A lot of content engineering challenges have pretty standard solutions, which has lead to some engineers saying agent engineering is solved/boring True to some extent, but there are a bunch of (fun) new ones emerging as agents evolve! https://t.co/N82RZTcFSD
#English text is being prepared.#Agents#Reasoning
#English text is being prepared.
Apple announced that the first batch of Apple Intelligence features will open for testing in China on November 4 with the iOS developer beta, and will officially launch in December with iOS 27.3. It is also partnering with Alibaba to integrate Qwen into Siri, Writing Tools, and Visual Intelligence. DeepSeek completed its second round of financing this month, raising approximately 105.5 billion yuan, with a post-investment valuation of 615.5 billion yuan; Tencent and CATL remain the two largest investors. Anthropic has submitted filings to list on Nasdaq under the ticker ANT, with market valuation expectations of 1.5 trillion to 2 trillion US dollars.
#Industry news#Agents#OpenAI
Apple announced that the first batch of Apple Intelligence features will open for testing in China on November 4 with the iOS developer Beta, and will officially launch in December with iOS 27.3. It is also partnering with Alibaba to integrate Qwen into Siri, Writing Tools, and Visual Intelligence. DeepSeek completed its second round of funding of approximately RMB 105.5 billion this month, with a post-investment valuation of RMB 615.5 billion; Tencent and CATL remain the two largest investors. Anthropic has filed with the SEC and plans to list on Nasdaq under the ticker "ANT".
#Industry news#Agents#DeepSeek
According to The Wall Street Journal, Anthropic has publicly acknowledged that Claude is already deeply involved in developing its own successor, which the industry calls recursive self-improvement. Anthropic's June report "When AI Builds Itself" showed that, as of May this year, more than 80% of the code merged into its production codebase was written by Claude, whereas when Claude Code was first launched in February 2025 that share was in the single digits; an internal survey in August this year showed that 26% of R&D work was at an AI-led stage. Co-founder Jack Clark estimated in May that there was a probability of over 60% that AI would develop a successor model without human help by the end of 2028, and on October 12 he moved that prediction forward to 2027, stressing that this was only his personal judgment. The article also mentions that about 1,300 AI practitioners signed a joint letter calling for a coordinated slowdown in R&D, and that critics believe the existing data reflect faster coding rather than genuine autonomous evolution.
#English text is being prepared.#Safety / alignment#Anthropic
OpenAI announced that starting in December this year, the quotas for its three paid plans—ChatGPT Plus, Pro, and Business—will be calculated as a combined total, with ChatGPT and Codex sharing the same quota pool and billed uniformly by token; consumption depends on the model selected and task complexity. Instant chat does not consume the shared quota, and users can check their remaining quota and reset time on the "Settings—Usage" page in ChatGPT. This adjustment is related to OpenAI integrating ChatGPT Work's agent capabilities into ChatGPT, and the company expects most users will not run out of quota more often than they do now.
#Product updates#OpenAI#Industry news
The People's Hospital of Ningbo University introduced Alibaba DAMO Academy's medical AI, discovering 8 cases of pancreatic cancer from CT images of more than 40,000 people, 6 of which were early-stage, with model sensitivity of 92.9% and specificity of 99.9%. DAMO Academy's PANDA model trains by registering contrast-enhanced CT images onto non-contrast CT, enabling opportunistic screening with ordinary outpatient images; its multi-cancer early screening technology has covered 9 countries and regions and served over 20 million people. A team from Harvard University's Massachusetts General Hospital published the FaceAge study in The Lancet Digital Health, using facial photos to predict survival risk in cancer patients; among 3,820 patients tested, more than half of the high-risk group died within 90 days after the photo was taken, but the researchers cautioned that the tool carries risks of skin color and ethnic bias.
#Papers/Research#Multimodal#Image generation
Blackstone told investors that Anthropic may not immediately obtain an investment-grade rating after going public, and that major rating agencies are more likely to assign a high-end junk rating in the BB range. If the rating falls into junk territory, Anthropic's prospects for financing its more than $800 billion computing power plan will become more complicated, because funding pools such as insurance companies and pension funds will narrow significantly and interest rates will be higher. In the $63 billion financing codenamed "North Star," more than $42 billion of senior loans are guaranteed by Broadcom, while $18 billion of subordinated debt is backed by Anthropic's lease payments; KBRA assigned a BB+ rating and Morningstar DBRS assigned a BB rating.
#Industry news#OpenAI#Anthropic
A team from Mass General Brigham in the United States published a study in The Lancet Digital Health, developing the foundation model FAHR-Face and fine-tuning FAHR-FaceAge, which estimates age from faces, and FAHR-FaceSurvival, which predicts survival risk in cancer patients. Among patients classified as the highest-risk 10%, 52.5% did not survive beyond 90 days, compared with 2.1% in the lowest-risk 10%; after adjusting for age, sex, and cancer stage, patients in the highest quartile still had 3.23 times the risk of death compared with the lowest quartile. The team has launched the public FaceAge Research Study project to collect more diverse data.
#Papers/Research#Multimodal#English text is being prepared.
A CCTV News investigation found that before Double 11, a large number of clothing seller show images on e-commerce platforms were AI-generated, with some model images showing distortions such as three legs and hand poses that violate human anatomy, yet shop customer service still claimed they were real photos. A clothing e-commerce manager said that generating a full set of images with AI takes only a few minutes and costs about 30 to 40 yuan, while sending clothes out for a photo shoot costs 600 to 800 yuan per set and takes at least 3 to 4 days. Xie Yongjiang of Beijing University of Posts and Telecommunications said that if merchants falsely claim real photos and the difference between AI images and the actual product is enough to affect purchasing decisions, it may constitute false advertising or even consumer fraud.
#English text is being prepared.#Image generation#Industry news
Digital twin engineering technician Jiang Lei is building a digital twin platform for industrial inspection robots at Qiteng Robotics, using 3D reconstruction to recreate real factories in virtual space so that robots can complete route planning and task point verification before arriving on site. Jiang Lei said that in related projects, the traditional model requires two to three weeks of on-site debugging, while the introduction of digital twin simulation can shorten this to two to three days. Digital twin engineering technician is one of 11 new occupations released by the Ministry of Human Resources and Social Security and other departments in September this year.
#Industry news#Embodied AI
Tibo's post consists of only one sentence, "Banked 1/2," with no further context provided.
#English text is being prepared.
Shanghai Yidian, together with Shanghai Data Group and others, launched the "Joint Action to Promote the Innovative and Collaborative Development of Shanghai's Data Elements," releasing five major actions at the main forum of the 2026 Global Data Merchants Conference, covering public data demand matching, scenario innovation in key fields, Data for Science, state-owned enterprise data development and utilization, and the smart contract open-source community. Its "One Question, Use Data" demand-collection Agent has been open for trial use during the conference, and the "Zhiai Saisi" platform released its first batch of 11 high-quality benchmark datasets, covering fields such as materials, biomedicine, embodied intelligence, quantum, and nuclear power. The Zhiai Saisi community element library has already put online more than 3,000 AI for Science element resources, including 61 datasets.
#Industry news#English text is being prepared.#Agents
Zhang Changping, fund manager at Western Lead Fund, believes that the recent volatility in the technology sector is mainly driven by capital trading and external sentiment disturbances, and has not changed the long-term growth logic of domestic substitution. He recommends focusing on the allocation value of sectors such as computing power, storage, wafer manufacturing, and semiconductor equipment and materials. He pointed out that the domestic computing power ecosystem is moving from single-point product validation toward software-hardware collaborative adaptation, while the storage industry is seeing a resonance between an upcycle and domestic technology breakthroughs.
#English text is being prepared.#Reasoning#English text is being prepared.
In the third quarter of 2026, Unitree Technology's revenue was 665 million yuan, up 69.93% year-on-year, while net profit attributable to shareholders was 113 million yuan, down 17.37% year-on-year and down about 49.7% quarter-on-quarter from 224 million yuan in the second quarter. In the first three quarters, revenue was 1.817 billion yuan, up 55.71% year-on-year, but net profit after deducting non-recurring gains and losses was 349 million yuan, down 18.78% year-on-year, and net operating cash flow fell 54.69% year-on-year. Sanhua Intelligent Controls and Tuopu Group saw their Q3 net profit attributable to shareholders fall 9.64% and 4.78% year-on-year respectively, while Leadshine Technology's net profit attributable to shareholders in the first three quarters rose 86.80% year-on-year.
#Industry news#Embodied AI
An Economic Daily article points out that AI data centers have already occupied about 70% of the world's advanced memory capacity, leaving smartphones and computers to compete for the remaining 30%, which has caused prices of general-purpose memory and flash memory to rise sharply. In the first half of this year, contract prices for general-purpose memory nearly doubled, flash memory prices rose by more than 50%, and the proportion of memory chips in the total bill of materials for smartphones rose from 10% to 15% to more than 30%. The article argues that HBM squeezing wafer capacity, long capacity expansion cycles, and capacity being locked up by long-term agreements make this cycle different from previous cycles driven by consumer electronics.
#English text is being prepared.#English text is being prepared.
Murata Manufacturing announced it will invest approximately 1 trillion yen (about 42.2 billion RMB) from fiscal 2026 to fiscal 2030 to expand MLCC production, with the goal of doubling MLCC sales in fiscal 2030 compared with fiscal 2025. Of this, 350 billion yen will go to plant infrastructure and 650 billion yen to production line equipment and process upgrades. The Okayama base in Japan will receive 150 billion yen for new production lines, which are expected to gradually begin production by the end of 2028. Price increases from original manufacturers of high-capacity MLCCs used in AI servers generally range from 15% to 35%.
#Industry news#English text is being prepared.
The DeepSeek Harness team published a technical article titled "Super-agent Design Exploration: Toward a More Autonomous Agent System" on its official WeChat account, proposing a dual-layer architecture, event-driven design, asynchronous multi-channel interaction, separation of storage and compute, and an external Watchdog that takes over session management and context synchronization, shifting the multitasking ceiling from human attention to the system layer. The team says existing Agents operate as single Sessions that are independent of one another, and a person can typically follow only 3 to 5 active Sessions at the same time. Based on this approach and DeepSeek Harness, an internal Super-agent instance has been running continuously, participating in issue reproduction and fixes, code review responses, and iteration on Harness itself; the team emphasizes that this is an engineering-practice-oriented discussion of architectural ideas, not an introduction to existing product features.
#Product updates#Agents#DeepSeek
The official DeepSeek Harness team released "Super-agent Design Exploration," proposing a two-tier architecture in which a Router layer uniformly handles human interaction and task dispatch while multiple Workers execute concurrently in isolated sessions, and summarizing four characteristics: continuous, efficient, proactive, and persistent. Experiments show that Super-agent's task completion time p50 and p95 were 12.4 minutes and 15.7 minutes respectively, while the single-Session approach under neutral prompts had a p95 of 29.5 minutes; the team noted that the experiments were based on synthetic workloads and a simplified environment, mainly evaluating scheduling and collaboration capabilities. The team also provided a prototype implementation named dsh-superagent-prototype, and made clear that it is intended to validate the technical approach and is not an official DeepSeek product.
#Papers/Research#Agents#DeepSeek
According to a report by the UK's Financial Times, Blackstone Group has told investors that Anthropic is more likely to receive a speculative-grade rating in the BB range after its IPO, rather than the investment-grade rating previously expected. This reversal emerged in Anthropic's $63 billion chip financing deal; a speculative-grade rating would restrict investment by institutions such as insurance companies and pension funds. The prospectus shows that Anthropic plans to invest more than $800 billion in computing infrastructure by 2045, and currently holds more than $112 billion in cash and cash equivalents.
#Industry news#Anthropic