IPVM, a U.S.-based security and surveillance industry research group, says the Chinese surveillance equipment maker Dahua is selling cameras with what it calls a “skin color analytics” feature in Europe, raising human rights concerns. 

In a report released on July 31, IPVM said “the company defended the analytics as being a ‘basic feature of a smart security solution.'” The report is behind a paywall, but IPVM provided a copy to VOA Mandarin. 

Dahua’s ICC Open Platform guide for “human body characteristics” includes “skin color/complexion,” according to the report. In what Dahua calls a “data dictionary,” the company says that the “skin color types” that Dahua analytic tools would target are ”yellow,” “black,” and ”white.”  VOA Mandarin verified this on Dahua’s Chinese website. 

The IPVM report also says that skin color detection is mentioned in the “Personnel Control” category, a feature Dahua touts as part of its Smart Office Park solution intended to provide security for large corporate campuses in China.  

Charles Rollet, co-author of the IPVM report, told VOA Mandarin by phone on August 1, “Basically what these video analytics do is that, if you turn them on, then the camera will automatically try and determine the skin color of whoever passes, whoever it captures in the video footage. 

“So that means the camera is going to be guessing or attempting to determine whether the person in front of it … has black, white or yellow — in their words — skin color,” he added.  

VOA Mandarin contacted Dahua for comment but did not receive a response. 

The IPVM report said that Dahua is selling cameras with the skin color analytics feature in three European nations. Each has a recent history of racial tension: Germany, France and the Netherlands.

‘Skin color is a basic feature’

Dahua said its skin tone analysis capability was an essential function in surveillance technology.  

 In a statement to IPVM, Dahua said, “The platform in question is entirely consistent with our commitments to not build solutions that target any single racial, ethnic, or national group. The ability to generally identify observable characteristics such as height, weight, hair and eye color, and general categories of skin color is a basic feature of a smart security solution.”  

IPMV said the company has previously denied offering the mentioned feature, and color detection is uncommon in mainstream surveillance tech products. 

In many Western nations, there has long been a controversy over errors due to skin color in surveillance technologies for facial recognition. Identifying skin color in surveillance applications raises human rights and civil rights concerns.  

“So it’s unusual to see it for skin color because it’s such a controversial and ethically fraught field,” Rollet said.  

Anna Bacciarelli, technology manager at Human Rights Watch (HRW), told VOA Mandarin that Dahua technology should not contain skin tone analytics.   

“All companies have a responsibility to respect human rights, and take steps to prevent or mitigate any human rights risks that may arise as a result of their actions,” she said in an email.

“Surveillance software with skin tone analytics poses a significant risk to the right to equality and non-discrimination, by allowing camera owners and operators to racially profile people at scale — likely without their knowledge, infringing privacy rights — and should simply not be created or sold in the first place.”  

Dahua denied that its surveillance products are designed to enable racial identification. On the website of its U.S. company, Dahua says, “contrary to allegations that have been made by certain media outlets, Dahua Technology has not and never will develop solutions targeting any specific ethnic group.” 

However, in February 2021, IPVM and the Los Angeles Times reported that Dahua provided a video surveillance system with “real-time Uyghur warnings” to the Chinese police that included eyebrow size, skin color and ethnicity.  

IPVM’s 2018 statistical report shows that since 2016, Dahua and another Chinese video surveillance company, Hikvision, have won contracts worth $1 billion from the government of China’s Xinjiang province, a center of Uyghur life. 

The U.S. Federal Communications Commission determined in 2022 that the products of Chinese technology companies such as Dahua and Hikvision, which has close ties to Beijing, posed a threat to U.S. national security. 

The FCC banned sales of these companies’ products in the U.S. “for the purpose of public safety, security of government facilities, physical security surveillance of critical infrastructure, and other national security purposes,” but not for other purposes.  

Before the U.S. sales bans, Hikvision and Dahua ranked first and second among global surveillance and access control firms, according to The China Project.  

‘No place in a liberal democracy’

On June 14, the European Union passed a revision proposal to its draft Artificial Intelligence Law, a precursor to completely banning the use of facial recognition systems in public places.  

“We know facial recognition for mass surveillance from China; this technology has no place in a liberal democracy,” Svenja Hahn, a German member of the European Parliament and Renew Europe Group, told Politico.  

Bacciarelli of HRW said in an email she “would seriously doubt such racial profiling technology is legal under EU data protection and other laws. The General Data Protection Regulation, a European Union regulation on Information privacy, limits the collection and processing of sensitive personal data, including personal data revealing racial or ethnic origin and biometric data, under Article 9. Companies need to make a valid, lawful case to process sensitive personal data before deployment.” 

“The current text of the draft EU AI Act bans intrusive and discriminatory biometric surveillance tech, including real-time biometric surveillance systems; biometric systems that use sensitive characteristics, including race and ethnicity data; and indiscriminate scraping of CCTV data to create facial recognition databases,” she said.  

In Western countries, companies are developing AI software for identifying race primarily as a marketing tool for selling to diverse consumer populations. 

The Wall Street Journal reported in 2020 that American cosmetics company Revlon had used recognition software from AI start-up Kairos to analyze how consumers of different ethnic groups use cosmetics, raising concerns among researchers that racial recognition could lead to discrimination.  

The U.S. government has long prohibited sectors such as healthcare and banking from discriminating against customers based on race. IBM, Google and Microsoft have restricted the provision of facial recognition services to law enforcement.  

Twenty-four states, counties and municipal governments in the U.S. have prohibited government agencies from using facial recognition surveillance technology. New York City, Baltimore, and Portland, Oregon, have even restricted the use of facial recognition in the private sector.  

Some civil rights activists have argued that racial identification technology is error-prone and could have adverse consequences for those being monitored. 

Rollet said, “If the camera is filming at night or if there are shadows, it can misclassify people.”  

Caitlin Chin is a fellow at the Center for Strategic and International Studies, a Washington think tank where she researches technology regulation in the United States and abroad. She emphasized that while Western technology companies mainly use facial recognition for business, Chinese technology companies are often happy to assist government agencies in monitoring the public.  

She told VOA Mandarin in an August 1 video call, “So this is something that’s both very dehumanizing but also very concerning from a human rights perspective, in part because if there are any errors in this technology that could lead to false arrests, it could lead to discrimination, but also because the ability to sort people by skin color on its own almost inevitably leads to people being discriminated against.”  

She also said that in general, especially when it comes to law enforcement and surveillance, people with darker skin have been disproportionately tracked and disproportionately surveilled, “so these Dahua cameras make it easier for people to do that by sorting people by skin color.”  
…

Virgin Galactic rocketed to the edge of space with its first tourists Thursday, including a former British Olympian who bought his ticket 18 years ago and a mother-daughter duo from the Caribbean.

The space plane glided back to a runway landing at Spaceport America in the New Mexico desert, after a brief flight that gave passengers a few minutes of weightlessness.

Cheers erupted from families and friends watching from below when the craft’s rocket motor fired after it was released from the plane that had carried it aloft. The rocket ship reached about 88 kilometers high.

Richard Branson’s company expects to begin offering monthly trips to customers on its winged space plane, joining Jeff Bezos’ Blue Origin and Elon Musk’s SpaceX in the space tourism business.

Virgin Galactic passenger Jon Goodwin, who was among the first to buy a ticket in 2005, said he had faith that he would someday make the trip. The 80-year-old athlete — he competed in canoeing in the 1972 Olympics — has Parkinson’s disease and wants to be an inspiration to others.

“I hope it shows them that these obstacles can be the start rather than the end to new adventures,” he said in a statement.

Ticket prices were $200,000 when Goodwin signed up. The cost is now $450,000.

He was joined by sweepstakes winner Keisha Schahaff, 46, a health coach from Antigua, and her daughter, Anastatia Mayers, 18, a student at Scotland’s University of Aberdeen. Also on board: two pilots and the company’s astronaut trainer.

It was Virgin Galactic’s seventh trip to space since 2018, but the first with a ticket-holder. Branson, the company’s founder, hopped on board for the first full-size crew ride in 2021. Italian military and government researchers soared in June on the first commercial flight. About 800 people are currently on Virgin Galactic’s waiting list, according to the company.

Virgin Galactic’s rocket ship launches from the belly of an airplane, not from the ground, and requires two pilots in the cockpit. Once the mothership reaches a height of about 15 kilometers, the space plane is released and fires its rocket motor to make the final push to just over 80 kilometers up. Passengers can unstrap from their seats, float around the cabin for a few minutes and take in the sweeping views of Earth, before the space plane glides back home and lands on a runway.

In contrast, the capsules used by SpaceX and Blue Origin are fully automated and parachute back down.

Like Virgin Galactic, Blue Origin aims for the fringes of space, quick ups-and-downs from West Texas. Blue Origin has launched 31 people so far, but flights are on hold following a rocket crash last fall. The capsule, carrying experiments but no passengers, landed intact.

SpaceX, is the only private company flying customers all the way to orbit, charging a much heftier price, too: tens of millions of dollars per seat. It’s already flown three private crews. NASA is its biggest customer, relying on SpaceX to ferry its astronauts to and from the International Space Station. since 2020.

People have been taking on adventure travel for decades, the risks underscored by the recent implosion of the Titan submersible that killed five passengers on their way down to view the Titanic wreckage. Virgin Galactic suffered its own casualty in 2014 when its rocket plane broke apart during a test flight, killing one pilot. Yet space tourists are still lining up, ever since the first one rocketed into orbit in 2001 with the Russians.

Branson, who lives in the British Virgin Islands, watched Thursday’s flight from a party in Antigua. He had held a virtual lottery to establish a pecking order for the company’s first 50 customers — dubbed the Founding Astronauts. Virgin Galactic said the group agreed Goodwin would go first, given his age and his Parkinson’s.
…

China will require all mobile app providers in the country to file business details with the government, its information ministry said, marking Beijing’s latest effort to keep the industry on a tight leash. 

The Ministry of Industry and Information Technology (MIIT) said late on Tuesday that apps without proper filings will be punished after the grace period that will end in March next year, a move that experts say would potentially restrict the number of apps and hit small developers hard. 

You Yunting, a lawyer with Shanghai-based DeBund Law Offices, said the order is effectively requiring approvals from the ministry. The new rule is primarily aimed at combating online fraud but it will impact all apps in China, he said. 

Rich Bishop, co-founder of app publishing firm AppInChina, said the new rule is also likely to affect foreign-based developers which have been able to publish their apps easily through Apple’s App Store without showing any documentation to the Chinese government. 

Bishop said that in order to comply with the new rules, app developers now must either have a company in China or work with a local publisher.  

Apple did not immediately reply to a request for comment. 

The iPhone maker pulled over a hundred artificial intelligence (AI) apps from its App Store last week to comply with regulations after China introduced a new licensing regime for generative AI apps for the country.  

The ministry’s notice also said entities “engaged in internet information services through apps in such fields as news, publishing, education, film and television and religion should also submit relevant documents.” 

The requirement could affect the availability of popular social media apps such as X, Facebook and Instagram. Use of such apps are not allowed in China, but they can be still downloaded from app stores, enabling Chinese to use them when traveling overseas. 

China already requires mobile games to obtain licenses before they launch in the country, and it had purged tens of thousands of unlicensed games from various app stores in 2020. 

Tencent’s WeChat, China’s most popular online social platform, said on Wednesday that mini apps, apps that can be opened within WeChat, must also follow the new rules. 

The company said that new apps must complete the filing before launch starting from September, while exiting mini apps have until the end of March.  

 
…

The White House on Wednesday said it had launched a multimillion-dollar cyber contest to spur use of artificial intelligence to find and fix security flaws in U.S. government infrastructure, in the face of growing use of the technology by hackers for malicious purposes.  

“Cybersecurity is a race between offense and defense,” said Anne Neuberger, the U.S. government’s deputy national security adviser for cyber and emerging technology.

“We know malicious actors are already using AI to accelerate identifying vulnerabilities or build malicious software,” she added in a statement to Reuters.

Numerous U.S. organizations, from health care groups to manufacturing firms and government institutions, have been the target of hacking in recent years, and officials have warned of future threats, especially from foreign adversaries.  

Neuberger’s comments about AI echo those Canada’s cybersecurity chief Samy Khoury made last month. He said his agency had seen AI being used for everything from creating phishing emails and writing malicious computer code to spreading disinformation.

The two-year contest includes around $20 million in rewards and will be led by the Defense Advanced Research Projects Agency, the U.S. government body in charge of creating technologies for national security, the White House said.

Google, Anthropic, Microsoft, and OpenAI — the U.S. technology firms at the forefront of the AI revolution — will make their systems available for the challenge, the government said.

The contest signals official attempts to tackle an emerging threat that experts are still trying to fully grasp. In the past year, U.S. firms have launched a range of generative AI tools such as ChatGPT that allow users to create convincing videos, images, texts, and computer code. Chinese companies have launched similar models to catch up.

Experts say such tools could make it far easier to, for instance, conduct mass hacking campaigns or create fake profiles on social media to spread false information and propaganda.  

“Our goal with the DARPA AI challenge is to catalyze a larger community of cyber defenders who use the participating AI models to race faster – using generative AI to bolster our cyber defenses,” Neuberger said.

The Open Source Security Foundation (OpenSSF), a U.S. group of experts trying to improve open source software security, will be in charge of ensuring the “winning software code is put to use right away,” the U.S. government said. 
…

U.S. President Joe Biden signed an executive order on Wednesday to impose restrictions on U.S. investments in some high-tech industries in China.

Biden’s executive order could again heighten tensions between the United States, the world’s biggest economy, and No. 2 China after a period in which leaders of the two countries have held several discussions aimed at airing their differences and seeking common ground.

The new restrictions would limit U.S. investments in such high-tech sectors in China as quantum computing, artificial intelligence and advanced semi-conductors, but apparently not in the broader Chinese economy, which recently has been struggling to advance.

Some investments in those sectors will require government notification; others will be prohibited, but details were not immediately clear.

“We’re pursuing a policy of de-risking with respect to the PRC by taking targeted national security actions, not decoupling our economies,” one official said, according to Reuters.

In a trip to China in July, Treasury Secretary Janet Yellen told Chinese Premier Li Qiang, “The United States will, in certain circumstances, need to pursue targeted actions to protect its national security. And we may disagree in these instances.”

Trying to protect its own security interests in the Indo-Pacific region and across the globe, National Security Adviser Jake Sullivan said in April that the U.S. has implemented “carefully tailored restrictions on the most advanced semiconductor technology exports” to China.

“Those restrictions are premised on straightforward national security concerns,” he said. “Key allies and partners have followed suit, consistent with their own security concerns.”

Sullivan said they are not, as Beijing has claimed, a “technology blockade.”
…

The company whose name became synonymous with remote work is joining the growing return-to-office trend.

Zoom, the video conferencing pioneer, is asking employees who live within a 50-mile radius of its offices to work onsite two days a week, a company spokesperson confirmed in an email. The statement said the company has decided that “a structured hybrid approach – meaning employees that live near an office need to be onsite two days a week to interact with their teams – is most effective for Zoom.”

The new policy, which will be rolled out in August and September, was first reported by the New York Times, which said Zoom CEO Eric Yuan fielded questions from employees unhappy with the new policy during a Zoom meeting last week.

Zoom, based in San Jose, California, saw explosive growth during the first year of the COVID-19 pandemic as companies scrambled to shift to remote work, and even families and friends turned to the platform for virtual gatherings. But that growth has stagnated as the pandemic threat has ebbed.

Shares of Zoom Video Communications Inc. have tumbled hard since peaking early in the pandemic, from $559 apiece in October 2020, to below $70 on Tuesday. Shares have slumped more than 10% to start the month of August. In February, Zoom laid off about 1,300 people, or about 15% of its workforce.

Google, Salesforce and Amazon are among major companies that have also stepped up their return-to-office policies despite a backlash from some employees.

Similarly to Zoom, many companies are asking their employees to show up to the office only part time, as hybrid work shapes up to be a lasting legacy of the pandemic. Since January, the average weekly office occupancy rate in 10 major U.S. cities has hovered around 50%, dipping below that threshold during the summer months, according to Kastle Systems, which measures occupancy through entry swipes.
…

U.S. police reform advocates have long argued that police-worn body cameras will help reduce officers’ excessive use of force and work to build public trust. But the millions of hours of footage that so-called “body cams” generate are difficult for police supervisors to monitor. As Shelley Schlender explains, artificial intelligence may help.
…

Pope Francis on Tuesday called for a global reflection on the potential dangers of artificial intelligence (AI), noting the new technology’s “disruptive possibilities and ambivalent effects.”  

Francis, who is 86 and said in the past he does not know how to use a computer, issued the warning in a message for the next World Day of Peace of the Catholic Church, falling on New Year’s Day.  

The Vatican released the message well in advance, as it is customary.  

The pope “recalls the need to be vigilant and to work so that a logic of violence and discrimination does not take root in the production and use of such devices, at the expense of the most fragile and excluded,” it reads.  

“The urgent need to orient the concept and use of artificial intelligence in a responsible way, so that it may be at the service of humanity and the protection of our common home, requires that ethical reflection be extended to the sphere of education and law,” it adds.  

Back in 2015, Francis acknowledged being “a disaster” with technology, but he has also called the internet, social networks and text messages “a gift of God,” provided that they are used wisely.  

In 2020, the Vatican joined forces with tech giants Microsoft MSFT.O and IBM IBM.N to promote the ethical development of AI and call for regulation of intrusive technologies such as facial recognition.
…

Two to tech groups on Monday backed TikTok Inc in its lawsuit seeking block enforcement of a Montana state ban on use of the short video sharing app before it takes effect on January 1.

NetChoice, a national trade association that includes major tech platforms, and Chamber of Progress, a tech-industry coalition, said in a joint court filing that “Montana’s effort to cut Montanans off from the global network of TikTok users ignores and undermines the structure, design, and purpose of the internet.”

TikTok, which is owned by China’s ByteDance, filed a suit in May seeking to block the first-of-its-kind U.S. state ban on several grounds, arguing it violates the First Amendment free speech rights of the company and users.
…

The use of sophisticated spyware to hack into the devices of journalists and human rights defenders during a period of conflict in Armenia has alarmed analysts.

A joint investigation by digital rights organizations, including Amnesty International, found evidence of the surveillance software on devices belonging to 12 people, including a former government spokesperson.

The apparent targeting took place between October 2020 and December 2022, including during key moments in the Nagorno-Karabakh conflict, Amnesty reported.

The region has been at the center of a decades-long dispute between Azerbaijan and Armenia, which have fought two wars over the mountainous territory.

Elina Castillo Jiménez, a digital surveillance researcher at Amnesty International’s Security Laboratory, told VOA that her organization’s research — published earlier this year — confirmed that at least a dozen public figures in Armenia were targeted, including a former spokesperson for the Ministry of Foreign Affairs and a representative of the United Nations.

Others had reported on the conflict, including for VOA’s sister network Radio Free Europe/Radio Liberty; provided analysis; had sensitive conversations related to the conflict; or in some cases worked for organizations known to be critical of the government, the researchers found.

“The conflict may have been one of the reasons for the targeting,” Castillo said.

If, as Amnesty and others suspect, the timing is connected to the conflict, it would mark the first documented use of Pegasus in the context of an international conflict.

Researchers have found previously that Pegasus was used extensively in Azerbaijan to target civil society representatives, opposition figures and journalists, including the award-winning investigative reporter Khadija Ismayilova.

VOA reached out via email to the embassies of Armenia and Azerbaijan in Washington for comment but as of publication had not received a response.

Pegasus is a spyware marketed to governments by the Israeli digital security company NSO Group. The global investigative collaboration, The Pegasus Project, has been tracking the spyware’s use against human rights defenders, critics and others.

Since 2021, the U.S government has imposed measures on NSO over the hacking revelations, saying its tools were used for “transnational repression.” U.S actions include export limits on NSO Group and a March 2023 executive order that restricts the U.S. government’s use of commercial spyware like Pegasus.

VOA reached out to the NSO Group for comment but as of publication had not received a response.

Castillo said that Pegasus has the capability to infiltrate both iOS and Android phones.

Pegasus spyware is a “zero-click” mobile surveillance program. It can attack devices without any interaction from the individual who is targeted, gaining complete control over a phone or laptop and in effect transforming it into a spying tool against its owner, she said.

“The way that Pegasus operates is that it is capable of using elements within your iPhones or Androids,” said Castillo. “Imagine that it embed(s) something in your phone, and through that, then it can take control over it.”

The implications of the spyware are not lost on Ruben Melikyan. The lawyer, based in Armenia’s capital, Yerevan, is among those whose devices were infected.

An outspoken government critic, Melikyan has represented a range of opposition parliamentarians and activists.

The lawyer said he has concerns that the software could have allowed hackers to gain access to his data and information related to his clients.

“As a lawyer, my phone contained confidential information, and its compromise made me uneasy, particularly regarding the protection of my current and former clients’ rights.” he said.

Melikyan told VOA that his phone had been targeted twice: in May 2021, when he was monitoring Armenian elections, and again during a tense period in the Armenia and Azerbaijan conflict in December 2022.

Castillo said she believes targeting individuals with Pegasus is a violation of “international humanitarian law” and that evidence shows it is “an absolute menace to people doing human rights work.”

She said the researchers are not able to confirm who commissioned the use of the spyware, but “we do believe that it is a government customer.”

When the findings were released this year, an NSO Group spokesperson said it was unable to comment but that earlier allegations of “improper use of our technologies” had led to the termination of contracts.

Amnesty International researchers are also investigating the potential use of a commercial spyware, Predator, which was found on Armenian servers.

“We have the evidence that suggests that it was used. However, further investigation is needed,” Castillo said, adding that their findings so far suggest that Pegasus is just “one of the threats against journalists and human rights defenders.”

This story originated in VOA’s Armenia Service.
…