Researchers at the University of Amsterdam conducted a novel experiment, creating a social network populated exclusively by artificial intelligence bots. The primary objective was to analyze how digital dynamics could be influenced by different moderation measures. The underlying idea was to test the effectiveness of tools capable of reducing two of the most serious problems of contemporary social media : polarization and toxicity in online interactions. To this end, six different strategies were tested. These included switching to a chronological feed, increasing the visibility of dissenting opinions, hiding subscriber statistics, and removing profile descriptions. For example, sorting the news feed chronologically reduced

Specialists at CrowdStrike conducted a series of experiments with the Chinese artificial intelligence system DeepSeek, testing its code generation based on query terms. They found that the results directly depended on the identity of the customer or associated organization. If the queries included neutral scenarios or mentioned the United States, the model produced clean, well-structured, and attack-resistant code. However, as soon as the project was linked to topics that provoked a negative reaction from the Chinese government, the quality of the solutions declined dramatically. The most notable examples involved queries from Falun Gong practitioners and organizations that mentioned Tibet, Taiwan, or the

Since March 2025, Insikt Group of Recorded Future observed CopyCop (also known as Storm-1516), a covert Russian influence network, creating at least 200 new fake media websites targeting the United States, France, and Canada, as well as websites impersonating media brands, parties, and political movements in France, Canada, and Armenia. CopyCop also created a regionalized network of websites posing as a fictitious fact-checking organization, publishing content in Turkish, Ukrainian, and Swahili, languages never before featured on the network. Including the 94 German-facing websites reported by Insikt Group in February 2025, this adds up to over 300 websites created by CopyCop operators since

Former CIA agent Dale Britt Bandler is at the center of a high-profile criminal case: he used his access to the agency’s top-secret systems, turning them into a “personal Google” for private profit. Court documents reveal a scheme under which the 68-year-old returned to the agency as a contractor after retiring in 2014 and began exchanging data with lobbying firms and foreign clients. From 2017 to 2020, Bandler earned approximately $360,000 while remaining a CIA agent with Top Secret/SCI clearance. This status granted him access to some of the U.S. government’s most closely guarded secrets. Among his clients was a foreign national

Microsoft has unveiled a new generation of data centers designed specifically for artificial intelligence. The flagship facility is called Microsoft Fairwater and is located in Wisconsin. It is the company’s largest and most technologically advanced facility, which Microsoft calls its “AI factory.” This is the first of a series of similar data centers the company is building in the United States and abroad. Fairwater occupies 315 acres in Mount Pleasant, New York. The site comprises three buildings, totaling approximately 1.2 million square feet. The complex required 46.6 miles of piling, 26.5 million pounds of steel, and 120 miles of underground cabling. Unlike

Microsoft has closed a critical bug that could have severely compromised its cloud environments: Dutch researcher Dirk-Jan Mollema discovered two interconnected flaws in the Entra ID (formerly Azure Active Directory) identity management service that, when combined, could have allowed an attacker to obtain global administrator rights and effectively take control of any Azure tenant. The first issue involved a little-known mechanism for issuing internal tokens, so-called Actor Tokens, used for authentication. The second involved a legacy Azure AD Graph interface that incorrectly verified the tenant a request was coming from, causing it to accept tokens from other users. The combination of these

Google DeepMind and OpenAI’s artificial intelligence have achieved gold-level results in a competition dubbed the “Programming Olympics.” The companies’ models demonstrated a level of proficiency comparable to top competitors in the finals of the International Collegiate Programming Contest (ICPC), held in early September. According to OpenAI, its latest model, GPT-5, solved all 12 problems, 11 of them on the first try. The company claims this result would have secured the system first place. DeepMind, meanwhile, reported that its reasoning model Gemini 2.5 Deep Think achieved the second-best result, solving a task that no human could have solved. AI Victory at the ICPC

Two young men involved in the Scattered Spider group have been charged as part of the National Crime Agency investigation into a cyber attack on Transport for London (TfL). On August 31, 2024, TfL was the subject of a network intrusion that investigators believe was carried out by members of the criminal collective. Thalha Jubair, 19, from East London, and Owen Flowers, 18, from Walsall, West Midlands, were arrested on Tuesday (September 16) at their home addresses by the NCA and City of London Police. Both appeared today (September 18) at Westminster Magistrates Court, after the Crown Prosecution Service authorised their charges

Chinese authorities have ordered the country’s largest tech companies, including Libaba, ByteDance, and Tencent, to stop purchasing Nvidia GPUs. The order, issued by the state cybersecurity regulator and the Cyberspace Administration of China, also applies to the RTX Pro 6000D models, designed specifically for the Chinese market. Companies were ordered not only to suspend testing of new solutions but also to cancel existing orders. Recall that in 2024, Nvidia generated approximately $17.1 billion in revenue from China, equivalent to 13% of the company’s total revenue, which amounted to approximately $130 billion. This move is part of Beijing’s broader policy aimed at reducing

Imagine waking up one morning and discovering that your sensitive financial data—contracts, customer lists, investment strategies—is exposed on a hidden site on the dark web, with a timer threatening to make it public unless you pay a ransom. This is exactly what happened to ten asset management firms in South Korea, victims of the “Korean Leak” campaign orchestrated by the Qilin ransomware group. The “Korean Leak” Campaign: The Ten Victims and the Exposed Data Through CTI and OSINT monitoring—with sources such as Ransomware.live and H4ckmanac—it emerged that the Qilin group targeted the South Korean asset management sector. Direct verification on their onion