All news

336Kr · AI22:01Opinion

AI score55

ByteDance, Tencent, and Kuaishou all speed up at the same time—how far is interactive content from an explosion?

ByteDance, Tencent, and Kuaishou have intensively advanced their interactive content layouts over the past month: Douyin launched the Interactive Space Plaza and an interactive drama zone, adapted popular AI short dramas such as "Gui Xu" and "File Zero" into interactive versions, and on October 19 released a call for cooperation on interactive dramas and interactive film-games; Tencent's Konghu TDream mini-program began closed beta testing with data wipe, and Zaohua Gongfang updated professional features such as the infinite canvas workflow on September 17; Kuaishou launched the "Kuaishou Interactive Content World" mini-program, showcasing interactive dramas, simulation raising games, and lightweight mini-games in one place. The article notes that entry points and creation tools have gradually improved, but the cost of multi-branch storylines in interactive dramas is higher than that of ordinary short dramas, the revenue-sharing mechanism is not yet clear, and whether users are willing to stop at key nodes to make choices still awaits market validation.

#English text is being prepared.#Video#Industry news

336Kr · AI22:01Opinion

AI score56

AI comic drama quotes fall to 200 yuan per minute, profit divergence between producers and distributors

Quotes for AI comic drama production have fallen from about 5,000 yuan per minute to 200 yuan or even a few dozen yuan, a drop of about 96%. Industry practitioner Chen Yuanhang believes the industry has severe content oversupply and production teams have lost bargaining power. In September, Douyin added 51,800 AI dramas and comic dramas, but only 1.46% of new dramas took 64.6% of the playback increment, and about 82% of new dramas had less than 1 million plays that month. Duiba, meanwhile, shifted to the distribution side; in the first half of the year, AI short drama revenue was 223 million yuan, accounting for 51.2% of total revenue, and it disclosed that this business has been profitable on a monthly basis for three consecutive months.

#English text is being prepared.#Video#Industry news

IIEEE Spectrum22:01Opinion

AI score58

I Used AI to Fabricate a Scientific Paper, and IEEE Access's Detection Stopped It

IEEE Spectrum author Benjamin Skuse used the free versions of ChatGPT and Gemini to generate, step by step, a fictional paper in the field of optical solitons. After submitting it to IEEE Access, automated integrity checks identified multiple fabricated references, and it was not published. In the second round, he switched to the automated research system FARS, which generated a research proposal in 21 minutes, completed computational experiments the next day, and wrote a finished draft in 2 hours. He removed the system disclosure, added his byline, and submitted again; it was still blocked because of reference problems, and reviewers also found a meaningless annotation, lhewl, appearing in a figure. The article also provides background data: in 2025, about 57% of published academic articles carried detectable traces of LLM use, while only 1 in every 40 papers with statistical evidence of AI use formally disclosed it.

#English text is being prepared.#Papers/Research#Safety / alignment

Hheise online22:01Opinion

AI score52

From the Four Color Theorem to OpenAI: Why Computer Proofs Are Always Questioned

heise online looks back at 1976, when Haken and Appel used an IBM mainframe and 1,200 hours to complete a proof of the Four Color Theorem; at the time, the mathematics community fiercely questioned whether a computer proof counted as a proof. The article uses this to introduce how, after OpenAI released more than 370 Resultatfamilien, Terence Tao and others raised a new layer of doubts about LLM-led mathematical proofs that are difficult to understand.

#English text is being prepared.#Reasoning#OpenAI

UUnite.AI22:01Opinion

AI score51

Carta's Chief Product Officer: SaaS's Next Competition Is Above MCP

Carta Chief Product Officer Vrushali Paunikar believes that MCP solves the problem of connecting AI with enterprise software, but how to complete tasks after connecting is the key; the plugin and skill layers need to encapsulate instructions, workflows, and deterministic computation. The article cites a Google benchmark: letting an AI coding agent access technical documentation via MCP, then supplementing with task-specific instructions, achieved a 96.3% pass rate, with 63% fewer tokens consumed per correct answer than standard prompting. She believes that when all applications can expose data through the same open standard, access itself is no longer scarce, and knowing how to turn access into reliable results is more valuable.

#English text is being prepared.#English text is being prepared.#Agents

CComputerworld22:00Opinion

AI score50

Microsoft and other vendors push agentic AI PCs, advocating hybrid local-cloud sharing of AI workloads

At its October 7 PC launch event, Microsoft proposed the concept of "hybrid intelligence," letting frontier cloud models handle the hardest tasks while lighter tasks run locally on Copilot+ AI PCs to reduce token costs. Analyst Jack Gold expects that in the next three to four years, about 10% to 15% of AI workloads will run on PCs and 40% to 45% in the cloud. In a Microsoft demo, GitHub Copilot used GPT-6 Luna in the cloud to analyze code, and then a sub-agent running MAI Code 1.1 Flash locally reproduced the bug; the model was compressed to 3-bit precision and reduced in size by nearly 80%. The article also notes that the first batch of agentic AI PCs are expensive, with the Surface Laptop Ultra starting at about $2,600, and that enterprises still need to customize models themselves.

#English text is being prepared.#English text is being prepared.#Agents

TThe Guardian Technology22:00Opinion

AI score67

Experts warn that social media is flooded with AI disinformation ahead of the US midterm elections

The Guardian reports that experts warn that large amounts of AI-driven disinformation have appeared on social media ahead of the US midterm elections in November, and that many posts may not be intercepted by moderation. The article cites cases including Texas Senate candidate James Talarico being targeted by fabricated rumors of hospitalization and suicide, and Pennsylvania candidate Janelle Stelson being accused of hiring undocumented immigrants, and notes that the scope of content moderation rules at Meta, YouTube and X is limited, with some teams having been laid off.

#English text is being prepared.#Safety / alignment#Google

99to5Google22:00Opinion

AI score68

Anthropic's prospectus shows revenue of $16.3 billion in the first half, but AI has yet to deliver returns outside the tech industry

Daniel Bader of 9to5Google, citing Anthropic's prospectus, reports that the company's revenue in the first half of this year was $16.3 billion, compared with $1.2 billion in the same period of 2025; it has raised $141 billion and has more than $112 billion remaining, with a rumored valuation approaching $1.5 trillion. The article also cites a chart-based argument from Apollo chief economist Torsten Slok, stating that outside the tech industry, AI has not brought profit or productivity gains to U.S. companies, and that 80% of the top 1% of U.S. companies by AI spending are venture-backed firms.

#Industry news#OpenAI#Anthropic

HHuxiu21:24Opinion

AI score72

Anthropic Files S-1 in Nasdaq Push, Valuation May Be Cut from $1.5 Trillion to $1.11–1.48 Trillion

On the evening of October 26, Anthropic filed its S-1 prospectus with the SEC, applying to list on Nasdaq under the ticker ANT, with the number of shares offered, price and target valuation still left blank. The article applies a 5% discount for earnings sustainability, an 8% discount for cloud channel revenue sharing, and a 15% governance discount for the founder's lock on 50.1% of voting rights; after multiplying the three cuts, the valuation is revised down from $1.5–2 trillion to $1.11–1.48 trillion, and if profitability turns negative again, the pessimistic range is $1.05–1.4 trillion. The prospectus shows that 2026Q2 GAAP operating profit turned positive at $974 million, but 80% of revenue comes from API calls, and in the first half of 2026 about 48% of revenue was settled through the AWS and Google Cloud marketplaces.

Same story · 128 reports

#Industry news#OpenAI#English text is being prepared.

WWallstreetcn21:03Opinion

AI score66

Citadel Securities Report: The AI Narrative Is Shaking, and Risks Masked by Growth Are Emerging

Citadel Securities warned in its latest report that as the logic of the AI trade is being shaken, the market's margin for error has become extremely limited, and higher interest rates make the question of returns on AI infrastructure investment even more urgent. The report cites Meta as an example: its third-quarter revenue grew 26% year over year, but capital expenditure reached $34.9 billion, exceeding operating cash flow of $32.8 billion in the same period, and quarterly free cash flow turned negative at $2.1 billion. The report also mentions that Ramp data show token usage continued to grow while related spending was nearly flat from August to September, and that DeepSeek's release of DeepInfer indicates software improvements can boost the effective output of existing hardware, adding uncertainty to demand for incremental capacity.

Same story · 3 reports

#Industry news#English text is being prepared.#English text is being prepared.

HHuxiu20:00Opinion

AI score82

Anthropic Files S-1 to List on Nasdaq Under ANT…

Anthropic publicly filed its S-1 prospectus on October 26, 2026, planning to list on Nasdaq under the ticker "ANT," with revenue of $16.286 billion and a net loss of $80.748 billion in the first half of 2026. The author, Xi Chunying, points out that about $72.388 billion of the net loss came from fair value changes in financing instruments such as convertible notes, derivative liabilities and SAFEs, while the operating loss for the same period was $693 million and compute and infrastructure spending was $8.945 billion. The article argues that the core test of this IPO is whether AI commercialization can outrun compute investment, and warns that long-term compute procurement and leasing contracts could squeeze profit margins if demand slows.

Same story · 128 reports

#Industry news#OpenAI#Anthropic

SShao Meng19:15Opinion

AI score52

Matt Pocock on coding agents in the AI era: Say clearly what you care about, and make your codebase worthy of the agent

In a 44-minute interview with @nateherk, Matt Pocock said that in the AI era, the most valuable thing isn't chasing the latest model, but stating requirements clearly and getting tests and code quality to the point where the agent can be relied upon. He tried spec-driven development for several months; the more detailed the specs became, the more code quality slid, so he believes the code itself is the agent's working environment. He also mentioned that he commonly uses Claude Code with Opus 5.5, with effort set to medium, and believes most people set effort too high.

#English text is being prepared.#Agents#English text is being prepared.

YYicai19:03Opinion

AI score60

South Korean chip exports become AI's "canary," institutions warn of a hundred-billion-dollar export decline risk

South Korea's chip exports in the first 20 days of October surged 222.5% year-on-year to $27.5 billion, accounting for 44.7% of total exports in the same period, and are regarded as a "canary" indicator of AI capital expenditure. The Bank of Korea estimates that if major US tech companies slow their AI capital expenditure next year, South Korea's semiconductor exports could be $50 billion to $100 billion lower than previously forecast, and next year's GDP growth could be reduced by about 0.4 percentage points. Bloomberg Economics estimates that AI-related capital expenditure commitments by five core companies including Amazon from 2023 to 2027 will exceed $1.6 trillion, while Oxford Economics presents a scenario in which US GDP in 2028 would be 2% below the baseline under a tech downturn.

#English text is being prepared.#Industry news

336Kr · AI19:01Opinion

AI score79

Looking at Zhipu and MiniMax's Growth, Valuation, and Long-Term Narrative Through Anthropic's Prospectus

Anthropic made its prospectus public on October 26, planning to list on Nasdaq under the ticker "ANT." In the first half of 2026, revenue was $16.286 billion, up about 1,300% year over year, operating loss was $693 million, and cash flow from operating activities was positive at $3.929 billion. The article compares it with Zhipu and MiniMax: Zhipu's first-half revenue was 954 million yuan, up 399.7% year over year; MiniMax's revenue was $117 million, up 283.1% year over year; both companies' operating cash flow remained net outflows. Based on a rough calculation using market cap and ARR, MiniMax's P/ARR is about 9.8x and Zhipu's about 18.9x, while Anthropic's latest round at a $965 billion valuation corresponds to about 14.4x.

Same story · 128 reports

#Industry news#OpenAI#Anthropic

336Kr · AI19:01Opinion

AI score56

OpenAI claims to have proven the mathematical existence of Yang-Mills field theory, filling in half of Yang Zhenning's regret from back then

OpenAI published a post claiming that its internal model proved the mathematical existence of Yang-Mills field theory in four-dimensional space, and also claimed to have solved the Kakeya conjecture and the Stein restriction conjecture, overturned the Exponential Time Hypothesis (ETH), and provided a mathematical existence proof for QCD and a grand unification completion containing all gauge fields and fermions of the Standard Model. The article sorts out the two remaining problems of Yang-Mills field theory, pointing out that OpenAI has only solved existence; the mass gap required by the Clay Mathematics Institute's Millennium Prize Problems has not been solved, gravity has not been included, and the physics-related materials have not been made public, with the conclusions not yet verified.

Same story · 17 reports

#Papers/Research#OpenAI#English text is being prepared.

336Kr · AI19:01Opinion

AI score74

OpenAI releases GPT-6.1 Terra with a 50% price cut—can Chinese large models still hold onto their low prices?

On October 27 Beijing time, OpenAI released GPT-6.1 Terra, with standard API pricing at $1 per million input tokens and $5 per million output tokens, 50% lower than the previous generation GPT-5.6 Terra, while complex task execution, factual accuracy, and coding capabilities all improved. The article compares the pricing of Kimi K3.1 Pro at $2 per million input tokens and $10 per million output tokens, pointing out that domestic models rely on MoE architecture and engineering efficiency to keep the cost structure low, but total compute supply remains a hard boundary; Kimi K3.1 Pro saw rate limiting and quota anomalies within less than 48 hours of launch. The author argues that the room for domestic models to respond lies in B2B scenarios, data compliance, localized deployment, and government-enterprise channels.

Same story · 32 reports

#Model releases#OpenAI#Reasoning

336Kr · AI19:01Opinion

AI score82

AI Giants' Capital Race: Anthropic Sprinting Toward IPO, OpenAI Rushing for Funding, DeepSeek Raised to 100 Billion

36Kr reviews the capital race among AI giants in fall 2026: Anthropic has submitted an S-1 to the SEC and plans to list on Nasdaq under the ticker ANT, with its valuation expectation falling from $2 trillion to the $1.5 trillion range; OpenAI is concurrently advancing about $30 billion in funding, with a target valuation exceeding $1.4 trillion. The article also mentions that DeepSeek's new funding target has been raised to about 100 billion yuan, and cites the IMF's research on circular financing and Gartner's estimate of an approximately $5 trillion revenue gap, discussing the risks behind this funding wave.

#Industry news#OpenAI#Anthropic

336Kr · AI19:01Opinion

AI score62

Guokr Investigation of AI Companion App Paid Tiers: Memory, Voice, and Reply Length Split Into Membership Levels

Guokr author Lan Jianfeng, based on a month of registration, chat, and screenshot research, outlines how AI companion apps split memory, voice duration, reply length, and intimacy levels into different membership tiers: one overseas app's free users can only retain 10 memories, while another overseas app caps free replies at 300 characters, with members reaching 600, 1,000, and up to 2,000 characters by tier. The article also compares domestic and international differences: domestic apps' free versions retain complete basic features with membership offering deeper features, while overseas apps mostly restrict features through unlock mechanisms, and cites the Interim Measures for the Administration of AI Anthropomorphic Interactive Services, announced in April this year and effective from July 15, to explain the domestic regulatory framework.

#English text is being prepared.#Industry news#Agents

336Kr · AI19:01Opinion

AI score58

AI health consultations such as Doubao's gain popularity: why are big tech companies redoing internet healthcare?

36Kr, republishing an article by Jin Cuodao, reported that Doubao's daily health consultations exceeded 20 million, surpassing the total daily offline outpatient visits of hospitals nationwide, which is just over 10 million, and that OpenAI disclosed that 300 million people ask ChatGPT health questions each week. The article compares this with the previous generation of internet healthcare platforms, which fell into losses because they needed to sign contracts with large numbers of doctors and could only handle follow-up visits, and argues that general-purpose large models taking on pre-hospital health Q&A at low cost is the reason big tech companies such as ByteDance and Ant are re-entering the market.

#English text is being prepared.#Industry news#Agents

336Kr · AI19:01Opinion

AI score70

Dolphin Research breaks down Anthropic's prospectus: 85% inference gross margin, trillion-dollar compute bill requires revenue to rise another 4 to 5 times

Dolphin Research, based on an analysis of Anthropic's prospectus, states that its revenue in the first half of 2026 was $16.3 billion, of which usage-based billing revenue was $13.1 billion, accounting for 80%, and inference gross margin was 85%, but training costs of $6.6 billion consumed about 40% of revenue. The article calculates that signed compute contracts and leases total about $1.01 trillion; if the compute cost ratio remains at the 55% break-even line, annual revenue would need to be about $210 billion, while 2Q26 annualized revenue is $46.2 billion. The author also compares Zhipu and MiniMax, arguing that Anthropic gains pricing power through SOTA performance, but rigid take-or-pay contracts mean it can only cut prices when competition worsens.

Same story · 128 reports

#Industry news#English text is being prepared.#English text is being prepared.