My Blog

Cortana, Microsoft’s virtual assistant designed to compete with Apple’s Siri and Google’s Assistant, is to be retired on mobile.

Instead it will focus on offering productivity help in Windows 10, Outlook and Teams.

The hashtag #RIPCortana was being used on Twitter, as people reminisced – or in some cases pointed out how forgettable the assistant had been.

Meanwhile, Apple’s Siri will no longer default to a female voice in English.

The use of a female voice for virtual assistants has long been controversial for gender-typing a helpful, virtual companion.

Cortana was unveiled in 2014 as a virtual assistant for Windows phones. It was named after the advanced artificial intelligence guide in Microsoft’s then best-selling Halo game series.

Three years later, Microsoft abandoned its smartphone operating system, although Cortana remained available for iPhones and Android devices.

The death of the voice assistant on most platforms was announced last summer and in January, Microsoft ended support for Cortana integration in the Harman Kardon Invoke speaker. It offered speaker owners who used Cortana, a $50 Microsoft gift card.

Ben Wood, chief analyst at research firm CSS Insight said: “There was a certain inevitability to Microsoft abandoning the consumer-centric variant of Cortana. Amazon’s Alexa and Google’s Assistant are the mass market voice assistants of choice, leaving little space for rivals.

“Even the mighty Apple has struggled to get traction with Siri despite making huge investments to drive the platform forward.

“Microsoft has made a sensible decision to double down on Cortana as a platform to aid productivity gains, closely tied to its business-centric tools and services. Increasingly Cortana will become deeply integrated into specific Microsoft platforms, rather than being a generic voice assistant designed to be all things to all people.”

Female voice
Rival Apple has added two new voices to its assistant Siri, as well as eliminating the default female voice in the latest version of iOS.

In 2019 a report by UNESCO suggested that using female voices by default for voice assistants “sends a signal that women are obliging, docile and eager-to-please helpers available at the touch of a button or with a blunt voice command”.

Apple said of its decision to put the onus on users to choose the voice of its assistant: “This is a continuation of Apple’s commitment to diversity and inclusion, and products and services that are designed to better reflect the diversity of the world we live in.”

In some countries and languages, Siri already defaults to a male voice.

The NHS has promised transparency in any future deals with US data firm Palantir after the coronavirus pandemic.

It follows a lawsuit by Open Democracy which accused the NHS of “sneaking through” a £23m contract with the firm.

The group said it had won the lawsuit, because “the government finally caved”.

But the NHS maintains it always acted legally, and that campaigners dropped the case “when they realised they didn’t have a leg to stand on”.

The lawsuit was withdrawn by Open Democracy and the NHS clarified it would follow the proper process in future.

Palantir: The controversial firm now worth £17bn
NHS faces legal action over data contract
Palantir helps analyse huge volumes of data generated by governments and businesses, and sorts through the information for useful insights, patterns and connections.

It has frequently been the subject of scrutiny by privacy campaigners.

Founded with support from the US Central Intelligence Agency in 2003, it has been linked to efforts to track undocumented migrant workers in America in recent years.

The deal
Palantir’s involvement in the NHS began in March 2020, on a short-term basis, to suggest how best to deploy resources during the coronavirus pandemic.

It analysed a so-called “datastore” of health information where personal details were “pseudonymised, anonymised or aggregated” to protect privacy.

In December, Palantir’s short-term contract was extended for two years.

The contract – seen by the BBC – sets out how the data analysis will be used for insights into dealing with the coronavirus pandemic.

‘Mission creep’
One clause also says “the services to be provided by Palantir extend to matters far beyond the response to the Covid-19 pandemic” -including Brexit, NHS workforce plans, and general government business.

Open Democracy, which labels Palantir a “spy-tech” company, argued this clause was “mission creep” and needed public consultation.

The NHS said it would have always followed due process for any non-Covid applications of the data, regardless of a legal challenge.

It did not rule out Palantir using data this way in future.

Impact assessment
The technical complaint was whether a new Data Protection Impact Assessment (DPIA) needed to be done for the revised deal.

A DPIA can involve public consultation, depending on what data is used and how.

One was completed for Palantir’s current contract in April last year, and the NHS confirmed it would conduct another if the contract was extended.

“Facing our lawsuit, the government has finally caved,” an Open Democracy blog said.

“They’ve pressed pause, committing not to extend Palantir’s contract beyond Covid without consulting the public.”

‘Spurious claim’
But the NHS told the BBC the contract had not been paused, and said proper checks would have always been conducted before extending the parameters of Palantir’s role.

A spokesman said: “Open Democracy have had to drop their court case unilaterally as it was apparent even to them that the NHS has always acted in accordance with its legal responsibilities.

“They, therefore, stood no chance of succeeding in their completely spurious claim… far from ‘winning’ this case, they had no choice but to drop it when they realised they hadn’t a leg to stand on.”

Data and how to manage and process it has been key to the battle against Covid-19 – just think of the charts shown by Chris Whitty and Sir Patrick Vallance at the Downing Street press conferences, and the government’s insistence that the timetable for easing lockdown is all about data, not dates.

So last spring, when the data about the disease was thin and managing it looked a bit of a nightmare, the government turned to a company that had experience at just this kind of work.

Palantir had already been a supplier to the public sector, notably working for Sunderland council to bring together various streams of data from social services and the police in an efficiency drive. It was also a controversial company, both for its work with ICE, the US border force working to deport undocumented immigrants, and because its founder Peter Thiel was one of the very few Silicon Valley figures to support President Donald Trump.

The government has always insisted that while Palantir’s tools were used to interrogate and understand the mass of data about the virus, the company was not being given any permanent access to sensitive health records, or the ability to use them for other purposes.

The decision to renew the contract with a £23m deal came as no surprise to another supplier of data services to the UK government.

“Once you’re in, you’re in”, he said, making the point that companies often underbid on a first contract just to get a foot in the door.

“You know you’re going to get the extensions, the improvements, which will make the whole thing profitable. The system is designed not to change. Colour me not surprised that Palantir got an extension.”

‘Fake’ accounts claiming to be Amazon workers have been praising their working conditions on Twitter.

Votes are currently being counted in Alabama to decide whether Amazon warehouse workers will form a union.

But last night, a series of anti-union tweets were sent from accounts claiming to be staff.

Twitter has now suspended many of the accounts, and Amazon has confirmed at least one is fake.

Most of the accounts were made just a few days ago, often with only a few tweets, all related to Amazon.

“What bothers me most about unions is there’s no ability to opt out of dues,” one user under the handle @AmazonFCDarla tweeted, despite a state law in Alabama which prevents this.

“Amazon takes great care of me,” she added.

The BBC is not responsible for the content of external sites.
View original tweet on Twitter
Another account – which later changed its profile picture after it was revealed to be fake – said: “Unions are good for some companies, but I don’t want to have to shell out hundreds a month just for lawyers!”Amazon FC
Many of the accounts involved used the handle @AmazonFC followed by a first name.

Amazon has previously used this handle for its so-called Amazon Ambassadors – real employees who are paid by the firm to promote and defend it on Twitter.

Amazon deploys Twitter ‘defence force’
‘Fake’ Amazon ambassadors baited online
An Amazon spokesperson told the BBC that Darla was not an official Ambassador, but has not responded to the BBC’s request for clarification on several other accounts.

“It appears that this is a fake account that violates Twitter’s terms,” the spokesperson said. “We’ve asked Twitter to investigate and take appropriate action.”

Fake accounts
Several of the high-profile accounts have been suspended by Twitter. It told the BBC that Amazon Ambassadors are subject to Twitter’s rules on spam and platform manipulation.

Accounts which impersonate or falsely claim to be affiliated with a company, can be temporarily suspended or removed.

Any parody account should have a disclaimer in its Twitter bio, the company added.

It is unclear whether the accounts are real employees, bots or trolls pretending to be Amazon Ambassadors.

Amazon hits back
Amazon’s executives and its official accounts have been tweeting in defence of the company in recent days, after negative reports of poor working conditions, including staff having to urinate in bottles.

The BBC is not responsible for the content of external sites.
View original tweet on Twitter
The official Amazon News account tweeted: “You don’t really believe the peeing in bottles thing, do you? If that were true, nobody would work for us.

“The truth is that we have over a million incredible employees around the world who are proud of what they do, and have great wages and health care from day one.”

The BBC is not responsible for the content of external sites.
View original tweet on Twitter
In response to a tweet by US Democrat senator Elizabeth Warren, which accused Amazon of “exploiting loopholes and tax havens”, the account said:

“You make the tax laws, we just follow them. If you don’t like the laws you’ve created, by all means change them.

“Here are the facts: Amazon has paid billions of dollars in corporate taxes over the past few years alone.”

The US has warned it could put tariffs of up to 25% on a host of UK exports in retaliation for a UK tax on tech firms.

Ceramics, make-up, overcoats, games consoles and furniture could all be hit, according to a list published by the Biden administration.

The duties are designed to raise $325m (£235.8m), the amount the US believes the UK will raise from US tech firms.

A UK government spokesperson said it wanted to “make sure tech firms pay their fair share of tax”.

They added: “Should the US proceed to implement these measures, we would consider all options to defend UK interests and industry.”

Washington is pressing ahead with the action, initiated under President Donald Trump, and has scheduled hearings on the list.

It argues the recently introduced digital services tax – which taxes tech firms on their revenues – has “unreasonable, discriminatory, and burdensome attributes”.

US suspends tariffs on single malt Scotch whisky
Sunak urges US to back digital services tax
Facebook agrees to pay France €106m in back taxes
Such actions have proceeded against similar taxes in India, Austria and Spain, but action against the European Union as a whole was dropped.

The US Section 301 action is designed to apply domestic political pressure within the UK and other countries over the imposition of such taxes.

The UK and US held talks about the digital services tax on 4 December, and UK government sources stressed that the tariff list was being seen as procedural, rather than an escalation.

The tariffs are now subject to a consultation in the US over the next few weeks.

‘Hugely disappointing’
UK ceramics are on the US Trade Representative’s list, including certain tiles, bathroom ware like sinks and bidets, as well as ceramics for laboratory uses.

About £17m of these products were exported to the US in 2020, and £24m in 2019 before Covid.

Trade group the British Ceramic Confederation said it was “monitoring developments closely, working with UK officials”.

Meanwhile Adam Mansell, head of the UK Fashion & Textile Association (UKFT), called the threat to UK-made overcoats “hugely disappointing”, noting the US had only removed separate tariffs on other types of fashion goods, such as British cashmere last month.

“At a time when we are trying to start discussions over a UK-US trade deal, it is extremely important that both governments get around the table to remove this threat as soon as possible,” he said.

“With the industry still struggling with the impact of Covid-19 and understanding the new trade arrangements with the EU, an additional burden on our exports couldn’t come at a worse time.”

‘Public frustration’
At the Budget, the Office for Budget Responsibility calculated the digital services tax would raise £300m in the current financial year, and as much as £700m in future years.

Brought in last April it taxes at 2% the revenues – not profits – of search engines, social media services and online marketplaces which derive value from UK users.

It followed years of claims in Europe and elsewhere that big tech firms do not pay enough tax in the countries where they operate.

Last August, Facebook agreed to pay the French government €106m (£95.7m) in back taxes to settle a dispute over revenues earned in the country.

Earlier that year, Facebook boss Mark Zuckerberg said he recognised the public’s frustration over the amount of tax paid by tech giants.

‘Temporary’
A UK government spokesperson said: “Like many countries around the world, we want to make sure tech firms pay their fair share of tax. Our digital services tax (DST) is reasonable, proportionate and non-discriminatory.

“It’s also temporary. We’re working positively with the US and other international partners to find a global solution to this problem and will remove the DST when that is in place.”

There are signs the Biden administration wants a more conciliatory relationship on trade with the UK than Donald Trump did.

Last month, Washington agreed to suspend tariffs on UK goods, including single malt whiskies, that were imposed in retaliation over subsidies to aircraft maker Airbus. However, the UK is still lobbying the US to drop duties on British steel brought in in 2018.

A cyber-attack has disrupted live broadcasts on Australia’s Channel Nine TV network, prompting concerns about the country’s vulnerability to hackers.

The broadcaster said it was unable to air several shows on Sunday, including Weekend Today.

Nine said it was investigating whether the hack was “criminal sabotage or the work of a foreign nation”.

Australia’s parliament was also investigating a possible cyber-attack in Canberra on Sunday.

Assistant Defence Minister Andrew Hastie said access to IT and emails at Parliament House had been cut as a precaution. He said this was done in response to issues affecting an “external provider”, without elaborating.

Australia parliament hit by cyber-hack attempt
Australia defence data in ‘extensive’ hack
“This is a timely reminder that Australians cannot be complacent about their cyber-security,” the minister told the News.com.au website on Sunday.

“The government acted quickly, and we have the best minds in the world working to ensure Australia remains the most secure place to operate online,” he said.

It’s not clear if the parliamentary outage and the cyber attack on Channel Nine were connected. The station’s presenters tweeted about their on-air issues.

The BBC is not responsible for the content of external sites.
View original tweet on Twitter
1px transparent line
The Australian government, institutions and major corporations have fallen victim to a string of cyber-attacks in recent years.

Last year, Prime Minister Scott Morrison also warned that Australian organisations were being targeted by a sophisticated foreign “state-based” hacker.

Australia’s parliament and political parties were also hit by a cyber attack in 2019.

Cyber intelligence experts say only a few states – China, Russia, Iran, and North Korea – have the capacity for such attacks and are not allied with Australia.

Nine TV presenter Karl Stefanovic joked about “the Russians” on the network’s Today Show on Monday.

Experts have long linked various hacks in Australia to China, including the parliament attack in 2019.

What did Channel Nine say about the cyber attack?
Nine, which also owns The Sydney Morning Herald and The Age newspapers, said its publishing and radio divisions were largely unaffected.

But its TV section was. At first, Nine said it was “responding to technical issues” affecting its live broadcasting.

Weekend Today, which runs from 07:00 to 13:00 local time (21:00 to 03:00 GMT) from Sydney, did not air.

Its online news site, 9news.com.au, was also affected.

On Sunday night, Nine confirmed there ha-d been a “cyberattack on our systems”.

“Our IT teams are working around the clock to fully restore our systems which have primarily affected our broadcast and corporate business units. Publishing and radio systems continue to be operational,” the company said in a statement.

A later report by Mark Burrows, a senior journalist for the network, said the company was “under attack by hackers”. He said emails and editing systems had gone down.

“I’m not surprised,” Mr Hastie, the assistant defence minister, told Nine. “Last year alone we had 60,000 reports to Australian cyber-security of cyber-crime. That’s one every 10 minutes.”

Nine told all its staff to work at home until further notice. It was able to its shows on Monday.

A firm that has the registered details of more than nine million chipped pets across the UK has been accused of losing its customers’ data.

Petlog is requesting that all users create a new account, but is not explaining why they need to do so.

One dog owner told the BBC he had logged on and received the details of someone else with the same name, including a phone number and address.

Petlog has not responded to the BBC’s requests for comment.

David Plant contacted the BBC after he saw messages on a Facebook group suggesting that Petlog had lost customer data and needed everyone to re-register.

He went on the website and typed his name in – but instead of being matched to his springer spaniel, Sally, he was provided with the details of another man with the same name, and his dog, Max. The details included the address and mobile phone number.

“This seems like a massive breach of GDPR (data-protection regulations),” Mr Plant told the BBC. “In theory I could register his dog to my address and claim him as mine.”

He said numerous messages and calls to Petlog had gone unanswered.

Another said the situation was creating “chaos for pet owners”.

“Probably thousands of pets with microchips inserted are no longer registered, leaving owners unable to be reunited with stolen or lost pets and are completely unaware of this,” Chris Boston told the BBC.

Anger over electric-shock dog collar YouTube clips
Households ‘buy 3.2 million pets in lockdown’
Professor of law at Newcastle University, Lilian Edwards, believes what has happened represents a breach of GDPR rules that should have been reported to the information commissioner’s office (ICO).

“It sounds like a massive database issue and [it] obviously contained personal data, so it is a breach and they should have notified the Information Commissioner’s Office within three days,” she said.

“They are not handling it very well and that is surprising because people are very attached to their pets.”

The ICO said it had no record of such a breach, although it said that not all such incidents needed to be reported.

One BBC reporter checked the details of their cat, Smudge, on the database and was unable to re-register its chip.

The site said: “Your pet’s details have not been lost. We just need to clean up some of your data first.”

The Petlog website added: “Our online services and website have been upgraded to ensure the database is secure. All data has been safely and securely migrated. As you may not have added your details to the database in the first instance, some of our security questions may not immediately match.”

Missing dogs
In the FAQ section, Petlog explained that for some customers with new details, such as a different email address, the process would take longer.

“Customers can fill in the ‘can’t see my pet’ form within their online account so the system can match their record and their new details with their pet’s record,” it said.

The firm added: “We understand this might be concerning, but we can give our reassurance that all pets are still safely on our microchip database and in the event of a pet going missing, reunification [sic] won’t be affected.”

But on its Facebook page, people disputed this.

“Two dogs found on M6 and couldn’t have their microchips details passed on as they weren’t on the system,” wrote one.

Many others said they had had difficulties re-registering, with some saying they no longer had a record of their pet’s chip ID.

“This is a complete and utter disgrace,” wrote one customer on Petlog’s Facebook group, asking why she had found out about the issues on social media.

“Petlog should have immediately contacted all customers directly to speed up correction of this dire cock-up.”

Kate Bevan, editor of Which? Computing magazine and a cat owner, said: “It’s concerning to discover that Petlog has apparently failed to keep track of the personal information of its customers and of the details of their beloved pets.

“All organisations are required by law to protect user information. Petlog must resolve the problems as a matter of urgency.”

Pet theft soars
Twitter account Missing Pets GB urged all owners to check their accounts immediately.

Petlog is managed by the UK Kennel Club, which in January issued a statement apologising for issues arising from the “implementation of our new database”.

At the time, chief executive Mark Beazley blamed the switchover to a new system for a range of problems and delays that customers were experiencing.

“I give you my assurance that we will resolve the remaining customer-service issues and offer further online improvements,” he said at the time.

Pet ownership has rocketed during the Covid pandemic, with UK households buying 3.2 million pets since it began, according to the Pet Food Manufacturers’ Association.

Pet theft has also increased during lockdowns, prompting campaigners to urge the government to introduce tougher penalties for the crime.

Anyone with a smartphone that uses facial recognition will know it does not really work with a mask on.

That can be frustrating – but although masks have undoubtedly thwarted the facial-recognition industry, the technology has also adapted.

It may sound strange but wearing a mask does not necessarily stop a computer from identifying someone.

And there are even examples of the pandemic being used as an excuse to use facial recognition.

Less accurate
Last year, as people began to increasingly wear masks around the world, the prevailing view was it represented a huge challenge to facial recognition.

It seemed obvious – algorithms designed to analyse faces would be less accurate if part of the face was concealed.

And studies backed this up.

The National Institute of Standards and Technology (NIST) tested 89 commercial facial-recognition algorithms – and found a 5-50% error rate in matching faces with digitally applied masks to photos of the same person.

‘That’s Orwellian’
But some facial-recognition technologies still work pretty well on those wearing a mask.

In January, a US Department of Homeland Security “controlled-scenario test” found one with a 96% success rate – although the results “varied greatly between systems”.

“Based on these results” the department said, “organisations that need to perform photo ID checks could potentially allow individuals to keep their masks on, thereby reducing the risk of Covid-19 infection.”

And although some police forces are using facial recognition less – London’s Metropolitan Police, for example, has not conducted a facial-recognition test for over a year – it is still being used, even, reportedly, at Black Lives Matter protests last summer.

“Even if the use of this technology is temporarily interrupted… that doesn’t obviate the threat that this technology poses both in the short and the long term,” Michael Kleinman, from Amnesty International, told BBC News.

“Anyone walking in front of a camera where police departments are running facial recognition – their face can be captured and they can be identified. That’s Orwellian.”

‘Touchless experience’
In the private sector, it is harder to tell whether the use of facial recognition has decreased over the past year – there is no directory, no list of when and where it is used.

But on Monday, it was announced Disney World was to trial facial-recognition technology for a month.

Its system takes an image of a customer’s face and converts it into a unique number, which is associated with the form of admission used for entry into the theme park.

The idea is to cut waiting times, with Disney saying it wants a more “touchless experience”.

The BBC is not responsible for the content of external sites.
View original tweet on Twitter
And crucially, the technology does not require guests to take off their masks.

Uncovered areas
Even before the pandemic, research had been under way on how facial recognition could work with masks.

In Japan, NEC had been working on a system for people who wear masks because of allergies.

And in January, it announced one it said was 99.9% accurate.

It works out whether someone is wearing a mask and then focuses on the uncovered areas, such as the eyes and forehead.

And the company wants to sell it for security checks in office buildings and other facilities.

“Touchless verification has become extremely important due to the impact of the coronavirus,” NEC told Reuters.

So once again, the pandemic, rather than hindering facial recognition, is being used as a reason to use it.

“Early on in the pandemic, we thought one of the very few silver linings could be the decline in facial-recognition technologies – but we’ve absolutely found that it’s the opposite,” Ella Jakubowska a campaign officer at pressure group European Digital Rights, says.

“The pandemic has unfortunately provided cover for companies to push out to what are effectively mass-surveillance infrastructures, under the guise of public health.

“A lot of companies have seen it almost [as a] challenge to develop technologies that can even more incisively identify people in public.”

Strict rules
It has also been announced facial recognition will be used at the Olympics, due to start this July in Japan.

It is not yet clear what for – but there will be strict rules restricting shouting and singing, and foreign visitors will be unable to attend.

Already, the technology had been used to check whether people were wearing masks at sporting events, the Japan Times reported.

Facebook too is reportedly considering building facial recognition into its upcoming smart glasses, despite privacy activists’ concerns it could be used by stalkers.

Facial recognition remains hugely controversial – and those controversies will remain, whether or not masks are being worn.

Facebook has removed a group of China-based hackers it says targeted members of the Uighur community living abroad.

It said hackers used malicious websites and apps to infect devices and allow for remote surveillance, with journalists and activists targeted.

A majority of the cyber attacks didn’t happen directly on Facebook but used the social media platform to share links to infected sites.

This is not the first time hackers have been accused of such activity.

The Uighurs are originally from the north-western region of Xinjiang in China and those targeted are currently living in places including Turkey, the United States, Australia and Canada.

“This activity had the hallmarks of a well-resourced and persistent operation, while obfuscating who’s behind it,” Facebook’s Mike Dvilyanski, head of cyber espionage investigations, and Nathaniel Gleicher, head of security policy, said in a blog post.

Facebook said it removed accounts – which totalled fewer than 100 – it found to have been created by the hackers, a group known as Earth Empusa or Evil Eye.

It believes fewer than 500 accounts were targeted.

Facebook says some of the ways the group infected devices included:

creating fake Uighur-themed apps for the Android app store, including a prayer app and a dictionary app
posing on Facebook as journalists, students, human rights advocates or members of the Uighur community, building trust and tricking them into clicking on malicious links
creating look-alike websites for popular Uighur and Turkish news websites
The Chinese Embassy in Washington has yet to comment.

China is facing mounting criticism from around the world over its treatment of the mostly Muslim Uighur population in Xinjiang.

Rights groups believe China has detained more than a million Uighurs over the past few years.

China denies allegations of abuse, saying camps in the region are “re-education” facilities used to combat terrorism.

A new Royal Navy surveillance ship is to be built to protect “critical” undersea cables.

Defence Secretary Ben Wallace warned “the lights could go out” if national infrastructure was lost, and the cables were “incredibly important”.

He also told the BBC’s Andrew Marr that Russia had “taken a deep interest” in the cables and the UK would be “deeply exposed” without further measures.

It comes ahead of Monday’s publication of the defence command paper.

The document will give more detail for the armed forces on the conclusions of the integrated review of the UK’s foreign and defence policies.

But some parts were already announced this week, including the lifting of the cap on the number of nuclear warheads the UK holds in its stockpile.

The government had previously committed to reducing the level to a maximum of 180 by the middle of the 2020s, but the move would allow the number to reach 260.

Mr Wallace said it would ensure the country’s nuclear deterrent was “credible”, and would still be lower than other nations – pointing to France, which has 300.

But Labour’s shadow foreign secretary Lisa Nandy said the proposal had “baffled” opposition parties and they would not support it until the measure had been justified by the government.

‘Risk of sabotage’
Hundreds of thousands of miles of undersea cables circle the globe, providing internet and communications links between nations and continents.

The Ministry of Defence said they are “vital to the global economy and communications between governments” and are at “risk of sabotage” due to “submarine warfare”.

The new Multi Role Ocean Surveillance ship will be fitted “with advanced sensors and will carry a number of remotely operated and autonomous undersea drones which will collect data”.

The vessel, staffed by 15 people and due to come into service in 2024, will carry out operations in both UK and international waters.

The MoD added it will also “be able to support with other defence tasks, including exercises and operations in the Arctic which will become an increasingly contested area”.

Undersea cables carry more than 90% of the world’s communications – including trillions of dollars worth of financial transactions every day.

There’s growing concern these underwater arteries could be vulnerable to attack.

British and US military and intelligence officers have recently warned of Russian submarines “aggressively operating” near Atlantic undersea cables.

The Ministry of Defence says there’s a risk of sabotage – which presents an existential threat to the UK.

As part of a wider defence review – the MoD will order a new Royal Navy surveillance ship to monitor this critical infrastructure.

2px presentational grey line
Mr Wallace told the BBC: “The lights could go out if we lose our critical national infrastructure across the board. Cables are one part of that critical national infrastructure and incredibly important.

“Russia has certainly taken a deep interest in those cables, not only to the United Kingdom but obviously to the continent of Europe.

“[The vessel’s] job is going to be to protect not only critical national infrastructure, but other things. It will be able to do other surveillance functions around the sea and everything else and I think it is really important that we invest in t hat because otherwise we are deeply exposed.”

Defence reforms ‘will help make UK match-fit’
UK to lift cap on nuclear stockpile after review
Prime Minister Boris Johnson has promised his plan for modernising the armed forces and foreign policy will help make the UK “match-fit”.

The Integrated Review, first announced in 2019, will set out the UK’s defence and foreign affairs priorities for the next decade or so, during which cyber warfare in particular is expected to become a greater threat.

Broader foreign policy from the review was announced this week, with Foreign Secretary Dominic Raab pledging to boost alliances in the Indo-Pacific region, describing it as “increasingly the geopolitical centre of the world”.

But more on how the Armed Forces could be overhauled will be announced on Monday, following a boost in funding late last year.

The defence secretary was pushed by Andrew Marr on what the new command paper would mean for the size of the forces, but Mr Wallace said that was a decision for Parliament.

He added: “What I can give you is the assurance that we have had a record settlement, so I am making decisions, not in an environment of falling tide like in previous cuts, but in an environment where I am going to make the decision to have the right Armed Forces to match our ambition and meet the threat.”

Peloton has warned parents to keep children away from its treadmills after the death of a child.

Boss John Foley called it “tragic accident” – one of a “small handful” of incidents in which children have been hurt on the firm’s exercise equipment.

He did not disclose further details about what happened.

In a separate incident, a 3-year-old suffered head injuries after getting trapped under a Peloton treadmill, but will make a full recovery.

The latest incident report into that accident, disclosed on Wednesday, said the child “was found to have tread marks on his back matching the slats of the treadmill”.

Mr Foley, Peloton’s founder, said children should stay away from the firm’s machines, which have surged in popularity during the pandemic.

“We design and build all of our products with safety in mind. But in order to help ensure that you and your family members stay safe with Peloton products in your home, we need your help,” he wrote in the letter to customers.

“Keep children and pets away from Peloton exercise equipment at all times. Before you begin a workout, double check to make sure that the space around your Peloton exercise equipment is clear.”

The US Consumer Product Safety Commission said it was investigating the incident.

Peloton sells cycling machines and treadmills that can be connected online to virtual fitness classes. Its business has boomed as people look for alternative ways to exercise during lockdowns and gym closures.

The firm has said the skyrocketing demand has created supply chain pressures, leading to backlogs in orders for the machines.

‘Devastating’
Its treadmills, which sell for more than $2,400 (£1,722), were designed for use by those aged 16 and older, who weigh at least 105 pounds, it said.

Mr Foley said the company did not intend to release more information about the accidents, in order to respect privacy of those involved.

“While we are aware of only a small handful of incidents involving the Tread+ where children have been hurt, each one is devastating to all of us at Peloton, and our hearts go out to the families involved,” he wrote.

The company is “currently assessing ways to reinforce our warnings about these critical safety precautions to hopefully prevent future accidents,” he added.

Last year, Peloton recalled about 27,000 bikes in the US, citing problems after clip-on pedals broke unexpectedly during use.

There were an estimated 22,500 treadmill-related injuries treated at US emergency departments among all ages, including around 2,000 involving children under the age of eight, according to the consumer product safety commission.

Between 2018 and 2020, it received reports of 17 deaths related to the use of a treadmill, including one involving a 5-year-old.