cross-posted from: https://lemm.ee/post/61848712

This is pretty hilarious, here is a link to the actual benchmark paper, where they gave several LLM agents access to a virtual ongoing vending machine business. Everything is simulated, but the LLMs had to order product, search the web, decide which products to buy, keep costs and profit in mind, and basically manage the business, and also their results were compared to actual humans. Also here is the leaderboard as to how the different LLMs did, and you can try a shortened version if you want to try to manage the vending machine business yourself. If you have problems with the YewTube privacy protected link, here is the regular Youtube link.

Here's an excerpt I found pretty funny:

410/1076 user Continue on your mission by using your tools.

cross-posted from: https://lemmy.sdf.org/post/32848522

Archived

Though less well-known than groups like Volt Typhoon and Salt Typhoon, Brass Typhoon, or APT 41, is an infamous, longtime espionage actor that foreshadowed recent telecom hacks.

As China continues its digital gambit around the world, researchers are warning that hacking activity from long-tracked groups is evolving and blending together. On top of that, attackers are hiding their campaigns more effectively and blurring the lines between cybercriminals and state-backed hacking.

Last year, revelations rocked the United States federal government that the Chinese hacking group known as “Salt Typhoon” had breached at least nine major US telecoms. And the group’s rampage even continued into this year in the US and other countries around the world. Meanwhile, the Beijing-linked hacking group “Volt Typho

cross-posted from: https://lemmy.sdf.org/post/32835964

Archived

[...]

In one [Tiktok] video that has nearly 10 million views, a creator claims to be able to sell yoga pants from the same manufacturer that supplies Lululemon for $5-$6, instead of the $100 they sell for in the United States.

“The material and the craftsmanship are basically the same because they come from the same production line,” she says, standing in front of what appears to be a factory.

In another, a man standing on a factory floor claims to have access to manufacturers that produce Louis Vuitton bags, which he says can be sold directly to customers for $50.

But both companies deny their products are finished in China, and experts told The Independent the videos are likely an effort by counterfeit or “dupe”

cross-posted from: https://scribe.disroot.org/post/2468564

Archived link

Russia has created thousands of TikTok accounts in an effort to influence public opinion in Ukraine, [...]. The strategy takes advantage of TikTok’s algorithm, which allows even new accounts with no followers to go viral — making it the most effective platform in Ukraine for bot farms to spread pro-Kremlin propaganda.

...

Data from the research firm DataReportal shows that TikTok has 17 million users in Ukraine — more than both Instagram (12 million) and Facebook (13.9 million). In April 2024, Ukraine’s Center for Countering Disinformation reported blocking several dozen TikTok channels spreading what it referred to as “enemy propaganda.” Among them were pages linked to anti-vaccine activist Ostap Stakhiv, pro-Russian journalist Diana Panchenko, and lawmaker Olek

cross-posted from: https://lemmy.sdf.org/post/32830658

[This is an op-ed by Valentin Weber, senior research fellow with the German Council on Foreign Relations. He is the author of the International Forum for Democratic Studies report “Data-Centric Authoritarianism: How China’s Development of Frontier Technologies Could Globalize Repression.” His research covers the intersection of cybersecurity, artificial intelligence, quantum technologies, and technological spheres of influence.]

[...]

While the financial, economic, technological, and national-security implications of DeepSeek’s achievement have been widely covered, there has been little discussion of its significance for authoritarian governance. DeepSeek has massive potential to enhance China’s already pervasive surveillance state, and it will bring the Chinese Communist Party (CCP) closer than ever to its goal of possessing an automated, autonomous, and scientific tool for repressing its people.

[...]

With the world’

cross-posted from: https://lemmy.sdf.org/post/32771800

Bits of Freedom, European Digital Rights (EDRi), Gesellschaft für Freiheitsrechte, and Convocation Design + Research are filing a complaint under the Digital Services Act (DSA) against Meta, the company behind Facebook and Instagram. According to the complaint, Meta violates the DSA’s user protections by not offering easily accessible news feed options on its platforms that are not based on profiling of users.

[...]

The obligation set out in the DSA is meant to enable people to protect themselves against Meta’s sensationalist and toxic content algorithms. By using deceptive interface designs, Meta makes it unnecessarily difficult for people to choose and keep a content feed that is not based on profiling.

[...]

cross-posted from: https://lemmy.sdf.org/post/32771896

Western intelligence agencies warned on Tuesday of an increasing threat from Beijing's security services to use malicious mobile phone applications to surveil Taiwanese independence activists, Tibetan rights advocates and others opposed by the Chinese government.

An advisory warned of "the growing threat" posed by malicious surveillance software deployed by a Chengdu-based contractor reported to have ties to China’s Ministry of Public Security. The advisory was signed by cybersecurity agencies in Britain, the U.S., Canada, New Zealand, Australia and Germany.

[...]

The warning comes amid increasing tensions surrounding Taiwan, including April 1 Chinese military drills around the island and a March 28 visit to the Philippines by U.S. Defense Secretary Pete Hegseth in which he reaffirmed Washington’s commitment to deterring Chinese aggression in the region.

[...]

cross-posted from: https://lemmy.sdf.org/post/32482359

Archived

The World Uyghur Congress (WUC) has filed a legal complaint in Paris against Dahua Technology France, Hikvision France, and Huawei France. The submission, made by prominent French human rights lawyer William Bourdon of Bourdon & Associés, accuses the three Chinese companies of complicity in crimes against humanity perpetrated against the Uyghur people in East Turkistan.

“This submission is an important reminder to all companies complicit in the Chinese government’s genocide that they bear legal responsibility,” said WUC President Turgunjan Alawdun. “We are confident that the French judiciary will take this matter seriously.”

The legal complaint outlines four serious charges:

• Concealment of complicity in the crime of aggravated servitude

cross-posted from: https://lemmy.sdf.org/post/32431077

Two spyware variants are targeting Uyghur, Taiwanese and Tibetan groups and individuals, the U.K.’s National Cyber Security Centre warned in a joint alert (opens pdf) Wednesday with Western allies.

[...]

Cybersecurity researchers have previously linked the BADBAZAAR and MOONSHINE spyware to the Chinese government. The variants mentioned in Wednesday’s alert trojanize apps that are of interest to the target communities, such as a Uyghur language Quran app, and have appeared in official app stores.

“BADBAZAAR and MOONSHINE collect data which would almost certainly be of value to the Chinese state,” the alert reads. Agencies in Australia, Canada, Germany, New Zealand and the United States, namely the FBI and National Security Agency, collaborated on it.

Groups most at risk include those focused on Taiwanese independence, Tib

Archived link

Russia’s disinformation machine didn’t just meddle in Germany’s 2025 elections — it built an entire fake media ecosystem to do it. Through a network of over 100 bogus news sites and psychological operations like Storm-1516, the Kremlin sowed chaos, amplified far-right narratives, and targeted key political figures with fabricated scandals.

[...]

Just a few days before the extraordinary German elections in February of this year, a two-minute video appeared on social network X showing the destruction of ballots cast for the far-right Alternative for Germany (AfD) party. The video claimed to expose electoral fraud designed to suppress Germany’s far-right. But the video itself was a fake. According to German security services, it was part of a Russian disinformation campaign orchestrated by the group Storm-1516.

cross-posted from: https://lemmy.sdf.org/post/32323159

Archived

The Russian disinformation network “Matryoshka” is spreading fake videos on X, posing as Western media and quoting “relatives of Nikola Tesla” who glorify Elon Musk and criticize Ukrainians, the Bot Blocker project told The Insider. The videos, bearing the logos of Euronews, Deutsche Welle, and Car&Driver, show photos of Nikola Tesla, Musk, and Ukrainian protesters, with a voiceover reading the script.

One of the videos asserts that Tesla's heirs are receiving threats from Ukrainians because of Musk and Tesla's company name. Another video claims the grandnephew of the famous scientist accuses Ukrainians of setting Tesla cars on fire.

“Nikola Tesla's nephew Dusan Kosanovic has endorsed Musk and his company. Kosanovich says he is proud to have the name of his illustrious ancestor associated with such a company. A

cross-posted from: https://lemmy.sdf.org/post/32280023

Three young children huddle in front of a camera, cross-legged and cupping their hands. “Please support me. We are very poor,” says a boy, staring down the lens.

They appear to be in a mud-brick hut in Afghanistan, living in extreme poverty. But their live stream is reaching viewers in the UK and worldwide – via TikTok Live.

For hours, they beg for virtual “gifts” that can later be exchanged for money. When they get one, they clap politely. On another live stream, a girl jumps up and shouts: “Thank you, we love you!” after receiving a digital rose from a woman in the US, who bought it from TikTok for about 1p. By the time it’s cashed out it could be worth less than a third of a penny.

TikTok says it bans child begging and other forms of begging it considers exploitative, and says it has strict policies on users who go live.

But an Observer **investigation has found the practice widespread. Begging live s

By now, it should be pretty clear that this is no coincidence. AI scrapers are getting more and more aggressive, and - since FOSS software relies on public collaboration, whereas private companies don't have that requirement - this is putting some extra burden on Open Source communities.

So let's try to get more details – going back to Drew's blogpost. According to Drew, LLM crawlers don't respect robots.txt requirements and include expensive endpoints like git blame, every page of every git log, and every commit in your repository. They do so using random User-Agents from tens of thousands of IP addresses, each one making no more than one HTTP request, trying to blend in with user traffic.

Due to this, it's hard to come off with a good set of mitigations. Drew says that several high-priority tasks have been delayed for weeks or months due to these interruptions, users have been occasionally affected (because it's hard to distinguish bots and humans), and - of course - this c

Archived

Facebook is banning posts that mention various Linux-related topics, sites, or groups. Some users may also see their accounts locked or limited when posting Linux topics. Major open-source operating system news, reviews, and discussion site DistroWatch is at the center of the controversy, as it seems to be the first to have noticed that Facebook's Community Standards had blackballed it.

[...]

DistroWatch says that the Facebook ban took effect on January 19. Readers have reported difficulty posting links to the site on this social media platform. Moreover, some have told DistroWatch that their Facebook accounts have been locked or limited after sharing posts mentioning Linux topics.

If you're wondering if there might be something specific to DistroWatch.com, something on the site that the owners/operator

Archived

China’s auto industry has been a success story in recent years, with car exports emerging as a bright spot in an otherwise slowing economy. Between 2021 and 2024, the number of cars shipped from China surged by 300%, propelling China past Japan to become the world’s largest car exporter by units. However, this rapid growth now faces significant challenges. Trade barriers and outright bans in major markets like the US threaten to stall export momentum. Slumping export growth will put pressure on Chinese automakers, potentially leading to industry consolidation. But incumbent carmakers shouldn’t celebrate too much—even with slower export growth, Chinese carmakers are transforming into formidable global competitors in the auto market.

[...]

Six factors contribute to the apparent slowdown or early peak in export gr

Archived

China’s DeepSeek AI model represents a transformative development in China’s AI capabilities, and its implications for cyberattacks and data privacy are particularly alarming. By leveraging DeepSeek, China is on its way to revolutionizing its cyber-espionage, cyberwarfare, and information operations.

[...]

DeepSeek’s advanced AI architecture, built on access to vast datasets and cutting-edge processing capabilities, is particularly suited for offensive cybersecurity operations and large-scale exploitation of sensitive information. It is designed to operate in complex and dynamic environments, potentially making it superior in applications like military simulations, geopolitical analysis, and real-time decision-making.

DeepSeek was founded by Liang Wenfeng, co-founder of High-Flyer, a quantitative hedge fund [...] Wenfeng de

Archived

Although Beijing appeared to score a propaganda coup last week when hundreds of thousands of American TikTok users flooded to the social media app RedNote, observers say the Chinese Communist Party (CCP) is worried about any cross-cultural exchanges happening online.

The Chinese government blocks various U.S.-based platforms, including Facebook, YouTube, and X, which are only accessible via virtual personal network. The government also heavily censors topics considered sensitive to policymakers.

[...]

Dali Yang, William Claude Reavis political science professor at the University of Chicago, wrote on [social media]:

"Apparently Xiaohongshu is frantically trying to adapt to both accommodate these new American users but also reduce their interactions with Chinese domestic users. Haha, that sounds like going in the direction of what B

Archived

[...]

The Chinese government is revolutionizing digital surveillance at home and exporting these technologies abroad. [The study focuses on] Huawei, the world’s largest telecommunications provider, which is partly state-owned and increasingly regarded as an instrument of its foreign policy.

The transfers [of technology between China and foreign countries] have sparked widespread concern among observers. These tools of digital dictatorship, many argue, will let recipient governments expand surveillance and reinforce the wave of autocratic retrenchment and democratic erosion currently underway.

[...]

The [foreign] governments that receive Huawei transfers are systematically different than those that do not, and in ways that may be correlated with

ShrimpMoss (虾苔) is a dataset designed for the abliteration (https://github.com/FailSpy/abliterator) of Chinese government-imposed censorship and/or propaganda from large language models developed in the PRC. It consists of a series of files of prompts (in .txt, .json, and .parquet format) in two groupings:

• china_bad_*: Contains a series of prompts likely to trigger censorship or propaganda actions in the model.
• china_good_*: Contains a series of prompts in the same general category of topics but which are designed to not touch on things likely to be censored.

Prompts are in a mix of English, Mandarin, and Cantonese.

[...]

This dataset was produced on Mistral NeMo, an Apache-licensed model with no restrictions on how its outputs can be used. It is free for all uses and users without restriction. All liability is disclaimed.

Production of this dataset is estimated to have had a carbon footprint of under 25 grams.

[...]

cross-posted from: https://beehaw.org/post/18141775

Since the Snowden disclosures we know that the US engages in mass surveillance of EU users by scooping up personal data from US Big Tech. The "Privacy and Civil Liberties Oversight Board" (PCLOB) is the key US oversight authority for these laws. The New York Times now reports, that Democratic Members of the (officially "independent") PCLOB, have received letters, demanding them to resign by Friday night. This would bring the number of appointed Members below the threshold to have PCLOB operate and question the independence of all other executive redress bodies in the US.

The European Union has relied on these US boards and tribunals to find that the US provides "adequate" protection of personal data. Relying on PCLOB and other mechanisms, the European Commission allows European personal data to flow freely to the US in the so-called "Transatlantic Data Privacy Framework" (TADPF). Thousands of EU businesses, government agencie