The most useful new evidence today is about what the latest filings do—and do not—say about common-owner earnings and cash conversion. Schneider Electric’s official 16 October release reports Q3 revenue of €11.460bn, up 16.9% organically, with triple-digit data-center demand and a higher 2026 target. That release was already in the 16 October research baseline. Its official PDF has no publication clock, so its place inside the exact rolling 24-hour window remains unconfirmed. Schneider’s model and valuation record remain on a separate ranking hold, despite the strong demand update. [Schneider Q3 release](https://www.se.com/ww/en/assets/pdf/release-Q3-revenues-2026)

TSMC’s 15 October release reports NT$1,494.24bn revenue and NT$763.06bn net income for Q3, alongside NT$29.43 diluted EPS per ordinary share and US$4.60 per ADR unit. Those units should remain separate unless the ADR ratio and currency bridge are applied. The event is earlier than the current 24-hour window, but it supports the pending 15% foundry allocation’s demand case; it does not resolve the 42.46x normalized earnings multiple or Taiwan/capex risk. [TSMC Q3 results](https://pr.tsmc.com/english/news/3346)

Two accounting checks change how the filings should be read. Lenovo’s unaudited June quarter reports a US$477.592m group loss, a US$608.600m common-owner loss, and US$131.008m of positive NCI income. Its US$1.075bn adjusted owner profit excludes a US$1.690bn warrant fair-value loss and other non-cash items. That adjusted figure is not GAAP earnings. The source confirms the issuer and quarter, but the accepted prior payload is unavailable here, so the separate Lenovo ranking hold remains. [Lenovo Q1 announcement](https://www1.hkexnews.hk/listedco/listconews/sehk/2026/0813/2026081300141.pdf)

Hua Hong’s full unaudited interim report resolves a prior extraction gap. It reports US$59.568m owner profit against US$72.957m NCI loss, producing a US$13.389m group loss. H1 CFO of US$468.481m compares with US$1,281.437m of PPE and intangible purchases, or negative US$812.956m before other flows. It also reports 1,737,646,356 basic weighted shares plus 56,408 dilutive options and confirms the May rename to Hua Hong Grace. These facts do not reconcile the changed source payload to the accepted baseline, so the ranking hold remains. [Hua Hong interim report](https://www1.hkexnews.hk/listedco/listconews/sehk/2026/0826/2026082600917.pdf)

Jabil’s September release is preliminary and unaudited. It pairs FY2026 common-owner income of US$1,042m with 106.9m diluted weighted shares; the US$9.747 quotient rounds to reported GAAP diluted EPS of US$9.75. Non-GAAP core diluted EPS of US$13.09 stays separate. This supports the 15% pending manufacturing position, but audited cash conversion and thin margins still matter. [Jabil FY2026 release](https://www.sec.gov/Archives/edgar/data/898293/000162828026063890/jbl-20260930ex991.htm)

Across the scenario set, Jabil (14.32x) and Amkor (13.11x) screen below TSMC (42.46x), Alphabet (36.28x), and Vertiv (31.14x). These are normalized-earnings scenarios, not current trailing P/E ratios or expected returns. Every modeled terminal-price gap is incomplete because it omits some interim cash and assets. The 15 research priorities were all covered in the batched searches, as were 20 rotating companies and 18 grouped supply-chain layers. Eleven of eleven requests succeeded, but all 220 returned leads had date-only ambiguity and none established exact 24-hour membership. No absence-of-news conclusion follows. The 16 October baseline already held Schneider’s release and TSMC’s 15 October results. The portfolio remains all cash until seven targets become eligible on 19 October; no rebalance is proposed. [HOYA FY2026 Integrated Report](https://www.hoya.com/ir/2026/en/review/it.html)

A ceasefire can be real in one part of a child’s life and painfully absent in another. That is the picture UNICEF presented on October 16: more food and a return to learning for many children in Gaza, alongside reported deaths, widespread displacement and sanitation systems that remain broken.

In [remarks delivered at the United Nations in Geneva](https://www.unicef.org.uk/press-releases/childhood-defined-by-struggle-for-survival-in-the-gaza-strip/), UNICEF deputy spokesperson Ricardo Pires said five boys and two girls had reportedly been killed in the preceding two weeks. At least 33 children were seriously injured in airstrikes, drone strikes and gunfire, he said. Two more boys died of earlier injuries. UNICEF’s reported total since the ceasefire announced in October 2025 now exceeds 340 children killed. These are the agency’s reported figures; they do not identify responsibility for each incident.

The same briefing described gains that deserve to be counted. UNICEF said more than 560,000 children had returned to learning since the ceasefire, including at more than 200 centers it supports. It said it was supplying safe water to 1.2 million people and that improved access to supplies had eased food insecurity. A statement that acknowledges those gains is more useful than one that pretends recovery has made no progress.

But a learning center cannot repair a city’s drainage. UNICEF says all six wastewater treatment plants are still out of service. Its October 16 remarks described the Sheikh Radwan stormwater basin as half full, with winter rain approaching. The agency warned that contaminated water could overflow into homes downstream. Pires said the basin could be drained within two weeks, but that Israel had not allowed the necessary equipment into Gaza. That account is UNICEF’s; this edition has not independently established the equipment’s status or obtained a response from Israeli authorities.

The practical distinction matters. Supplying drinking water to a family addresses today’s need. Restoring treatment plants and drainage reduces the danger that tomorrow’s rain will bring through the door. UNICEF says spare parts, generators and engine oil are needed to keep existing services running and to restore others. A promise to permit aid is therefore only part of the story. The test is whether the specific equipment reaches the workers who can use it before the basin fills further.

The agency estimates that about 800,000 children remain displaced, many in tents or damaged buildings. It also estimates that 60,000 children have lost one or both parents. Such estimates describe a vast need; they cannot describe every child’s circumstances. Pires told the story of a 16-year-old caring for her younger sister after their mother died. It was a UNICEF account of a family its staff met, not an interview conducted by this paper.

There is a tendency to ask whether a ceasefire has succeeded as if the answer must be a single word. The evidence UNICEF presented calls for a harder set of questions. Are children safer this week? Can they study? Can families get food? Will sewage enter their shelter when it rains? The answers can move in different directions at once. Winter will expose the difference between an improvement that can be measured now and a repair that has yet to be made.

The World Health Organization and INTERPOL [announced a cooperation agreement on October 16](https://www.who.int/news/item/16-10-2026-interpol-and-who-sign-agreement-to-strengthen-global-health-and-security-protection). They plan to combine public-health and law-enforcement expertise in cases where a threat crosses both fields. Their list includes falsified medical products, trafficking for organ removal, unsafe or fraudulently labeled food and unusual biological events.

That is a wide brief. A falsified medicine may require laboratory evidence, supply-chain tracing and a criminal investigation. An unusual biological event may require health agencies to investigate before anyone knows whether a crime occurred. The two organizations say their agreement provides a framework for cooperation in such situations and at major international gatherings.

The announcement does not say that a particular trafficking network has been stopped or that a new response system is operating in every member country. Those would be outcomes to report later. For now, the dated fact is the agreement and the work it proposes. WHO also lists counterfeit protective equipment and illegally produced alcohol among the goods covered by the planned cooperation.

The public interest lies in how the organizations handle uncertainty. Health officials need to share urgent warnings when products or events may cause harm. Investigators need evidence they can test. Those tasks can support one another, but neither agency’s October 16 release shows how quickly information will move between them in a real case. That will be the agreement’s meaningful measure: whether a warning reaches the right people, with enough detail to help them act.

NASA’s account of Orion’s return from the Moon offers a rare, clear comparison between a test and its follow-up. In an [October 15 post-flight assessment](https://www.nasa.gov/missions/artemis/artemis-2/rigorous-analysis-confirms-nasas-artemis-ii-heat-shield-performance/), the agency reported nine spots of heat-shield char loss after Artemis II, compared with 103 after Artemis I. NASA says the shield protected the four astronauts and the spacecraft as expected during reentry.

The agency changed Orion’s reentry trajectory after examining the Artemis I shield. It shortened the maximum distance the craft could travel through the atmosphere before splashdown, limiting time in the temperature range associated with the earlier damage. NASA says the Artemis II review found cracking and char loss reduced by a factor of ten. That is the agency’s engineering conclusion from the flight and its inspections.

The returned hardware remains a source of evidence. NASA says engineers have extracted samples, performed internal scans and continued measuring how much protective material wore away. Those checks matter because the question is not simply whether the crew arrived safely. Engineers also need to know whether their models predicted what happened and whether the manufacturing process will produce consistent shields for later missions.

Artemis II showed that a changed flight path could reduce the damage NASA observed on the earlier mission. NASA also says it has made design changes for Artemis III. The October assessment supports confidence in this flight’s performance; the continuing tests will help determine how broadly its lessons apply.

A stained-glass window can make its design process disappear. The finished surface looks inevitable; the drawings that guided it usually stay in storage. The Metropolitan Museum of Art plans to reverse that view in [*Designing the Gilded Age*](https://www.metmuseum.org/pt/press-releases/designing-gilded-age), an exhibition scheduled to open October 22 and run through February 7, 2027.

The museum says the show will bring together nearly 120 works from its collection and selected loans. Working sketches and watercolor presentations will sit within a broader account of Louis Comfort Tiffany’s collaborative studio. The Met also plans to reunite a stained-glass window with original wisteria transoms discovered in an Arizona home, presenting the parts together for the first time in at least half a century, according to its announcement.

This is an appealing shift in emphasis. A drawing shows choices before they harden into an object: an arrangement tried, a color proposed, a commission imagined. The museum says some designs document works that were lost, changed or never built. Such sheets can tell visitors about possibilities that a display of finished lamps and windows alone would miss.

The announcement is a description of a forthcoming exhibition, so any judgment about the installation must wait until it opens. Its premise is already worth attention. Tiffany remains the name most visitors will recognize. Looking at the studio’s drawings offers a chance to ask whose work made that name visible in glass.

A wind-driven brush fire tore through neighborhoods on the border of San Clemente and San Juan Capistrano on Friday. Fire officials said it damaged or destroyed six homes and forced about 16,000 people to leave their homes before nightfall.

The Orange County Fire Authority said crews responded shortly after 11:30 a.m. to reports of a fire near Portico Del Norte. Hot temperatures, low humidity, dry brush and Santa Ana winds blowing from the interior toward the coast drove its rapid growth. By Friday evening it had grown to about 570 acres. CNN reported that the fire authority listed it as zero percent contained. Officials said two firefighters suffered significant injuries and were in stable condition at hospitals.

“In anticipation of the fire weather that we’re currently experiencing, we had upstaffed a bunch of resources and placed them in strategic areas,” OCFA Battalion Chief Mark Backhoff told CBS Los Angeles. Capt. Steve Concialdi said crews had knocked down most of the fire by early evening. He warned that the danger was not over: “When we have Santa Ana winds, they could blow embers a mile, even greater.”

Officials gave different numbers of homes under evacuation orders in different briefings. Figures ranged from about 4,600 to 6,000 homes across zones in both cities. Evacuation shelters opened in both cities. The Capistrano Unified School District dismissed ten schools early. Rep. Mike Levin, a Democrat whose district includes part of the area, said his own family had evacuated their San Juan Capistrano home.

“I’ve lived here since 1995, and this is the first time we’ve really ever had a fire in this area,” resident Robert Morris told KCBS. “It’s a wake-up call.”

Gov. Gavin Newsom said the state had secured federal FEMA assistance for the fight. San Diego Gas & Electric reported two outages near the fire affecting nearly 3,000 customers combined. The South Coast Air Quality Management District warned that smoke would make air unhealthy for sensitive groups in parts of south Orange County until Saturday morning.

The Norte Fire is one of several fires burning across Southern California. In Malibu, the Bryson Fire, which started Thursday evening, had burned about 150 acres and was 17 percent contained. More than 7,000 structures were in areas under evacuation orders or warnings, according to Los Angeles County Fire. In San Diego County, the Henderson Fire had burned about 2,080 acres at Marine Corps Base Camp Pendleton and was 20 percent contained, according to Cal Fire.

A heat advisory remains in effect until 8 p.m. Saturday, with temperatures forecast to reach up to 95 degrees. Forecasters expect the winds to weaken later in the weekend. Until they do, officials are asking residents to follow evacuation orders and stay off the Interstate 5 corridor near the fire.

For years, export controls in the AI race have mostly been an American tool. On Friday, China published a list of its own.

China’s Ministry of Commerce and Ministry of Science and Technology jointly issued Announcement No. 48. It proposes adding 40 technologies to the country’s catalog of technologies whose export is banned or restricted: 23 in artificial intelligence, 12 in solar manufacturing and five in biomedicine. The public can comment until Nov. 15. One part does not wait for the consultation: starting immediately, anyone exporting a listed technology must register the deal with the commerce ministry first. According to a translation of the notice published by Geopolitechs, that includes exports that have already started but are not yet complete.

What would be banned, and what restricted

The draft bans the export of core technology for large AI models trained on China’s “important” national data. It also bans AI agents designed for highly sensitive settings such as national defense and critical information infrastructure. AI tools that can write or alter malicious code on their own to attack critical systems face a total ban.

The draft would restrict, rather than ban, the export of frontier models above certain thresholds. According to The Decoder’s reading of the annex, the restrictions would apply to a model’s weights, source code, architecture and training methods. They would kick in if a model was trained with at least 10^26 floating-point operations of computing power, has 2 trillion or more parameters, or ranks in the global top 10. AP noted that Moonshot’s Kimi K3 and Alibaba’s Qwen 3.8-Max already exceed 2 trillion parameters, though the proposal names no specific models. Some design technologies for AI and memory chips are also on the list.

The draft leaves some activity uncontrolled. Technology that has already been made public through open-source releases or papers is exempt. So is using a model through an API, an app or a hosted software service. According to The Decoder, installing restricted technology on servers abroad would need a license. Releasing it as open source would qualify for automatic licensing.

Why it matters

“The Chinese government is implementing such necessary measures with the aims of reinforcing the security bottom line, preventing misuse and abuse, and ensuring that AI remains under human control,” the commerce ministry said in a statement.

Lian Jye Su, chief analyst at Omdia, told AP the move could be seen as a response to Western countries placing sovereignty requirements on AI and other key technologies. Dean W. Ball, quoted by The Decoder, said the move does not mean China is shutting down open-weight models. He said it does show the government “establishing the optionality to shut down future open-weight models at will.”

That optionality is the real news. Many companies and governments outside China have built products on Chinese open-weight models. They now know that future versions could come with conditions. AP noted that the draft arrived as Washington and Beijing reported progress in talks on AI safety, including a channel for notifying each other of AI incidents that threaten national security.

The proposal also landed during a bad week for chip stocks. CNN reported that chip shares had not recovered since Chinese memory maker CXMT said earlier this week that it had made progress on advanced memory chips.

In his [October 16 World Food Day message](https://india.un.org/en/324550-world-food-day-2026), UN Secretary-General António Guterres said about 645 million people faced hunger in 2025 and nearly 2.7 billion people could not afford a healthy diet. The theme for the day asks the world to innovate now to nourish tomorrow. Fine. But an invention has not fed anyone merely because it has been announced.

The affordability figure should sit at the center of every food-technology pitch. A tool that raises yields may help. A better forecast may help. Artificial intelligence may help, as the UN message suggests. Each claim should eventually meet the same ordinary question: did nutritious food become easier for families to obtain? If the answer is unknown, the product is a proposal, not a victory.

This is not an argument against research. It is an argument for a demanding measure of its value. Governments and funders can count products launched, farms enrolled and money committed. They should also ask what a household can buy after rent, transport and other essentials. That measure may reveal barriers that a new technology alone cannot remove.

World Food Day is useful when it keeps that distance visible. Hunger is experienced at a table, not in a strategy document. Any program that claims to narrow the gap should be able to show where, when and for whom the next healthy meal became affordable.

On Friday in Bangkok, the managing director of the International Monetary Fund offered a metaphor that was always going to be quoted. Artificial intelligence, Kristalina Georgieva said, is a Formula 1 race car that might leave the track and run people over. “We are racing, and the question is how do we put barriers?” she said, according to the Associated Press. “So if a car was to fly out it doesn’t kill people.”

On the same panel, Jack Clark, Anthropic’s co-founder and head of policy, went further. “I think it’s worse than you describe. I think it’s Formula 1 at night without maybe brakes in the cars,” he said. “And it’s remarkable to me that the toys I buy for my children are regulated more than the literal superintelligence I work on at my day job.”

Those lines are vivid and easy to repeat, and they are about the wrong risk. A crash is dramatic, and it is the kind of danger institutions already know how to talk about. The more revealing sentence came later in the session, when Georgieva described what an IMF study found about jobs.

She said about 10 percent of jobs might be transformed or newly created, and that repetitive work would disappear. A third category would be service jobs created by people who benefit from the investment boom. “The net impact is positive,” she said. Then she added: “services don’t pay a lot. So you get young people who normally would have gone into professional routine jobs now serving coffee in Starbucks. And that is not what I think young people in the world aspire for.”

Call it the barista clause. The world’s most important economic institution has said in public that the AI transition may look positive in the aggregate statistics while moving a generation of graduates from the entry-level office jobs that start careers into counter work that does not lead anywhere. That is not a car crash. It is a slow loss of status that will never be a single news event, and that is why it is dangerous. Nobody holds a press conference about a career that never started.

Tiff Macklem, governor of the Bank of Canada, said the same panel’s quiet part out loud. He compared AI to trade, which raised global wealth while costing millions of manufacturing workers their livelihoods. “You have to find a way to help the losers,” he said. “Without a system to help them cope with the loss you get a backlash.”

We already ran that experiment. The help for trade’s losers was small, late and full of paperwork, and the backlash has shaped politics ever since. If the plan for AI is to say “net positive” again and promise retraining later, nobody should be surprised by the anger that follows.

Where I part ways with the panel is the metaphor. A race car on a track has spectators safely behind a barrier. AI has no grandstand. Most people are on the track, and the question is not only whether the car crashes. It is who owns the track and who collects the admission. This week we learned the chipmaker at the bottom of the supply chain keeps more than half of every dollar it books, while the companies that borrowed to build data centers are paying junk-bond yields. Profits are concentrating at a few points in the system. A serious transition policy would ask those points to pay for the barriers, and not only for safety rails against a crash. It would also fund the first rung of the career ladder that this technology is removing.

Clark said the decisions “cannot only be companies making these decisions.” He is right. It cannot only be central bankers on panels either. The test of whether the AI economy is working is not global growth. It is whether a 24-year-old with a degree can find work that leads somewhere. By the IMF’s account, that risk is now explicit, but its scale remains unknown.