufacturing such as basic authentication and encryption, without which children’s safety and security is at risk.
This amendment explicitly places child security front and centre in this Bill. In other legislation involving the internet and digital issues, such as the Online Safety Bill, the Government have imposed more onerous duties on those delivering services to children than to adults. This amendment would be entirely consistent with that approach—very much in the spirit of understanding that our children and young people are more vulnerable and therefore need more protection from harms.
I turn next to Amendment 5. The eagle-eyed among your Lordships will spot that it is very similar to Amendment 4, proposed by the noble Baroness, Lady Merron, and set out very elegantly by the noble Lord, Lord Bassam. In fact, I would suggest that, largely, its construction is better than ours because they managed to do the same thing in fewer words. I will speak to Amendment 5 but my comments apply to Amendment 4 as well.
Amendment 5 seeks to ensure that:
“Regulations under this section must include provision that all security requirements specified in accordance with this Act are included as essential requirements in statutory conformity assessments and marking procedures under the Radio Equipment Regulations 2017 … and in any other such assessments and procedures applicable to relevant connectable products.”
I am speaking to the spirit of both these amendments. Amendment 5—similar to that of the noble Lord, Lord Bassam—follows on from the advice and help of Which? I thank that organisation, which has really
been at the forefront of the consumer issues involved. In essence, the amendment picks up on three of the issues that the Minister tells us will be dealt with in SIs as soon as the Bill becomes an Act, but it takes the rather stronger approach of placing them in the Bill.
Paragraph (a) of proposed new subsection (2A) goes further than the general principle in specifying that passwords are not to be weak. As Which? explains, many smart products push the user to create a password themselves, rather than use a default password. However, they then allow weak and easily guessable passwords to be created, meaning that the risk of compromise stays high.
One of the outcomes of this amendment would be the introduction of a requirement for responsible password policy guidance to be adopted by the industry to ensure that security liability is not simply passed from the device manufacturer to the consumer. The Bill and associated guidance should be amended to clarify that every individual device must have a unique or user-set password that meets effective complexity requirements.
Paragraph (b) of proposed new subsection (2A) seeks to avoid the risk of disclosures going into a black hole or taking many years to fix. The Bill and associated guidance should be amended to make clear what is required of manufacturers, importers and distributors on provision of disclosure policy information, particularly around vulnerabilities. The appointed regulator should also clearly define and distribute a risk assessment framework for vulnerabilities that removes any sense of subjectivity and ensures that the response is effectively mandated.
Paragraphs (c) and (d) of our proposed new subsection concern the length of time a product is supported. The Government should introduce mandatory minimum support periods for smart products and consider whether these periods should reflect how long consumers, on average, continue to use such products. There is a precedent here. New ecodesign and energy labelling requirements came into force in England, Scotland and Wales in 2021. They include a requirement for electronic display items, including televisions, to be provided with firmware and security update support for a minimum of eight years after the last unit of a model has been placed on the market. A consistent approach to support periods for a range of products therefore needs to be considered, and it has already been considered in this other legislation.
Customers need absolute clarity on the support period manufacturers will offer, so that they are able to make more informed purchasing decisions. There must be a clear definition of what the “point of sale” means and how this relates to the definitions of “supply” in Clause 55. Without clearer specifications on what form the transparency requirements will take, there is a risk that this information could be hidden, obfuscated or even mislead. This amendment is designed to probe the Government’s thinking on these very important issues.
Finally, and very briefly, as a signatory to Amendment 2, I give it my full support.