President Trump cast American elections as under siege last night, describing a system riddled with vulnerabilities that hostile foreign actors and unauthorized immigrants are exploiting.
The dark, foreboding 25-minute address from the East Room, which repeated fraud theories that have been debunked, served two main purposes, Axios’ Alex Isenstadt and Marc Caputo report:
Build support for his SAVE America Act, which would require proof of citizenship for voter registration and is stalled in the Senate.
Return to a topic that fixates him perhaps more than any other: The 2020 election, which he lost to Joe Biden.Photo: Saul Loeb/AFP via Getty Images
Citing newly released “raw” intelligence, Trump claimed that China carried out “the largest compromise of election data in history” during the 2020 election — obtaining 220 million U.S. voter files and creating “ballots for Biden.”
One big catch: Voter rolls listing names and addresses are readily available in nearly every state. Some even post them online to promote transparency.
He also accused the intelligence community — the “Deep State” — of withholding documents describing China’s activities from him when he was in his first term as president.
By blaming the intel community of 2020, he excused his own administration — presumably including current CIA director John Ratcliffe, who was DNI director at the time — from catching what he now calls a huge threat.Speech headlines in today’s Wall Street Journal, Washington Post.
Friction point: Some in Trump’s political operation believe that talking about voter fraud will motivate his voters to turn out in November. But outside of the White House, party leaders and pollsters strongly believe that swing voters don’t want to hear about it.
“It’s a stupid, stupid move,” said one Republican pollster who works on several campaigns and has tested the effectiveness of the “stolen election” narrative.
“Even swing voters who think something wasn’t good about the election, when they listen to Trump, just have an eyeroll,” the pollster said.
China has become one of Africa’slargest development financiers. Since 2000, Chinese and other state-backed institutions have committed more than US$180 billion in loans to African countries. The money has been used to finance roads, railways, power stations, ports, water infrastructure and industrial projects. Agriculture has also become part of this expanding partnership. Food systems specialist Adrino Mazenda analysed Chinese loans to African countries between 2000 and 2024 to find out about China’s spend in agriculture.
What and where is China investing in when it comes to agriculture?
China’s agricultural funding mainly goes to farm development, fisheries, irrigation, mechanisation, agricultural technology and rural infrastructure.They invest less in agro-processing and storage. Southern African countries (Angola, Zambia, Zimbabwe and Mozambique) receive the most in loans.They’re followed by east Africa (Ethiopia, Kenya and Tanzania) and west Africa (Nigeria and Ghana). Chinese institutions also invest in agriculture in Egypt.
My study compared agricultural lending from Chinese institutions with lending to other sectors. It also looked at who received the loans, where the money went, and the kinds of farming projects that were funded.
I found that Chinese lending to agriculture has expanded across Africa. Between 2000 and 2024, Chinese lenders financed 41 agricultural loans worth approximately US$2.26 billion across Africa.
Most agricultural finance supported farm schemes, production estates, fisheries, mechanisation and irrigation infrastructure.
Farm schemes accounted for almost 36% of agricultural lending.
Fisheries received 29% of the loans.
Investment in storage and cold-chain infrastructure made up only 3% of China’s loans.
Loans for agro-processing facilities were less than 2%.
Larger loans were more frequently channelled through government-affiliated agencies and other non-sovereign entities rather than national governments.
Our mission is to share knowledge and inform decisions.
Overall, agricultural funding is a relatively small part of China’s much larger development finance portfolio. Transport, energy and other infrastructure sectors consistently received much larger commitments.
What are the shortcomings?
My study found that many investments improved farming infrastructure, such as irrigation systems and roads.However, there was not enough investment in food processing, storage, transport and markets. These are needed to build strong agricultural industries.
I also found that decisions about agricultural loans were driven more by whether projects were practical and who was applying for the money. Whether countries had the capacity to carry out the work was also key for investors. Funding wasn’t shaped by any plan to transform agriculture across Africa.
This fits a broader pattern by Chinese lenders to fund projects they believe can be delivered, rather than following a wider strategy for developing a particular sector.
This is a problem because many African countries need major investment to modernise agriculture. They often do not have enough money to pay for irrigation, farm machinery, storage facilities, transport infrastructure and food processing plants on their own. Loans from international development partners are therefore important. But whether these loans lead to lasting development depends not only on how much money is available, but also on the kinds of projects they fund.
Made by Adrino Mazenda on ChatGPT
For agriculture to drive economic growth and improve food security in Africa, the sector needs to change. This means more than producing more crops. It also requires investment in markets, research, agricultural extension services and institutions that connect farmers to local and international buyers. For example, smallholder farmers need to be connected to global supply chains.
But my research found that China’s investments fall short in this area. Although China’s funding strengthened agricultural production, it provided comparatively little support for storage, processing and market systems. These are widely recognised as essential for agricultural transformation.
What are the solutions?
The long-term contribution of Chinese finance to African countries will depend on two things.
Firstly, the amount of investment. Secondly, whether future lending supports the wider systems that connect production, processing and markets. Investing only in farming infrastructure is unlikely to produce the bigger changes needed to create a more productive and competitive agricultural sector.
African governments have an opportunity to negotiate financing that goes beyond financial boosts to individual projects. Greater investment in the following would strengthen agricultural value chains and increase the long-term benefits of external finance:
storage facilities
agro-processing
cold-chain systems
transport networks
research
extension services
market development.
Governments also need to make sure that their agriculture, finance and planning departments work together. This would help ensure that loans from international lenders support each country’s long-term plans for agriculture instead of funding disconnected projects.
Governments should also be more transparent about how they borrow money and how projects are carried out. This would make it easier for the public to hold them accountable. It would also help ensure that the money leads to lasting improvements in agriculture.
Development partners can choose to support agricultural financing models that connect production with processing, post-harvest infrastructure and market access. In this way, their investment in a country’s agriculture can generate wider economic benefits.
Climate change, population growth and food insecurity are placing increasing pressure on African food systems. The question is whether development finance is structured to create value long after individual projects have been completed. This will require building productive, competitive and resilient agricultural systems.
The American Dream now runs through the skilled trades, according to Jamie Dimon—and his bank is investing $24 million to ramp up training in Philadelphia.Jeenah Moon/Bloomberg via Getty Images
“We need 300,000 electricians, welders, etc. to build ships in the next five or 10 years,” Dimon told CNBC on Tuesday, speaking from the Philadelphia Navy Yard.
Unlike many traditional white-collar career paths,skilled trades apprenticeships often pay workers while they train—putting them on a path to six-figure salaries without taking on college debt.
“It fits what we call the American dream: getting kids skills or all workers’ skills that they have jobs that could pay 80-, 90-, $100,000 a year after you know a year or two of training. This lifts up America. It helps build the defense industry,” Dimon added.
The Philadelphia Navy Yard alone currently employs about 16,000 workers, but Dimon said that workforce could double over the next five years as domestic shipbuilding ramps up.
To help fuel that growth, JPMorgan announceda $24 million investment through loans and philanthropic grants to support a new submarine manufacturing and assembly facility expected to create 450 permanent jobs. The funding will also expand workforce training and apprenticeship programs for “thousands” of prospective welders, electricals, pipefitters, and other skilled trades workers needed at the Navy Yard.
The U.S. can’t catch South Korea without more skilled workers
Joining Dimon was Pennsylvania Sen. David McCormick—husband of Meta president Dina Powell McCormick—who argued that while artificial intelligence has fueled fears about the future of white-collar work, demand for skilled trades has never been stronger.
“If you’re an experienced welder or electrician, we can’t get enough of you,” he said. “We have to have [a] workforce that meets this incredible demand.”
But Philadelphia embodies the uphill battle the industry faces. The yard was purchased by Hanwha, a South Korean ship manufacturer, in 2024 for $100 million and currently only delivers one to one-and-a-half ships per year. In South Korea, ships are delivered “basically one a week,” according to the CEO of Hanwha Philly Shipyard, David Kim. The apprenticeship can also accommodate only about 20 trainees at a time, compared with roughly 400 simultaneously at Hanwha’s facilities in South Korea.
Workforce struggles extend well beyond Pennsylvania.
In the Hampton Roads region of Virginia, there is an estimated shortage of 10,000 shipyard workers, which is expected to grow to 40,000 by 2030, reported WHRO.
At Huntington Ingalls Industries—an American shipbuilding company that operates major yards in Newport News, Virginia and Pascagoula, Mississippi—over $110 million is invested annually in workforce development. And according to CEO Chris Kastner, it’s becoming a growing priority.
“We hired over 1,600 shipbuilders in the first quarter,” Kastner said during HII’s 2026 Q1 earnings call in May. “We also graduated nearly 200 apprentices from our apprentice schools this year, and our apprentice schools are now at full enrollment. I’m confident that as our workforce continues to stabilize, our workforce will become more proficient.”
Meta, Lowe’s and BlackRock are investing hundreds of millions into the skilled trades—and presenting Gen Z with opportunity
Shipbuilding is just one sector eager for skilled workers. But as demand surges in construction, including energy projects, AI data centers and semiconductor facilities, many companies are now pouring hundreds of millions of dollars into training the next generation of electricians, plumbers, machinists, and more.
Meta also entered the effort in June, with the launch of a $115 million training program called America’s Workforce Academy. The effort seeks to specifically train data center technicians, and after completing the 5-week program, a job is guaranteed on one of Meta’s sites.
“America needs hundreds of thousands of skilled tradespeople—electricians, mechanics, fiber technicians, and more—and this program creates clear, accessible pathways into those careers,” Rachel Peterson, Meta’s vice president of data centers, said at the time.
The stakes go beyond one individual company or sector. By 2030, an estimated 2.1 million skilled trades jobs could go unfilled, according to an analysis of U.S. Department of Education data by JLL. And the economic losses could reach $1 trillion annually.
The Fortune 500 Innovation Forum will convene Fortune 500 executives, U.S. policy officials, top founders, and thought leaders to help define what’s next for the American economy, Nov. 16-17 in Detroit. Apply here.
The real message to be drawn from Trump’s address to the nation tonight is that he will call into question the votes of every state and city that chooses a Democratic senator or representative in the 2026 midterm elections. He’ll push Republican governors and mayors not to certify the results. He’ll demand recounts and audits.
We’ve been here before, but this time he’s even less restrained than he was in 2020 and is surrounded by people who will do his bidding.
His address tonight was absurd. It was riddled with so many lies that I’m reluctant to dignify them with rebuttals, but you should have them.
He mentioned a newly-declassified investigation into a voter registration group in Muskegon, Mich. that apparently had invited fraudulent registrations in 2020 — but Trump didn’t mention that the applications had been caught and none of them had resulted in any ballots being sent out incorrectly. The F.B.I. closed the investigation, stating “the investigation to date did not identify a criminal violation or a priority threat to national security.”
He alleged, once again, that foreign powers have hijacked votes, or that federal or state officials plotted to rig either the 2020 election.But no evidence has ever emerged showing that vote counts have been manipulated or corrupted. Intelligence reports, state audits of vote tallies and lawsuits have repeatedly affirmed official results in 2020 and other years. Nothing suggests China manipulated votes. Instead, U.S. intelligence assessment says China “probably also continued longstanding efforts” to gather information on U.S. voters and public opinion and to use that information to influence U.S. policy “as it has during all election cycles since at least 2008.”
The most significant foreign influence operations occurred in the 2016 presidential election and were conducted by Russia, in favor of Trump, according to the Mueller report. To the extent that this and other reports appeared to cast doubt on the legitimacy of Trump’s victory, they have had the effect of fueling his distrust of U.S. intelligence agencies.
Despite his repeated assertions that U.S. elections are not secure, Trump during his second term has significantly cut the budget of the Cybersecurity and Infrastructure Security Agency, including its election work. That’s because Trump grew contemptuous of the agency, and the government’s election security work generally, after it validated the integrity of the 2020 election.
So the entire performance tonight was fake — an extension of his Big Lie that the 2020 election was “stolen” from him.
It was also a commercial for the “Save America Act,” which would make it harder for many American citizens to vote. Voters would have to prove their citizenship in person upon registering to vote, with documents such as an enhanced form of REAL ID (a state ID card compliant with federal regulations) that indicates American citizenship; a birth certificate; a passport or military identification card.
An estimated 9 percent of eligible voters, or 21.3 million Americans, either do not have documents that prove their citizenship, such as passports and birth certificates, or cannot retrieve them in a day or less, according a study by the Center for Democracy and Civic Engagement at University of Maryland and the Brennan Center for Justice. And 45 states do not issue the kind of enhanced driver’s license indicating citizenship status that would be needed to verify voting eligibility.
The point is that American democracy is acutely endangered by a sociopath who will stop at nothing to get the results he wants.
This means that you and I and every other patriotic American have to do whatever we can to ensure free and fair elections, and fight Trump’s torrent of lies and authoritarian moves.
If you’re anything like me, you’re warn out by Trump. You’d like nothing better than to tune him out. I get it. But American democracy is seriously on the line here. We must keep up — and accelerate — the fight.
China and Europe have typically led the charge on open models. Now, another US company is joining the fray with a significant entrant into the race.
On Wednesday, Thinking Labs, the AI startup founded by ex-OpenAI CTO Mira Murati, launched its first model: Inkling. The model sets itself apart from leading AI labs in the US in a very important way: it is open-weight, allowing customers to customize it directly to their needs through fine-tuning and training. To suit that purpose, the model was built to perform across a wide range of areas. Or, as Thinking Machines describes it, it’s a “generalist model”that can reason about text, images, and audio rather than being proficient in just one domain. This can be observed in benchmark performance, which isn’t best-in-class but is consistent across specialties, as seen in the image below. Other model features include:
Parameters: Mixture-of-Experts transformer model with 975 billion total parameters, 41 billion active.
Context window: Up to one million tokens
Pretrained: On 45 trillion tokens of text, images, audio and video, allowing it to be proficient across all three.
Lightweight counterpart: Inkling-Small, a lighter-weight model with 12 billion active parameters.
Fine-tuning: To make customization accessible, the company says it is making Inkling available on Tinker, the company’s training API platform.
Cost-efficiency: Inkling’s “controllable thinking effort” allows users to customize the cost/performance curve.Until now, most open-source models have been developed in other countries, with China leading the way. These models are good for the ecosystem because they allow people and organizations to more deeply specialize the models, adjust their behavior, run them locally, and even save money. Though, as Thomas Randall, Research Director at Info-Tech Research Group, told The Deep View, they may not be for everyone. “If organizations have use cases where cost control, sovereignty, fine-tuning, deployment flexibility, or domain-specific performance matter, then open-source models are relevant,” said Randall. “For the highest-risk or most complex work (e.g., legal, medical, cyber, or regulated decisions), closed frontier model providers are preferable because of safety tooling, support models, auditability, and more mature deployment ecosystems.” Still, for the US to compete in the open-weights model space, it has a lot left to do, despite recent entrants from not only Thinking Machines but Nvidia, Google and OpenAI. A closer look at benchmark performance shows that many of China’s leading labs, which have also taken a generalist approach with their open-weight models, have outperformed Inkling on benchmarks. That includes GLM 5.2, DeepSeek V4 Pro, and Kimi K2.6. Kimi K3 also just released on Thursday, and it’s Kimi’s most capable model to date, with 2.8 trillion parameters.Competition is the lifeblood of any industry. This is especially true for the state of AI we are in now, where enterprises and AI builders are prioritizing cost, and open-weight models provide a tangible solution while applying pressure on the leading labs to be more cost-conscious. Just this month, OpenAI and Meta have both released new models, each highlighting their cost efficiency. Still, it is important to keep in mind the broader backdrop, including the rising costs of finite resources such as data centers and energy. Open-weight models still need the same compute power, even if they cost less than proprietary models.
Photo: Saul Loeb/AFP via Getty Images
Photo: Saul Loeb/AFP via Getty Images
Speech headlines in today’s Wall Street Journal, Washington Post.
Competition is the lifeblood of any industry. This is especially true for the state of AI we are in now, where enterprises and AI builders are prioritizing cost, and open-weight models provide a tangible solution while applying pressure on the leading labs to be more cost-conscious. Just this month, OpenAI and Meta have both released new models, each highlighting their cost efficiency. Still, it is important to keep in mind the broader backdrop, including the rising costs of finite resources such as data centers and energy. Open-weight models still need the same compute power, even if they cost less than proprietary models.
canisgallicus 