Archive for Legislation

Proposed Legislation To Make IoT Devices More Secure

Digital Minister Margot James has proposed the introduction of legislation that could make internet-connected gadgets less vulnerable to attacks by hackers.

What’s The Problem?

Gartner predicts that there will be 14.2 billion ‘smart’, internet-connected devices in use worldwide by the end of 2019.  These devices include connected TVs, smart speakers and home appliances. In business settings, IoT devices can include elevators, doors, or whole heating and fire safety systems in office buildings.

The main security issue of many of these devices is that they have pre-set, default unchangeable passwords, and once these passwords have been discovered by cybercriminals the IoT devices can be hacked in order to steal personal data, spy on users or remotely take control of devices in order to misuse them.

Also, IoT devices are deployed in many systems that link to and are supplied by major utilities e.g. smart meters in homes. This means that a large-scale attack on these IoT systems could affect the economy.

New Law

The proposed new law to make IoT devices more secure, put forward by Digital Minister Margot James, would do two main things:

  • Force manufacturers to ensure that IoT devices come with unique passwords.
  • Introduce a new labelling system that tells customers how secure an IOT product is.

The idea is that products will have to satisfy certain requirements in order to get a label, such as:

  • Coming with a unique password by default.
  • Stating for how long security updates would be made available for the device.
  • Giving details of a public point of contact to whom cyber-security vulnerabilities may be disclosed.

Not Easy To Make IoT Devices Less Vulnerable

Even though legislation could put pressure on manufacturers to try harder to make IoT devices more secure, technical experts and commentators have pointed out that it is not easy for manufacturers to make internet-enabled/smart devices IoT devices secure because:

Adding security to household internet-enabled ‘commodity’ items costs money. This would have to be passed on to the customer in higher prices, but this would mean that the price would not be competitive. Therefore, it may be that security is being sacrificed to keep costs down – sell now and worry about security later.

Even if there is a security problem in a device, the firmware (the device’s software) is not always easy to update. There are also costs involved in doing so which manufacturers of lower-end devices may not be willing to incur.

With devices which are typically infrequent and long-lasting purchases e.g. white goods, we tend to keep them until they stop working, and we are unlikely to replace them because they have a security vulnerability that is not fully understood. As such these devices are likely to remain available to be used by cybercriminals for a long time.

What Does This Mean For Your Business?

Introducing legislation that only requires manufacturers to make relatively simple changes to make sure that smart devices come with unique passwords and are adequately labelled with safety and contact information sounds as though it shouldn’t be too costly or difficult.  The pressure of having, by law, to display a label that indicates how safe the item is could provide that extra motivation for manufacturers to make the changes and could be very helpful for security-conscious consumers.

The motivation for manufacturers to make the changes to the IoT devices will be even greater when faced with the prospect of retailers eventually being barred from selling products that don’t have a label, as is the plan with this proposed legislation.

The hope from cybersecurity experts and commentators is that the proposal isn’t watered-down before it becomes law.

GDPR Says HMRC Must Delete Five Million Voice Records

The Information Commissioner’s Office (ICO) has concluded that HMRC has breached GDPR in the way that it collected the biometric voice records of users and now must delete five million biometric voice files.

What Voice Files?

Back in January 2017, HMRC introduced a system whereby customers calling the tax credits and Self-Assessment helpline could enrol for voice identification (Voice ID) as a means of speeding up the security steps. The system uses 100 different characteristics to recognise the voice of an individual and can create a voiceprint that is unique to that individual.

When customers call HMRC for the first time, they are asked to repeat the vocal passphrase “my voice is my password” to up to five times to register before speaking to a human adviser.  The recorded passphrase is stored in an HMRC database and can be used as a means of verification/authentication in future calls.

It was reported that in the 18 months following the introduction of the system, HMRC acquired 5 million peoples’ voiceprints this way.

What’s The Problem?

Privacy campaigners questioned the lawfulness of the system and in June 2018, privacy campaigning group ‘Big Brother Watch’ reported that its own investigation had revealed that HMRC had (allegedly) taken the five million taxpayers’ biometric voiceprints without their consent.

Big Brother Watch alleged that the automated system offered callers no choice but to do as instructed and create a biometric voice ID for a Government database.  The only way to avoid creating the voice ID on calling, as identified by Big Brother Watch, was to say “no” three times to the automated questions, whereupon the system still resolved to offer a voice ID next time.

Big Brother Watch highlighted the fact that GDPR prohibits the processing of biometric data for the purpose of uniquely identifying a person, unless there is a lawful basis under Article 6, and that because voiceprints are sensitive data but are not strictly necessary for dealing with tax issues, HMRC should request the explicit consent of each taxpayer to enrol them in the scheme (Article 9 of GDPR).

This led to Big Brother Watch registering a formal complaint with the ICO.

Decision

The ICO has now concluded that HMRC’s voice system was not adhering to the data protection rules and effectively pushed people into the system without explicit consent.

The decision from the ICO is that HMRC now must delete the five million records taken prior to October 2018, the date when the system was changed to make it compliant with GDPR.  HMRC has until 5th June to delete the five million voice records, which the state’s tax authority says it is confident it can do long before that deadline.

What Does This Mean For Your Business?

Big Brother Watch believes this to be the biggest ever deletion of biometric IDs from a state database, and privacy campaigners have hailed the ICO’s decision as setting an important precedent that restores data rights for millions of ordinary people.

Many businesses and organisations are now switching/planning to switch to using biometric identification/verification systems instead of password-based systems, and this story is an important reminder that these are subject to GDPR. For example, images and unique Voiceprint IDs are personal data that require explicit consent to be given, and that people should have the right to opt out as well as to opt-in.