DeepSeek AI News & Updates

DeepSeek Valuation Soars to $45B in First Funding Round Amid Chinese AI Competition

DeepSeek is raising its first venture capital round at a potential $45 billion valuation, led by Chinese state investment funds and tech giants Tencent and Alibaba. The Chinese AI lab gained prominence for developing efficient large language models that match top U.S. models while using significantly less compute and running on Huawei chips. The funding aims to retain talent through equity compensation amid intense competition for AI researchers.

DeepSeek Releases V4 Models With 1.6 Trillion Parameters, Approaching Frontier Performance at Lower Cost

Chinese AI lab DeepSeek has released preview versions of its V4 large language models, including V4 Pro with 1.6 trillion parameters, making it the largest open-weight model available. The models reportedly close the gap with leading frontier models on reasoning benchmarks while offering significantly lower pricing, though they trail state-of-the-art models by approximately 3-6 months in knowledge tests. The release comes amid U.S. accusations that China is stealing American AI intellectual property through proxy accounts.

Anthropic Exposes Massive Chinese AI Model Distillation Campaign Targeting Claude

Anthropic has accused three Chinese AI companies (DeepSeek, Moonshot AI, and MiniMax) of creating over 24,000 fake accounts to conduct distillation attacks on Claude, generating 16 million exchanges to copy its capabilities in reasoning, coding, and tool use. The accusations emerge amid debates over US AI chip export controls to China, with Anthropic arguing that such attacks require advanced chips and justify stricter export restrictions. The incident raises concerns about AI model theft, national security risks from models stripped of safety guardrails, and the effectiveness of current export control policies.

DeepSeek Introduces Sparse Attention Model Cutting Inference Costs by Half

DeepSeek released an experimental model V3.2-exp featuring "Sparse Attention" technology that uses a lightning indexer and fine-grained token selection to dramatically reduce inference costs for long-context operations. Preliminary testing shows API costs can be cut by approximately 50% in long-context scenarios, addressing the critical challenge of server costs in operating pre-trained AI models. The open-weight model is freely available on Hugging Face for independent verification and testing.

OpenAI Implements Strict Security Measures Following DeepSeek Model Copying Allegations

OpenAI has significantly enhanced its security operations to prevent corporate espionage, implementing measures like information tenting, biometric access controls, and offline systems for proprietary technology. The security overhaul was accelerated after Chinese startup DeepSeek allegedly copied OpenAI's models using distillation techniques in January.

Chinese AI Lab DeepSeek Allegedly Used Google's Gemini Data for Model Training

Chinese AI lab DeepSeek is suspected of training its latest R1-0528 reasoning model using outputs from Google's Gemini AI, based on linguistic similarities and behavioral patterns observed by researchers. This follows previous accusations that DeepSeek trained on data from rival AI models including ChatGPT, with OpenAI claiming evidence of data distillation practices. AI companies are now implementing stronger security measures to prevent such unauthorized data extraction and model distillation.

DeepSeek Releases Efficient R1 Distilled Model That Runs on Single GPU

DeepSeek released a smaller, distilled version of its R1 reasoning AI model called DeepSeek-R1-0528-Qwen3-8B that can run on a single GPU while maintaining competitive performance on math benchmarks. The model outperforms Google's Gemini 2.5 Flash on certain tests and nearly matches Microsoft's Phi 4, requiring significantly less computational resources than the full R1 model. It's available under an MIT license for both academic and commercial use.

DeepSeek's R1-0528 AI Model Shows Enhanced Capabilities but Increased Government Censorship

Chinese AI startup DeepSeek released an updated version of its R1 reasoning model (R1-0528) that nearly matches OpenAI's o3 performance on coding, math, and knowledge benchmarks. However, testing reveals this new version is significantly more censored than previous DeepSeek models, particularly regarding topics the Chinese government considers controversial such as Xinjiang camps and Tiananmen Square. The increased censorship aligns with China's 2023 law requiring AI models to avoid content that "damages the unity of the country and social harmony."

DeepSeek Releases Updated R1 Reasoning Model with MIT License on Hugging Face

Chinese AI startup DeepSeek has released an updated version of its R1 reasoning AI model on Hugging Face under a permissive MIT license, allowing commercial use. The updated model contains 685 billion parameters, making it a substantial upgrade that requires significant computational resources to run.

DeepSeek Emerges as Chinese AI Competitor with Advanced Models Despite Export Restrictions

DeepSeek, a Chinese AI lab backed by High-Flyer Capital Management, has gained international attention after its chatbot app topped app store charts. The company has developed cost-efficient AI models that perform well against Western competitors, raising questions about the US lead in AI development while facing restrictions due to Chinese government censorship requirements.