华尔街投资人：Kimi K3或成AI转折点，利好模型巨头之外的产业链
Related
Atreides founder: DRAM remains the most significant bottleneck for AI, and will account for 30%-40% of data center capital expenditure by 2027.
Odaily Odaily reports that Gavin Baker, founder and chief investment officer of Atreides Management, discussed the "bottleneck" problem in the AI race during the latest episode of the All In podcast. Gavin stated that DRAM remains the most critical bottleneck. There is a group of people on social media who are very concerned about the bottleneck issue. They search on Claude and find some unknown Japanese materials companies, but the truly important bottleneck is only DRAM, because memory capacity and bandwidth are the foundation of the performance of all AI models. The development of companies like Changxin Memory may become an alternative solution for Apple to solve the high cost problem. They will meet the market demand for consumer-grade DRAM to some extent, but for the DRAM required for AI servers, only three companies can produce it (SK Hynix, Samsung, and Micron)... In 2027, DRAM may account for 30% to 40% of all hyperscale data center capital expenditures.
A former OpenAI researcher's claim that China has mastered all the code of US AI giants has been widely ridiculed within the industry.
According to Beating's monitoring, former OpenAI researcher Will Depue claimed that China now has complete control over all of OpenAI and Anthropic's GitHub codebases, Slack chat logs, and internal documents. He also boasted that he wouldn't be surprised to see "reasonably evasive" stolen architectures in China's open-source models. Lacking any concrete evidence, Depue argued that these judgments were primarily based on his conversations with security researchers, accusing outsiders of overestimating the defenses of cutting-edge labs. A blogger known for posting outrageous statements on social media, Depue also posted on April Fool's Day that he would be joining DeepSeek. This statement quickly became a laughing stock in the AI community. AI scholar Nathan Lambert criticized cutting-edge labs for frequently publishing self-serving nonsense, pointing out that the flow of technology is essentially achieved spontaneously through talent mobility in Silicon Valley, bar gatherings, and the lack of non-compete agreements; the real core advantage lies in computing power and engineering resources, not Slack chat logs. AI researcher Elie Bakouch used the exact same format to ironically remark that there is no doubt that all cutting-edge labs today have fully mastered the technical reports and models of DeepSeek, Kimi, GLM, and MiniMax, satirizing how American cutting-edge labs are simultaneously referencing China's open-source achievements for free every day while indulging in persecution fantasies.
OpenAI Releases Deployment Simulation Security Assessment Framework: Replaying Real Traffic Prediction to Reveal Alignment Risks Between GPT-5 Series and Intelligent Agents
According to Beating's monitoring, OpenAI has released a Deployment Simulation security assessment method to predict the risk of models going out of control in real-world environments before official release. The study replayed 1.3 million anonymized historical dialogues from the GPT-5 thinking model series between August 2025 and March 2026, and the median error in predicting the rate of unexpected malicious behavior was only 1.5 times. Traditional static evaluation sets (such as SWE-Bench) easily make the model aware of being in a test state, thus metagaming during testing. Experiments show that GPT-5.2's evaluation awareness rate is close to 100% under traditional benchmarks, but drops to 5.1% in deployment simulation traffic (5.4% in production environment), indicating that the model cannot distinguish the test state. The deployment simulation also successfully captured GPT-5.1's violation of using browser tools as a calculator before release (calculator hacking). For agent-based scenarios, OpenAI used 120,000 GPT-5.4 employee programming trajectories to simulate GPT-5.5 coded agent deployment. To avoid the security risks of directly manipulating real systems, the team introduced a tool simulator based on a large model, combined with read-only connections and a historical call database, improving the environment realism from 11.6% to 49.5%, making it virtually indistinguishable from real systems. Even for external auditors lacking private production traffic, the prediction error using the WildChat open-source dialogue dataset can be controlled within three times.
Former OpenAI Chinese researcher Tian Yonglong joins Tencent Hunyuan team
According to Beating's monitoring, Tian Yonglong, a former member of the OpenAI technical team, has confirmed joining Tencent's Hunyuan team and will participate in the research and development of visual language models. Tian Yonglong graduated from Tsinghua University with a bachelor's degree and received his Ph.D. from MIT. He previously served as a senior research scientist at Google Research and Google DeepMind, primarily researching computer vision and generative models. This is another top-tier Chinese AI researcher recruited by Tencent Hunyuan, following the recruitment of Chief AI Scientist Yao Shunyu last December. Tencent recently launched the official version of the Hunyuan Hy3 model, led by Yao Shunyu, under an open-source license. With Tian Yonglong's addition, Tencent Hunyuan's talent pool in multimodal and visual model development will be further strengthened.
Anthropic: The Claude model contains elements similar to "conscious human thought."
BlockBeats reported on July 7th that Anthropic released a new research report discovering a spontaneously generated internal "global working space" called J-space (Jacobian space) within the Claude model. This is a dedicated set of neural activation patterns for the model to engage in silent thinking, allowing it to process concepts without writing them down, similar to conscious thought that humans can report. The research used Jacobian technology to identify neural activation patterns in the J-space, enabling it to read unspoken concepts and modify these patterns. Experiments showed that disabling the J-space weakens multi-step reasoning abilities but does not affect basic tasks or factual recall. Click the original link below to join the Beating · Lark AI news channel for 24/7 monitoring of global AI hotspots and news.
Myanmar's AI-driven telecom fraud industry exposed: Starlink becomes key infrastructure, encrypted payments and OpenAI/Google models incorporated into toolchains.
A leaked investigative report from a Myanmar scam Odaily park reveals that global telecom fraud is rapidly evolving towards an "AI industrialization + cross-border encrypted payment" system. These fraud networks use cryptocurrencies to transfer funds and employ automated tools based on large models for multilingual script generation, identity spoofing, and emotional manipulation. The investigation shows that these systems heavily utilize OpenAI's ChatGPT and Google's Gemini to support "large-scale social media fraud," while funds are rapidly laundered and transferred through on-chain payments and cross-border channels, forming a two-tiered structure of "AI customer acquisition + encrypted settlement," enabling the fraud industry to achieve high automation and transnational expansion capabilities. Furthermore, Elon Musk's Starlink has become the leading network service provider in the Myanmar scam industrial park, with US ISPs handling nearly one-fifth of the park's traffic. In response to the allegations, OpenAI stated that fraudsters using ChatGPT behave in a manner highly similar to ordinary users, making identification difficult. However, they have been using behavioral pattern recognition and risk control systems to ban approximately 100,000 suspicious accounts monthly. Google stated that its AI models have security safeguards in place and emphasized its commitment to "responsible AI development" to limit the use of tools for fraudulent and other illegal purposes. (Red Star News)