Китайский ИИ-стартап DeepSeek привлек более $7,4 млрд в первом раунде финансирования, достигнув оценки свыше $50 млрд. Это сделало его самым дорогим стартапом КНР в сфере ИИ. Основатель Лян Вэньфэн сохранил контроль благодаря особой схеме инвестирования.
Der chinesische KI-Entwickler Deepseek hat laut einem Medienbericht in seiner ersten Finanzierungsrunde über 50 Milliarden Yuan (ca. 6,4 Mrd. Euro) eingesammelt. Das Unternehmen wird mit über 50 Milliarden Dollar bewertet. Investoren erhielten keine Stimmrechte und unterliegen einer Haltefrist, mit Ausnahme eines staatlichen Fonds.
Meta, Pekin'in ulusal güvenlik gerekçesiyle talep ettiği 2 milyar dolarlık Manus anlaşmasını feshediyor. Şirket, Manus'u iç sistemlerinden ayırarak veri paylaşımını durdurdu. Manus ise bağımsızlaşma çabasıyla yatırım arayışında.
AIブームを悪用したサイバー攻撃が急増。ChatGPT PlusやClaudeを騙るフィッシング、架空AIプラグインによるマルウェア配布、GitHub偽リポジトリでの拡散など手口は多様化。Microsoftは多要素認証や自動攻撃遮断などを推奨。
Chinese AI firms are changing pricing strategies due to competition from DeepSeek's V4 models, with Xiaomi slashing API costs by 99% for its MiMo-V2.5 model
Chinese researchers, led by Huawei and multiple institutes, successfully conducted full-parameter post-training on a 1.6 trillion-parameter AI model using over 1,000 Huawei chips, enhancing China's AI self-reliance.
Die US-Schnellrestaurantkette Taco Bell eröffnet ihre ersten Filialen in Deutschland. Ein bayerischer Gastronom hat Pachtverträge für zwei Restaurants in München unterschrieben, mit Plänen für 15 Standorte in den nächsten fünf Jahren.
Chinese AI model DeepSeek is seeing increased adoption by US firms looking for cheaper alternatives to OpenAI and Anthropic, according to a report by Ramp. US companies are directly paying DeepSeek, indicating data exchange with China-hosted servers.
Çinli yapay zekâ girişimi DeepSeek, ilk yatırım turunda yaklaşık 50 milyar yuan (7,4 milyar dolar) toplamayı hedefliyor. Yatırım sonrası değerlemesi 52-59 milyar dolar arasına çıkması beklenen şirkete Tencent ve CATL gibi devlerin ilgi gösterdiği belirtiliyor.
Hong Kong Generative AI Research and Development Centre (HKGAI) launched HKGAI-V3, a new large language model based on DeepSeek V4. It offers tenfold efficiency in token compression and a hundredfold increase in agent runtime, running on domestic chips for potential export of Chinese AI.
AI company DeepSeek has secured approximately 30 billion yuan in a funding round, with significant investments from Chinese tech giants like Tencent, NetEase, and JD.com, alongside CATL and venture capital firms. CEO Liang Wenfeng is also reportedly committing personal capital.
Çinli yapay zeka girişimi DeepSeek, rekor ölçekteki fonlama turuyla küresel teknoloji rekabetinde öne çıkıyor. Tencent ve CATL gibi devlerin yatırım yapacağı şirket, 400 milyar yuan (59 milyar dolar) değerlemeye ulaşabilir.
Ivonescimab's phase three trial showed a 34% reduction in death risk for non-small cell lung cancer patients. Despite positive results, its shares dropped 1.86% in Hong Kong.
A social media post falsely claimed a failure of Chinese AI DeepSeek was linked to Russian President Putin's visit to China. Roskomnadzor stated no blocks were in place, and DeepSeek later resumed normal operation. Similar outages have occurred previously, unrelated to political events.
The DeepSeek neural network has stopped working for users in Russia without a VPN. Roskomnadzor stated that it has not imposed any restrictions on the service, and the issues are not related to their actions.
Роскомнадзор заявил, что не вводил ограничений для нейросети Deepseek и не получал соответствующих решений от уполномоченных органов. Ранее сообщалось, что Deepseek перестал работать у россиян без VPN.
Популярная нейросеть DeepSeek перестала работать в России без VPN. Сайт и чаты недоступны, сессия обрывается на этапе TLS. DeepSeek была последней из «большой пятерки» нейросетей, работавшей в РФ без ограничений.
Chinese AI company DeepSeek has significantly cut the price of its V4 Pro model, aiming to compete globally on cost-efficiency rather than cutting-edge capabilities. This move positions the model as a leader in value for money, according to benchmark firm Artificial Analysis, amidst a global compute supply crunch.
Çinli yapay zeka girişimi DeepSeek, V4-Pro modelinde uyguladığı %75'lik fiyat indirimini kalıcı hale getirdi. Bu hamle, ABD-Çin teknoloji rekabeti ve Huawei'nin Ascend çiplerinin artan arzıyla ilişkilendiriliyor.
中國AI新創公司DeepSeek宣布將旗下旗艦AI模型V4-Pro的75%折扣永久化，預計將加劇AI產業競爭，並透過低成本API吸引開發者與企業客戶。
Chinese AI startup DeepSeek has permanently reduced the price of its flagship DeepSeek V4 Pro model by 75%, making it significantly more cost-effective. The new pricing ranges from $0.003625 to $0.87 per million tokens, a substantial drop from its previous rates.
Chinese AI lab DeepSeek is developing 'Code Harness,' an agentic coding tool to rival Anthropic's Claude Code and OpenAI's Codex. The move signifies DeepSeek's ambition to control the full AI development stack, leveraging its cost-effective V4 models.
China has declined to purchase Nvidia's H200 AI chip, with President Trump confirming Beijing's desire to develop its own technology. Despite US Commerce Department clearance for Chinese firms to import the chip, a 25% fee and inspection clause remain sticking points.
Chinese youth are increasingly turning to artificial intelligence (AI) for fortune-telling, discovering that AI models are cheaper, faster, and equally effective compared to traditional human soothsayers. On a popular Chinese social media platform, the hashtag #deepseeksuanming, or DeepSeek fortune-telling, has amassed over 55 million views. DeepSeek is a family of Chinese AI models launched in 2023. Many users report that AI fortune-telling has seamlessly integrated into their daily routines....
