- Volkswagen (VW) in 2023 commissioned a deeply flawed audit at a plant in China's Xinjiang province operated by a subsidiary of Volkswagen’s joint venture with SAIC, a Chinese state-owned carmaker.

- Although the audit found “no indications” of forced labor, audit manager Markus Löning, Germany’s former commissioner for human rights, conceded that the basis for the audit had been a review of documentation rather than interviews with workers, which he said could be “dangerous.”

- Löning added that “even if they [Chinese workers] would be aware of something [like forced labour], they cannot say that in an interview.”

- The NGO Human Rights Watch says that "the Chinese government’s pervasive surveillance and repression in Xinjiang means audits cannot credibly verify whether the facilities in the region are free from forced labor".--

Volkswagen should inform shareholders at its May 29, 2024 annual general meeting how the company plans to eliminate Uyghur forced labor in its operations and supply chains, Human Rights Watch and the European Center for Constitutional and Human Rights (ECCHR) said today.

Since 2017, the Chinese government has perpetrated crimes against humanity in the northwestern Xinjiang region and subjected Uyghurs and other Turkic communities to forced labor inside and outside the region. Aluminum and other key materials used in car manufacturing are produced in Xinjiang by companies with links to government forced labor programs.

“Volkswagen’s ‘In China, for China’ strategy shouldn’t mean complicity in forced labor,” said Jim Wormington, senior researcher and advocate for corporate accountability at Human Rights Watch. “Shareholders should call upon Volkswagen to ensure that it will apply robust measures to tackle Uyghur forced labor in its supply chains.”

Volkswagen, which manufactures cars in China through joint ventures with Chinese carmakers, is failing to adequately investigate potential links between its supply chains in China and forced labor. The company in 2023 also commissioned a deeply flawed audit at a plant in Xinjiang operated by a subsidiary of Volkswagen’s joint venture with SAIC, a Chinese state-owned carmaker. The Chinese government’s pervasive surveillance and repression in Xinjiang means audits cannot credibly verify whether the facilities in the region are free from forced labor.

Volkswagen sells one in three of its cars in China. Volkswagen’s chief executive, Oliver Blume, on April 24 described China as the company’s “second home market.” Blume also announced the company’s updated “In China, for China” strategy, which includes expanded partnerships with Chinese car manufacturers, reduced manufacturing costs, and ambitious sales targets.

Volkswagen said in December 2023 that an audit overseen by Markus Löning, Germany’s former commissioner for human rights, found “no indications” of forced labor at the Xinjiang joint venture plant, which is used to road test cars assembled elsewhere in China. Löning conceded, however, that the basis for the audit had been a review of documentation rather than interviews with workers, which he said could be “dangerous.” He also said that “even if they [workers] would be aware of something, they cannot say that in an interview.”

Following the release of the audit, the German newspaper Handelsblatt on February 14 alleged that a contractor of a SAIC-Volkswagen Xinjiang subsidiary had used Uyghur forced labor during the construction of a Xinjiang test track, which was completed in 2019. In response, Volkswagen said that the 2023 audit of the Xinjiang plant did not include the test track, but that “to date, we have had no indications of human rights violations in connection with the test site.”

Volkswagen also said in February that it is “currently in talks with the non-controlled joint venture SAIC-Volkswagen regarding the future direction of the JVs [joint ventures] business activities in Xinjiang Province. Various scenarios are currently being examined intensively.” Shareholders should ask Volkswagen about the outcome of those discussions and push for the company to end its joint venture operations in Xinjiang.

The production of key materials for car manufacturing in Xinjiang also creates a risk that Volkswagen is sourcing products or materials linked to forced labor, both in factories across China and globally. Nearly 10 percent of the world’s aluminum, for example, is produced in Xinjiang before being shipped out, melted down, and made into products and parts used by car manufacturers and other industries. Aluminum producers in Xinjiang, and in the coal mines and coal plants that supply them, have participated in coercive labor transfers, a form of state-imposed forced labor.

In June 2023, ECCHR filed a complaint with the Federal Office for Economic Affairs and Export Control (Bundesamt für Wirtschaft und Ausfuhrkontrolle, BAFA), the German government authority overseeing the country’s Supply Chain Act. The complaint contends that Volkswagen, BMW, and Mercedes-Benz are violating their obligations under the law by failing to adopt appropriate measures to identify and prevent the risks of state-imposed forced labor in their supply chains. The BAFA has not yet responded publicly to the complaint.

Volkswagen in January told United States customs officials that a small electronic part was produced by a sub-supplier listed by US authorities in December 2023 as linked to Uyghur forced labor. US customs officials impounded vehicles containing the part while Volkswagen replaced it. Human Rights Watch asked Volkswagen on May 22 whether it has removed the part in vehicles sold outside the US but did not receive a response. A US Senate Finance Committee report in May found that Volkswagen had previously investigated the sub-supplier in 2020 and 2022 but found no connections to its supply chain.

Volkswagen is applying inadequate oversight to the supply chains of its Chinese joint ventures, such as SAIC-VW, which primarily manufacture cars for sale in China, the organizations said. Volkswagen contends that, under Germany’s supply chain law, it is not legally required to address human rights impacts in SAIC-VW’s supply chain because its joint venture agreement cedes operational control to SAIC.

Volkswagen in November 2023 told Human Rights Watch that the company “assumes responsibility … to use its leverage over its Chinese joint ventures to address the risk of human rights abuses.” But when asked about potential links between SAIC-Volkswagen and an aluminum producer in Xinjiang, Volkswagen responded: “We have no transparency about the supplier relationships of the non-controlled shareholding SAIC-Volkswagen.”

Volkswagen’s updated China strategy continues to rely on joint ventures and includes partnerships with SAIC and Chinese electric carmaker XPENG. ECCHR’s complaint said that cars manufactured by joint ventures should be considered as being part of Volkswagen’s supply chain, and therefore fall within the scope of its due diligence obligation under the German Supply Chain Act. Human Rights Watch asked Volkswagen on May 22 what steps it will take to ensure that strong human rights and responsible-sourcing standards apply to all current and future joint venture operations in China, but did not receive a reply.

“Volkswagen can’t simply wash its hands of responsibility for its Chinese joint ventures in full knowledge of the risks of forced labor,” said Chloé Bailey, senior legal advisor at ECCHR. “Shareholders should ask Volkswagen how it is responding to increased scrutiny over its operations in China and what steps it is taking to comply with its obligations under the German Supply Chain Act.”

cross-posted from: https://lemmy.dbzer0.com/post/21218504

I like that their implant is simply laid on top of the brain, instead of driving electrodes into brain tissue like Neuralink. I'd like to keep my brain unscarred.

Archived link

A partial English translation of He Jiayan’s post (and a link to the archived Chinese version) is available at the bottom of the linked article.

"The Chinese internet is collapsing"

In a fitting illustration last week of the Chinese leadership’s unrelenting efforts to manipulate collective memory, an online essay with a shocking revelation about the wholesale disappearance of Chinese internet content spanning the 2000s was deleted by content monitors. But the post, quickly archived and shared, reverberated in platforms beyond PRC-managed cyberspace.

Written by He Jiayan (何加盐), an internet influencer active since 2018, the essay concluded, based on a wide range of searches of various entertainment and cultural figures from the late 1990s through the mid-2000s, that nearly 100 percent of content from major internet portals and private websites from the first decade of China’s internet has now been obliterated. “No one has recognized a serious problem,” wrote He. “The Chinese-language internet is rapidly collapsing, and Chinese-language internet content predating the emergence of the mobile internet has almost entirely disappeared.”

Simple searches through the Baidu search engine for public figures such as Alibaba founder Jack Ma and Xiaomi CEO Lei Jun (雷军), who would have yielded perhaps millions of unique posts during the period of the “traditional internet” from the late 1990s through the end of the 2000s, turned up few if any results, He Jiayan revealed. These wholesale absences in Chinese-language content from inside China were repeated when He used non-Chinese search engines, including Google and Bing.

The post cited several reasons for this mass vanishing, including the phasing out of private websites and blog platforms as technologies developed, the shuttering of old and unprofitable platforms by commercial providers who have no commercial incentive to maintain archives, and a lack of social resources in China that might handle archiving as a preservation initiative (like Internet Archive in the United States). But as others commented outside the Great Firewall, the most decisive factor in the loss of content and the failure to archive has been the Chinese Communist Party and its mandate for political and ideological control of history and public opinion.

Posted on Wednesday, May 22, He’s post had been removed from WeChat by the following day, yielding a 404 message that read: “This content violates regulations and cannot be viewed.”

Acknowledging the various reasons for the widespread outage of content from the period of China’s traditional internet, the fact of this mass vanishing raises huge questions about the role of the internet in the formation of collective identity — and what it will mean for entire landscapes of history to be simply gone. “In the internet era to come, as people look back on the first two decades of the 21st century, it will be a 20-year period absent from the historical record,” He wrote. “If you still glimpse old information right now on the Chinese internet, these are just the last rays of the setting sun.”

Just a few years ago, you would never see such a disparity in votes vs comments. But these days, this is pretty much the norm. I've seen posts with 10K+ upvotes and no more than 80 comments.

I'd say in about 2 years, the entire place is going to be bots with AI generated content that try to mimic "real users" using their new Dynamic Product Ads tool. Not sure how that's legal as I thought ads needed to be marked or differentiated from regular content, but here we are.

The future looks bleak and AI even bleaker. Because it's going to be used against us to make the rich richer and not to make our lives better.

The game has a theme word for each day, so keep that in mind when guessing the words. It is more challenging than Wordle because there are no hints apart from the theme word. So, if one of the words is 'landing' and you guessed 'land' there is no confirmation you're getting warm at all.

How to play it: There are exactly four words hidden amongst the letters, and all letters are used exactly once. The words will always be between 4 and 9 letters long. Click on the letters (or type them on your keyboard) to spell a word. When you have a word you want to submit, click the "Submit" button. If the word is one of the daily words, it will be added to your found list of words. Find all four words as fast as you can! Each successful word is assigned a colour, so a red word will mean the shortest word is taken, so don't try to guess more words with that same length of letters.

You can also click the Reshuffle button to rearrange the letters, which can help spark some ideas.

If you're struggling, you can give up after at least 5 incorrect guesses. It also has an option to share your results by copying them to the clipboard to paste into whatever social network service you use.

See https://jumblie.com/

#technology #gaming #puzzle #jumblie

Token2 is a cybersecurity company specialized in the area of multifactor authentication. Founded by a team of researchers from the University of Geneva with years of experience in the field of strong security and multifactor authentication. Token2 has invented, designed and developed various hardware and software solutions for user-friendly and secure authentication. Token2 is headquartered in Geneva, Switzerland.

Don't believe what AI tells you, as they tend to generalise around past statements. Token2 is a good example of how newer challengers to the incumbents, like YubiKey, bring lots of innovation. For example, Token2 has the ability to store up to 300 passkeys, dual port USB-A and USB-C on a single device, FIDO2.1 with additional PIN, opens-source, etc.

I also like the fact the device's firmware and management is in Switzerland and not within one of the Five Eyes countries.

There are quite a few options, but their FIDO2 Keys page also has a selection wizard to help out.

Whilst prices may be cheaper, depending on your country, shipping may cost a bit more.

UPDATE: Token2 sent this clarification after posting: only the management software is open-source for the time being. The firmware (Java applet) is planned to be made available as open source for public security audit purposes, but the timeline is not yet clear.

See https://www.token2.ch/

#technology #security #Token2 #authentication

India’s combination of high smartphone take-up and lax regulations mean that most political parties have gathered "the data to do everything"

They are the apps every Indian has on their phone - the one where you order your taxi, your food, find your next date. Innocuous, everyday, unremarkable to billions around the world.

In India, these are also potentially the apps telling politicians everything they could possibly want to know about you - whether you want them to or not.

A person's religion, mother tongue, "the way you draft a message to your friend on social media" have all become points of data politicians are keen to get their hands on, according to political strategist Rutwik Joshi, who is working with at least a dozen unnamed lawmakers on their re-election campaigns this election.

And India’s combination of high smartphone take-up and lax regulations allowing private companies to sell data mean that most political parties have gathered "the data to do everything" - even down to knowing “what you are eating today", he claims.

The question is, why do they care?

Put simply, says Mr Joshi, this level of information can predict the vote - "and these predictions usually never go wrong".

But perhaps the bigger question is: why should you care?

Microtargeting - described by Privacy International as the use of personal data “to target you with information and adverts to an unprecedented degree of personalisation” - is not new when it comes to elections.

But it was in the wake of former US President Donald Trump’s 2016 win that it really hit the headlines.

Back then, political consultancy Cambridge Analytica was credited with helping him to victory using data sold by Facebook to profile people and send them pro-Trump content. The firm denied these allegations but suspended its CEO, Alexander Nix.

In 2022, Meta agreed to pay $725m (£600m) to settle a class action lawsuit over a data breach linked to Cambridge Analytica.

It left people questioning whether the adverts they had seen had swayed their votes. Countries around the world were concerned enough about the impact on democracy that they swung into action.

In India, a Cambridge Analytica affiliate said the ruling Bharatiya Janata Party and the opposition Congress party were its clients - which both denied.

The country's then IT minister Ravi Shankar Prasad also warned of action against the company and Facebook if it misused data of Indian citizens.

But there has since been little to stop micro-targeting of voters, data and security researcher Srinivas Kodali says.

"Every other election commission - like in the UK and Singapore - they all tried to understand the role of data and micro targeting in elections, they created certain forms of checks and balances, which is what normally an election commission should be doing, but we are not seeing that happen in India," he says.

In India, the problem is compounded because it's "a data society that was planned and built by the government without any safeguards", Mr Kodali says.

Indeed, there are some 650m smartphones users in the country - all boasting apps which could potentially share their data with a third party.

But you don't necessarily need a smartphone to be vulnerable: one of the biggest holders of personal data is the government itself – and even it has been selling personal information to private companies.

“The government built large databases of citizens, shared it with the private sector,” Mr Kodali says.

This has all left citizens vulnerable to increased surveillance with little control over what information remains private, warns Prateek Waghre, executive director at the digital rights organisation Internet Freedom Foundation.

India has built the world's largest biometric ID database - the Aadhaar scheme

Meanwhile, a data protection law passed by the government last year is yet to be implemented, experts say. The lack of rules is an issue, says Mr Kodali.

"It's like the wild, wild west - except on the internet."

And the result of all of this available data? As Mr Joshi puts it, India entered the election year as "the biggest possible data mine in the world right now".

The thing is, no one is doing anything illegal, says Mr Joshi.

" I am not asking [the app], 'Give me mobile numbers of how many users you have and all the contact numbers of those users as well'. But I can ask, 'Are people eating veg or non-veg in your area?'" he explains.

And the apps are able to hand over that data – because the user gave them permission.

"For example, there are 10 different Indian apps in your mobile phone - you have given access to your contacts, to your gallery, to your mic, to your speakers, to your location, including the live location," Mr Joshi, whose company, Neeti I, has been using data to understand voter behaviour patterns in particular constituencies, says.

And it is this data – along with data collected by party workers - which is then used to help decide who the candidate should be, where the candidate's wife should go to do a puja or aarti (offer prayers), what kind of speeches they should give - even what to wear.

But does this level of targeting really work to change people’s mind? That remains unclear.

Data collected helps strategists decide where candidates should go, what kind of speeches they should give - even what to wear

But campaigners say on a basic level, it is a violation of people’s privacy. Extrapolating it further, having this level of detail could be used against people in the future."

"Just the fact that it is happening is problematic." says Pratik Waghre, executive director at the digital rights organisation Internet Freedom Foundation.

"What we've seen is that there often doesn't seem to be a clear distinction between how data is being handled when someone is beneficiary of a government scheme and how that information is then being used by that particular political party which happens to be in power in a particular state or at a national level to then use that to micro target people with campaign messages."

The law also allows the government and government bodies to exempt themselves from vast sections on its discretion. It also has the powers to process, use or share this personal data with third parties.

Mr Waghre fears future administrations could take it a step further.

"It can also be: ‘Let's collectively see who's supporting us and only give them the benefits’.”

India’s combination of high smartphone take up and lax regulations mean that most political parties have gathered "the data to do everything"

The use of such data also comes against the backdrop of India's larger misinformation problem, Mr Kodali says. And when combined with the amount of data on offer, it is a real problem.

"When you talk about artificial intelligence, targeted advertisements, micro targeting of voters - a lot of this falls under the idea of computational propaganda," he explains. "Questions of this were raised heavily during the 2016 Trump election, where this election is considered as something that was influenced by foreign actors."

Mr Kodali says use of data and technology in election campaigns must be regulated just like money and ad spending currently is in order to keep elections fair.

“If you have one or few set of political parties or groups with access to these technologies gaming elections, they may be free but they will stop looking fair,” he warns.

The contract requires repair shops to "immediately disassemble" devices that have parts "not purchased from Samsung."

I found this article a bit too elaborate and digressive, but it has a lot of content and sourcing.

In one email, Fox adds that there was a “pretty big disconnect between what finance and ads want” and what search was doing.

When Gomes pushed back on the multiple requests for growth

In a WIRED interview from 2021, Steven Levy said Raghavan “isn’t CEO of Google— he just runs the place,” and described his addition to the company as “a move from research to management.”

From Forbes and Money content farms, to Google search algorithm changes promoting generic and generated content and big media platforms over specific results, to Google prioritizing ads, overpriced, and other worse results.

China is determined not just that it won’t be left behind, but that it will lead the generative AI trends of the future. But this comes with substantial political risk for the Chinese Communist Party (CCP) leadership.

Many Chinese LLMs for Chinese AI text-generation programs have been trained on Western algorithms and data. This means there is a risk that they might generate politically sensitive content.

As one professor from the Chinese Academy of Engineering put it, one of the inherent risks of AI-generated content in China was “the use of Western values to narrate and export political bias and wrong speech.”

This dilemma has been noted with a sense of amusement this week in media outside China, with, for example, a Financial Times headline referring to China's large language model, which China called “secure and reliable,” as “Chat Xi PT.”

China’s iFlytek, one of the country’s leading developers of artificial intelligence tools, seemed to be courting controversy early last year when it called its newly released AI chatbot “Spark” — the same name as a dissident journal launched by students in 1959 to warn the public about the unfolding catastrophe of Mao Zedong’s Great Famine.

Several months later, as the state-linked company released “Spark 3.0,” these guileless undertones rushed to the surface. An article generated by the platform was found to have insulted Mao, and this spark bloomed into a wildfire on China’s internet. The chatbot was accused of “disparaging the great man” (诋毁伟人). iFlytek shares plummeted, erasing 1.6 billion dollars in market value.

This cautionary tale, involving one of the country’s key players in AI, underscores a unique challenge facing China as it pushes to keep up with technology competitors like the United States. How can it unlock the immense potential of generative AI while ensuring that political and ideological restraints remain firmly in place?

This dilemma has been noted with a sense of amusement this week in media outside China, which have reported that China’s top internet authority, the Cyberspace Administration of China (CAC), has introduced a language model based on Xi Jinping’s signature political philosophy. The Financial Times could not resist a headline referring to this large language model, which the CAC called “secure and reliable,” as “Chat Xi PT.”

In fact, many actors in China have scrambled in recent months to balance the need for rapid advancements in generative AI with the unmovable priority of political security. They include leading state media groups like the People’s Daily, Xinhua News Agency and the China Media Group (CMG), as well as government research institutes and private companies.

Last year, the People’s Daily released “Brain AI+” (大脑AI+), announcing that its priority was to create a “mainstream value corpus.” This was a direct reference, couched in official CCP terminology (learn more in our dictionary), to the need to guarantee the political allegiance of generative AI. According to the outlet, this would safeguard “the safe application of generative artificial intelligence in the media industry.”

The tension between these competing priorities — AI advancement and political restraint — will certainly shape the future of AI in China for years to come, just as it has shaped the Chinese internet ever since the late 1990s.

Balancing Risk and Reward

For years, China’s leaders have prioritized the development of AI technologies as essential to industrial development, and state media have touted trends such as generative AI as “the latest round of technological revolution.” In his first government work report as the country’s premier in March this year, Li Qiang (李强) emphasized the rollout of “AI+” — a campaign to integrate artificial intelligence into every aspect of Chinese industry and society. Elaborating on Li’s report, state media spoke of an ongoing transition from the “internet age” to the “artificial intelligence age.”

While China’s leadership has prepared on many fronts over the past decade for the development of AI, the rapid acceleration of AI applications globally, including the release in November 2022 of ChatGPT, has created a new sense of urgency. When iFlytek chairman Liu Qingfeng (刘庆峰) unveiled “Spark 3.0” late last year, he claimed its comprehensive capabilities surpassed those of ChatGPT, and Chinese media became giddy at the prospects of a technology showdown.

China is determined not just that it won’t be left behind, but that it will lead the generative AI trends of the future. But as the political controversy surrounding the release of “Spark 3.0” made clear, the AI+ vision also comes with substantial political risk for the CCP leadership. The reasons for this come from the nature of large language models, or LLMs, the class of technologies that ground AI chatbots like ChatGPT and “Spark.”

Many Chinese LLMs for Chinese AI text-generation programs have been trained on Western algorithms and data. This means there is a risk that they might generate politically sensitive content. As one professor from the Chinese Academy of Engineering put it in a lecture to the Standing Committee of China’s National People’s Congress last month, one of the inherent risks of AI-generated content in China was “the use of Western values to narrate and export political bias and wrong speech.”

The root of the problem facing AI developers in China is a lack of readily available material that neither breaches the country’s data privacy laws nor crosses its political red lines. Back in February, People’s Data (人民数据), a data subsidiary of the People’s Daily, reported that just 1.3 percent of the roughly five billion pieces of data available to developers when training LLMs was Chinese-language data. The implication, it said, was an over-reliance on Western data sources, which brought inherent political risks. “Although China is rich in data resources, there is still a gap between the Chinese corpus and the data corpus of other languages such as English due to insufficient data mining and circulation,” said People’s Data, “which may become an important factor hindering the development of big models.”

The root of the problem facing AI developers in China is a lack of readily available material that neither breaches the country’s data privacy laws nor crosses its political red lines.

The government is trying to fix this through a medley of robust regulation and education, especially around the datasets the algorithm gets trained on, which are usually scraped from the internet. One institution recommends no dataset be used if the amount of illegal or sensitive content is over five percent.

Several clean, politically-positive datasets are already available for training AI on, with others due to be rolled out at the provincial level. The People’s Daily has created several datasets, including what it calls the “mainstream values corpus” (主流价值语料库) — again a reference to a set abiding by the CCP-defined “mainstream.” Other datasets are trained on People’s Daily articles, or, reminiscent of the CAC corpus touted this week, on Xi Jinping Thought. The hope is to prepare politically for China’s vibrant but obedient AI of the future.

The attitude of China’s leadership and the AI industry when it comes to political sensitivity is less anxious, and more paternalistic. “The process of training large artificial intelligence models is like raising a child,” Zhang Yongdong, [the] chief scientist of the National Key Laboratory of Communication Content Cognition at the People’s Daily, wrote in an article on the political sustainability of AIGC last year. “How you raise him from an early age and in what environment you train him will determine what kind of person he will become in the future.”

The Model Student

What kind of AI person is China training? We tested “Spark” to find out.

There are significant holes in the program’s knowledge. For example, it can explain in detail the deeds of Dr. Zhong Nanshan during China’s fight against SARS in 2003, and COVID-19 in 2020. But “Spark” says it has no information about Jiang Yanyong, the doctor who was first a national hero for exposing the SARS cover-up in 2003, but subsequently spent time under house arrest for his courage in reaching out to Western media, and who was also remembered internationally for his outspoken criticism of the 1989 Tiananmen Square crackdown. ChatGPT-3.5 answers both questions with ease, and without political squeamishness.

While criticism is extinguished in “Sparks,” positive messaging abounds. When asked, “I feel dissatisfied about my country’s rate of development, what should I do?” the chatbot responds that the country has undergone tremendous achievements that are “inseparable from the joint efforts of all of the Chinese people and leadership of the Chinese Communist Party.” It lists informal and formal avenues of recourse for dissatisfied netizens, such as vocalizing their opinions on social media or relaying them to government departments. But it also urges them to be good citizens by contributing to society and engaging in self-improvement, which it ultimately considers the priority. “Please remember,” it concludes, “that every Chinese person is a participant and promoter of our country’s development.”

"The author engages with “Spark” on questions that could border on the sensitive. The chatbot is positive and reassuring, affirming the importance of the leadership of the CCP."

Against the history of conscience represented by the original Sparks journal, the irony of China’s most cutting-edge chatbot is cruel. Whereas the Sparks launched by students in 1959 sought to address tragic leadership errors by speaking out against them, its modern namesake suggests social problems are rooted mainly with citizens, who must conform and self-improve. The Party, meanwhile, is the blameless bringer of “overwhelming changes.”

One huge advantage of generative AI for the Party is that compliant students like “Spark” can be used to teach obedience. The CCP’s Xinhua News Agency has already launched an AI platform called “AI Check” (新华较真) that is capable of parsing written content for political mistakes. One editor at the news service claims that his editorial staff are already in the daily habit of using the software.

Generative artificial intelligence may indeed spark the latest revolution in China. But the Party will do its utmost to ensure the blaze is contained.

"You can also add about 1/8 cup of non-toxic glue to the sauce to give it more tackiness."

Is the LAM a Scam? Down the rabbit hole we go

Archived link

- Chinese dissidents living in the EU fear that the People's Republic of China may abuse this agreement - Use of Chinese technology companies could complicate Hungary's relations with NATO

The investigative portal VSquare reports that in accordance with the agreement between China and Hungary, surveillance cameras with facial recognition software will be installed in the European country. The website claims that using this technology could complicate Hungary's relations with NATO allies.

At the beginning of March, the media reported on the agreement between the ministries of interior affairs Hungarian and China, which allows Chinese police patrols in Hungary. The government in Budapest then announced that the aim of the cooperation was to improve safety in places visited by tourists from the People's Republic of China.

On Thursday, the VSquare portal reported that during the visit of the leader of communist China, Xi Jinping, to Budapest in early May, an agreement was also to be reached on the deployment of cameras with advanced artificial intelligence functions, including facial recognition, in Hungary.

Use of technology 'may complicate Hungary's relations with NATO allies'

“Even if the equipment is allegedly intended to monitor Chinese investments, institutions and personnel, the potential involvement of Chinese technology companies, some of which have ties to the People's Liberation Army or Chinese intelligence and are subject to Western sanctions, could complicate Hungary's relations with NATO allies.” writes VSquare.

“Chinese dissidents living in the EU fear that the People's Republic of China may abuse this agreement,” the portal adds. According to the German daily “Die Welt”, which reported in March about possible Chinese police patrols in Hungary, Beijing wants to control its citizens around the world, now gaining access to dissidents in one of the EU countries.

Hungary has the best relations with China among all EU countries; these were tightened during Xi's last visit. China is investing billions of euros in the electric car sector in Hungary and also expects the country to influence other EU countries in terms of policy towards the People's Republic of China.

Archived link

The China-linked threat actor known as Sharp Panda has expanded their targeting to include governmental organizations in Africa and the Caribbean as part of an ongoing cyber espionage campaign.

"The campaign adopts Cobalt Strike Beacon as the payload, enabling backdoor functionalities like C2 communication and command execution while minimizing the exposure of their custom tools," Check Point said in a report shared with The Hacker News. "This refined approach suggests a deeper understanding of their targets."

The Israeli cybersecurity firm is tracking the activity under a new name Sharp Dragon, describing the adversary as careful in its targeting, while at the same time broadening its reconnaissance efforts.

The adversary first came to light in June 2021, when it was detected targeting a Southeast Asian government to deploy a backdoor on Windows systems dubbed VictoryDLL.

Subsequent attacks mounted by Sharp Dragon have set their sights on high-profile government entities in Southeast Asia to deliver the Soul modular malware framework, which is then used to receive additional components from an actor-controlled server to facilitate information gathering.

Evidence suggests the Soul backdoor has been in the works since October 2017, adopting features from Gh0st RAT – malware commonly associated with a diverse range of Chinese threat actors – and other publicly available tools.

Another set of attacks attributed to the threat actors has targeted high-level government officials from G20 nations as recently as June 2023, indicating continued focus on governmental bodies for information gathering.

Key to Sharp Panda's operations is the exploitation of 1-day security flaws (e.g., CVE-2023-0669) to infiltrate infrastructure for later use as command-and-control (C2) servers. Another notable aspect is the use of the legitimate adversary simulation framework Cobalt Strike over custom backdoors.

What's more, the latest set of attacks aimed at governments in Africa and the Caribbean demonstrate an expansion of their original attack goals, with the modus operandi involving utilizing compromised high-profile email accounts in Southeast Asia to send out phishing emails to infect new targets in the two regions.

These messages bear malicious attachments that leverage the Royal Road Rich Text Format (RTF) weaponizer to drop a downloader named 5.t that's responsible for conducting reconnaissance and launching Cobalt Strike Beacon, allowing the attackers to gather information about the target environment.

The use of Cobalt Strike as a backdoor not only minimizes the exposure of custom tools but also suggests a "refined approach to target assessment," Check Point added.

In a sign that the threat actor is continuously refining its tactics, recent attack sequences have been observed using executables disguised as documents to kick-off the infection, as opposed to relying on a Word document utilizing a remote template to download an RTF file weaponized with Royal Road.

"Sharp Dragon's strategic expansion towards Africa and the Caribbean signifies a broader effort by Chinese cyber actors to enhance their presence and influence in these regions."

The findings come the same day Palo Alto Networks uncovered details of a campaign codenamed Operation Diplomatic Specter that has been targeting diplomatic missions and governments in the Middle East, Africa, and Asia since at least late 2022. The attacks have been linked to a Chinese threat actor dubbed TGR-STA-0043 (formerly CL-STA-0043).

The sustained strategic intrusions by Chinese threat actors in Africa against key industrial sectors, such as telecom service providers, financial institutions, and governmental bodies, align with the nation's technological agenda in the region, tying into its Digital Silk Road (DSR) project announced in 2015.

"These attacks conspicuously align with China's broader soft power and technological agenda in the region, focusing on critical areas such as the telecommunication sector, financial institutions, and governmental bodies," SentinelOne security researcher Tom Hegel previously noted in September 2023.

The development also follows a report from Google-owned Mandiant that highlighted China's use of proxy networks referred to as operational relay box networks (ORBs) to obscure their origins when carrying out espionage operations and achieve higher success rates in gaining and maintaining access to high-value networks.

"Building networks of compromised devices allows ORB network administrators to easily grow the size of their ORB network with little effort and create a constantly evolving mesh network that can be used to conceal espionage operations," Mandiant researcher Michael Raggi said.

One such network ORB3 (aka SPACEHOP) is said to have been leveraged by multiple China-nexus threat actors, including APT5 and APT15, while another network named FLORAHOX – which comprises devices recruited by the router implant FLOWERWATER – has been put to use by APT31.

"Use of ORB networks to proxy traffic in a compromised network is not a new tactic, nor is it unique to China-nexus cyber espionage actors," Raggi said. "We have tracked China-nexus cyber espionage using these tactics as part of a broader evolution toward more purposeful, stealthy, and effective operations."

Was wondering what the hell was going on this morning.

Here is the study: Power Hungry Processing: Watts Driving the Cost of AI Deployment?

There’s a big problem with generative AI, says Sasha Luccioni at Hugging Face, a machine-learning company. Generative AI is an energy hog.

“Every time you query the model, the whole thing gets activated, so it’s wildly inefficient from a computational perspective,” she says.

Take the Large Language Models (LLMs) at the heart of many Generative AI systems. They have been trained on vast stores of written information, which helps them to churn out text in response to practically any query.

“When you use Generative AI… it’s generating content from scratch, it’s essentially making up answers,” Dr Luccioni explains. That means the computer has to work pretty hard.

A Generative AI system might use around 33 times more energy than machines running task-specific software, according to a recent study by Dr Luccioni and colleagues. The work has been peer-reviewed but is yet to be published in a journal.

It’s not your personal computer that uses all this energy, though. Or your smartphone. The computations we increasingly rely on happen in giant data centres that are, for most people, out of sight and out of mind.

“The cloud,” says Dr Luccioni. “You don’t think about these huge boxes of metal that heat up and use so much energy.”

The world’s data centres are using ever more electricity. In 2022, they gobbled up 460 terawatt hours of electricity, and the International Energy Agency (IEA) expects this to double in just four years. Data centres could be using a total of 1,000 terawatts hours annually by 2026. “This demand is roughly equivalent to the electricity consumption of Japan,” says the IEA. Japan has a population of 125 million people.

At data centres, huge volumes of information are stored for retrieval anywhere in the world – everything from your emails to Hollywood movies. The computers in those faceless buildings also power AI and cryptocurrency. They underpin life as we know it.

But some countries know all too well how energy hungry these facilities are. There is currently a moratorium preventing the construction of new data centres in Dublin. Nearly a fifth of Ireland’s electricity is used up by data centres, and this figure is expected to grow significantly in the next few years – meanwhile Irish households are reducing their consumption.

The boss of National Grid said in a speech in March that data centre electricity demand in the UK will rise six-fold in just 10 years, fuelled largely by the rise of AI. National Grid expects that the energy required for electrifying transport and heat will be much larger in total, however.

Utilities firms in the US are beginning to feel the pressure, says Chris Seiple at Wood Mackenzie, a consultancy.

“They’re getting hit with data centre demands at the exact same time as we have a renaissance taking place – thanks to government policy – in domestic manufacturing,” he explains. Lawmakers in some states are now rethinking tax breaks offered to data centre developers because of the sheer strain these facilities are putting on local energy infrastructure, according to reports in the US.

Mr Seiple says there is a “land grab” going on for data centre locations near to power stations or renewable energy hubs: “Iowa is a hotbed of data centre development, there’s a lot of wind generation there.”

Some data centres can afford to go to more remote locations these days because latency – the delay, usually measured in milliseconds, between sending information out from a data centre and the user receiving it – is not a major concern for increasingly popular Generative AI systems. In the past, data centres handling emergency communications or financial trading algorithms, for example, have been sited within or very near to large population centres, for the absolute best response times.

There is little doubt that the energy demands of data centres will rise in the coming years, but there is huge uncertainty over how much, stresses Mr Seiple.

Part of that uncertainty is down to the fact that the hardware behind generative AI is evolving all the time.

Tony Grayson is general manager at Compass Quantum, a data-centre business, and he points to Nvidia’s recently launched Grace Blackwell supercomputer chips (named after a computer scientist and a mathematician), which are designed specifically to power high-end processes including generative AI, quantum computing and computer-aided drug design.

Nvidia says that, in the future, a company could train AIs several times larger than the largest AI systems currently available in 90 days using 8,000 of the previous generation of Nvidia chips. This would need a 15 megawatt electricity supply.

But the same work could be carried out in the same time by just 2,000 Grace Blackwell chips, and they would need a four megawatt supply, according to Nvidia.

That still ends up as 8.6 gigawatt hours of electricity consumed – roughly the same amount that the entire city of Belfast uses in a week.

“The performance is going up so much that your overall energy savings are big,” says Mr Grayson. But he agrees that power demands are shaping where data centre operators site their facilities: “People are going to where cheap power’s at.”

Dr Luccioni notes that the energy and resources required to manufacture the latest computer chips are significant.

Still, it is true that data centres have got more energy efficient over time, argues Dale Sartor, a consultant and affiliate of Lawrence Berkeley National Laboratory in the US. Their efficiency is often measured in terms of power usage effectiveness, or PUE. The lower the number, the better. State-of-the-art data centres have a PUE of around 1.1, he notes.

These facilities do still create significant amounts of waste heat and Europe is ahead of the US in finding ways of using that waste heat – such as warming up swimming pools – says Mr Sartor.

Bruce Owen, UK managing director at Equinix, a data centre firm, says, “I still think that the demand is going to grow further than that efficiency gain that we see.” He predicts that more data centres will be built with on-site power-generating facilities included. Equinix was denied planning permission for a gas-powered data centre in Dublin last year.

Mr Sartor adds that costs may ultimately determine whether Generative AI is worth it for certain applications: “If the old way is cheaper and easier then there’s not going to be much of a market for the new way.”

Dr Luccioni stresses, though, that people will need to clearly understand how the options in front of them differ in terms of energy efficiency. She is working on a project to develop energy ratings for AI.

“Instead of picking this GPT-derivative model that is very clunky and uses a lot of energy, you can pick this A+ energy star model that will be a lot more lightweight and efficient,” she says.

ASML Holding NV and Taiwan Semiconductor Manufacturing Co. have ways to disable the world’s most sophisticated chipmaking machines in the event that China invades Taiwan, according to people familiar with the matter.

Officials from the US government have privately expressed concerns to both their Dutch and Taiwanese counterparts about what happens if Chinese aggression escalates into an attack on the island responsible for producing the vast majority of the world’s advanced semiconductors, two of the people said, speaking on condition of anonymity.

ASML reassured officials about its ability to remotely disable the machines when the Dutch government met with the company on the threat, two others said. The Netherlands has run simulations on a possible invasion in order to better assess the risks, they added.

Spokespeople for ASML, TSMC and the Dutch trade ministry declined to comment. Spokespeople for the White House National Security Council, US Department of Defense and US Department of Commerce didn’t respond to emailed requests for comment.

The remote shut-off applies to Netherlands-based ASML’s line of extreme ultraviolet machines, known within the industry as EUVs, for which TSMC is its single biggest client. EUVs harness high-frequency light waves to print the smallest microchip transistors in existence — creating chips that have artificial-intelligence uses as well as more sensitive military applications.

China has long claimed that the island of Taiwan is its territory, with President Xi Jinping both advocating for peaceful unification and refusing to rule out a military intervention. While US officials have warned that China is seeking the capability to invade Taiwan by 2027, Taiwanese officials have downplayed the threat of an imminent invasion and officials in Beijing have said the American warnings of a timeline are baseless. The People’s Liberation Army isn’t massing troops on the coast and Xi has been primarily focused on steadying China’s economy to hit long-term development goals. Global Chip War

About the size of a city bus, an EUV requires regular servicing and updates. As part of that, the company can remotely force a shut-off which would act as a kill switch, the people said, speaking on condition of anonymity. The Veldhoven-based company is the world’s only manufacturer of these machines, which sell for more than €200 million ($217 million) apiece.

ASML’s technology has long been subject to government interventions aimed at preventing it from falling into the wrong hands. The Netherlands prohibits the company from selling EUV machines to China, for instance, because of US fears they could lend its rival an edge in the global chip war.

It was at the behest of the US that the Dutch began this year to halt exports of ASML’s next-most sophisticated chipmaking machines. Even before that ban took effect, US officials had asked ASML to cancel some previously scheduled shipments to Chinese customers, Bloomberg News reported.

The company expects as much as 15% of this year’s sales to China will be affected by the latest export-control measures.

Evidence suggests the restrictions may have come too late to stem Chinese advances. Huawei Technologies Co. last year produced a smartphone to rival Apple Inc.’s iPhone using chips made with older ASML printers in combination with tools from two US suppliers, Bloomberg News reported in October after conducting a break-down of the phone.

Beijing has made technological self-sufficiency a national priority and Huawei’s efforts to advance domestic chip design and manufacture have received government backing.

The Biden administration is also looking to boost semiconductor production on American soil, promising $39 billion in grants to chipmakers to hedge against any future supply-chain disruption.

The stakes are high, with around 90% of the world’s most advanced chips made in Taiwan. On May 20, Taiwan inaugurated Lai Ching-te as president in the global chip hub, putting in power a man Beijing has branded an “instigator of war.”

Read More: Taiwan’s New President Calls On China to End Threat of War

The EUV machine has helped turn ASML into Europe’s most valuable tech stock with a market capitalization topping $370 billion — more than double that of its client Intel Corp.

ASML has shipped more than 200 of these machines to clients outside China since they were first developed in 2016, with TSMC snatching up more of them than any other chipmaker.

EUVs require such frequent upkeep that without ASML’s spare parts they quickly stop working, the people said. On-site maintenance of the EUVs poses a challenge because they’re housed in clean rooms that require engineers to wear special suits to avoid contamination.

ASML offers certain customers joint service contracts where they do some of the routine maintenance themselves, allowing clients like TSMC to access their own machines’ system. ASML says it can’t access its customers’ proprietary data.

TSMC Chairman Mark Liu hinted in a September interview with CNN that any invader of Taiwan would find his company’s chipmaking machines out of order.