U.S. lawmakers are pressing to tighten controls on China’s access to advanced AI chips, with several proposals gaining attention amid scrutiny of Nvidia and its former CEO Jensen Huang. The bills would expand congressional review of export-rule changes, require tracking of advanced processors and reporting of suspected diversions, pressure allies such as the Netherlands and Japan to restrict chipmaking equipment sales and servicing, and regulate Chinese firms’ access to computing power rented from overseas data centers. The Remote Access Security Act passed the House in January and awaits Senate consideration; lawmakers may attach proposals to annual defense legislation.
OpenAI told lawmakers it supports a bipartisan package to limit China’s access to advanced chips and chipmaking equipment, including equipment from ASML. Anthropic has also backed tougher export controls. Nvidia’s allies oppose the proposals, arguing they could harm companies connected to its business. The debate has unsettled markets: Hong Kong-listed Z.AI fell 10.88% and MiniMax 8.73%, while Nvidia shares dropped as much as 14% on Thursday before closing down 4.3%. CFO Colette Kress became interim CEO as Nvidia’s board seeks a permanent replacement. Chinese firms’ use of overseas data centers to rent computing power is one target of proposed restrictions.
Market commentators offered differing assessments of the fallout. Futurum Group CEO Daniel Newman said the AI sector remains resilient and projected Nvidia would hold 41% market share within two years; he said AMD and Broadcom could need two to three years to shift the complex global supply chain. European shares slipped 0.2% on Friday after a record close, with the Stoxx 600 at 662.64 points and its technology sub-index down 0.6%. The index nevertheless rose more than 2% for the week, its strongest weekly gain since June, and Goldman Sachs raised its 12-month forecast from 695 to 700 points.
Other reports describe AI’s effects across business and technical work. A November 13 report says 74% of small and medium-sized businesses use AI at least occasionally. Of those users, 60% reported producing more without hiring, 72% finishing tasks faster, and 45% taking on work beyond their previous capabilities. More businesses skipped a planned full-time hire because of AI than laid off a worker—13% versus 3.3%. Customer service was the most common use, followed by marketing, bookkeeping and sales; 23% gained customers seeking human-performed work, while 14% took jobs correcting AI output.
A freight-classification study used 5,964 records and 47 features to test six machine-learning approaches and explanations from SHAP and LIME. Tree models averaged 86.8% test accuracy, compared with 83.3%–85.0% for other models. The leading explanation features had about four times the effect on predictions as randomly selected features, but models with similar accuracy often disagreed about which features mattered. Researchers recommend checking explanation stability and faithfulness before using them to guide freight decisions.
In engineering, SoftInWay says its AxSTREAM AI tools speed early design work while retaining physics-based checks and engineering judgment. A pre-trained compressor model added in October uses 15 design variables, speed and mass flow to produce 51 flow-path parameters, performance maps, and surge and choke limits. Trained on 34,305 designs and about 4.2 million performance points, most test predictions came within 3% of results from the company’s 2D streamline solver. The work began more than five years ago with NASA-backed research alongside Raytheon Technologies and was validated against Purdue University Zucrow Laboratories compressor data; 3D CFD and mechanical design remain part of the process.
Several company updates round out the news. Singapore-headquartered X Star Technology announced a nearly A$10 million investment in AutoSwift to scale AI WINGMAN. Paloren listed U.S. corporate AI workshops starting at US$4,500 and department programs from US$18,000. eMemory reported Q3 2026 revenue of TWD 1,210 million, up 27.1% year over year, record earnings per share and security-related revenue growth of more than 200%; management expects royalty growth supported by advanced nodes, PUF deployments and AI-security demand. Figure AI, valued at US$39 billion in its September 2025 Series C after raising more than US$1 billion in committed capital, had announced no IPO date as of November 3, 2026, and no public S-1 appeared in SEC filings.
Key Takeaways
- Congress is considering four measures to limit China’s access to advanced AI chips, chipmaking equipment and rented overseas computing power; the Remote Access Security Act passed the House and awaits Senate action.
- OpenAI supports a bipartisan package of chip and equipment restrictions, and Anthropic also supports tougher export controls; Nvidia’s allies oppose the proposals.
- Chinese AI shares fell after the Nvidia news: Z.AI dropped 10.88% and MiniMax 8.73%; Nvidia fell as much as 14% before closing down 4.3%, with Colette Kress serving as interim CEO.
- Daniel Newman said Nvidia could hold 41% market share within two years and estimated AMD and Broadcom may need two to three years to shift the global supply chain.
- Small-business AI use is widespread in a November 13 report: 74% use it at least occasionally, and 60% of users say they produce more without adding staff.
- A freight-model study found tree models averaged 86.8% accuracy, but explanations from similarly accurate models often disagreed about which features drove predictions.
- SoftInWay’s AxSTREAM AI compressor model was trained on 34,305 designs and about 4.2 million performance points; most test predictions were within 3% of its 2D solver results.
- X Star Technology is investing nearly A$10 million in AutoSwift to scale AI WINGMAN; Paloren’s U.S. AI workshops start at US$4,500.
- eMemory reported Q3 2026 revenue of TWD 1,210 million, up 27.1% year over year, with security-related revenue up more than 200%.
- Figure AI had no announced IPO date as of November 3, 2026; its September 2025 Series C raised more than US$1 billion in committed capital at a US$39 billion valuation.
Congress Moves Toward Tighter Limits on Nvidia Chips for China
U.S. House Speaker Mike Johnson said support is growing in Congress for limiting exports of advanced Nvidia chips to China. Lawmakers from both parties called for tougher restrictions after reports linked Nvidia CEO Jensen Huang to a businesswoman whom U.S. authorities suspect of spying for China. Nvidia chips help power advanced AI development, including China’s growing AI industry. Senior White House officials have also discussed whether Commerce Secretary Howard Lutnick should leave his post.
Huang’s Resignation Fuels Calls to Restrict Nvidia Chip Sales to China
Support for tighter U.S. controls on advanced Nvidia chips is growing in Congress after CEO Jensen Huang stepped down. Huang resigned after reports of an alleged relationship with Alice Huang, a Chinese businesswoman whom U.S. authorities suspect of spying for Beijing. The news also raised scrutiny of Nvidia’s business ties and Commerce Secretary Howard Lutnick, while Chinese AI stocks fell as investors feared new restrictions. Lawmakers are debating whether chip exports could pose national security risks, while others argue that U.S. companies should keep access to the Chinese market.
Chinese AI Stocks Fall as Nvidia Turmoil Raises Chip Supply Fears
Chinese AI shares fell on Friday as Nvidia CEO Jensen Huang’s resignation raised fears that Washington could further limit China’s access to advanced computing. Hong Kong-listed Z.AI fell 10.88% and MiniMax dropped 8.73%, while Tencent, Alibaba and other shares also declined. Nvidia shares fell as much as 14% on Thursday before closing down 4.3%; Chief Financial Officer Colette Kress became interim CEO as the board seeks a permanent replacement. U.S. export rules block direct sales of the most advanced processors to China, but some Chinese firms rent computing power from overseas data centers. Congress is considering the Remote Access Security Act, which would let the Commerce Department regulate such access, while Chinese chipmakers have recently gained as investors expect tighter controls to encourage local alternatives.
Four U.S. Bills Could Tighten Rules on AI Chips for China
U.S. lawmakers from both parties are pushing proposals to limit China’s access to advanced AI chips after reports about former Nvidia CEO Jensen Huang. The AI Overwatch Act would give Congress more power to review export rule changes, while the Chip Security Act would require companies to track advanced processors and report suspected diversions. The Match Act would pressure U.S. allies, including the Netherlands and Japan, to restrict sales and maintenance of advanced chipmaking equipment for China. The Remote Access Security Act would let the Commerce Department regulate Chinese firms that rent computing power from overseas data centers; it passed the House in January and awaits Senate consideration. Lawmakers are considering adding the proposals to annual defense legislation.
OpenAI Backs U.S. Effort to Restrict China’s Access to AI Chips
OpenAI has told lawmakers it supports a bipartisan package of measures to limit China’s access to advanced AI chips and chipmaking equipment. Congress is considering three proposals that would restrict future sales of high-end chips, reduce the diversion of Nvidia processors, and limit access to equipment from ASML. Lawmakers may attach the measures to annual defense legislation that must pass before year’s end. OpenAI’s support was under consideration before the scandal that led to Jensen Huang’s departure as Nvidia CEO. Anthropic has also supported tougher export controls, while Nvidia’s allies have opposed the proposals, saying they could harm companies tied to its business.
Nvidia CEO resignation shakes up the AI market
On November 12, 2026, Veneroso Wealth Management CEO Mario Veneroso discussed the market impact of NVIDIA CEO Jensen Huang’s resignation on Fox Business’s “Making Money.” He examined what the news could mean for the broader artificial intelligence trade.
Daniel Newman says AI trade can withstand NVIDIA troubles
Futurum Group CEO Daniel Newman told Taylor Riggs that the artificial intelligence sector remains resilient despite recent controversies involving NVIDIA. Newman said NVIDIA is projected to hold a 41% market share within two years. He added that competitors AMD and Broadcom would need two to three years to shift the complex global supply chain.
Study tests whether AI explanations for freight decisions can be trusted
Researchers tested machine learning models that classify shipments as Air or Air Charter, or as Truck or Ocean, using 5,964 records and 47 features from the PEPFAR Supply Chain Management System Delivery History dataset. They compared three Multi-Layer Perceptron models, Logistic Regression, Random Forest, and gradient boosting, and assessed SHAP and LIME explanations across 100 random trials. Tree models had the highest average test accuracy at 86.8%, compared with 83.3% to 85.0% for the other models. The most important features identified by SHAP or LIME had about four times the effect on predictions as randomly selected features, but models with similar accuracy often disagreed about which features mattered. The study says explanation stability and faithfulness should be checked before using explanations to guide freight decisions.
Most small businesses using AI handle more work without new hires
A report dated November 13, 2026, found that 74% of small and medium-sized businesses use AI at least occasionally. Among businesses that use it, 60% said they produce more work without adding staff, and 72% said they finish tasks faster. More users skipped a planned full-time hire because of AI than laid off a worker, 13% compared with 3.3%; 45% said AI helped them take on work beyond their previous abilities. Customer service was the most common use, followed by marketing, bookkeeping, and sales. The report also found that 23% gained customers who wanted a person to do the work, while 14% took on jobs fixing AI-generated output.
Singapore’s X Star Technology invests nearly A$10 million in AutoSwift
Singapore-headquartered automotive finance technology company X Star Technology is investing close to A$10 million in AutoSwift. The November 12, 2026 announcement says the investment will help scale AI WINGMAN.
European stocks slip as Nvidia news raises AI investment concerns
European stocks fell 0.2% on Friday after reaching a record high the day before, as worries about AI investment spread after Nvidia CEO Jensen Huang’s surprise resignation. The Stoxx 600 stood at 662.64 points at 08:08 GMT, while its technology sub-index dropped 0.6%. Shares of chipmakers Siltronic and Aixtron fell more than 1%, and Richemont shares dropped 7% after reporting slower sales growth. Despite Friday’s decline, the Stoxx 600 gained more than 2% during the week, its strongest weekly rise since June, and Goldman Sachs raised its 12-month forecast for the index from 695 to 700 points.
AxSTREAM AI speeds up early engineering design work
SoftInWay says AxSTREAM AI helps engineers build models, explore designs and run early analyses faster, while keeping engineering judgment and physics-based checks in place. Its AI assistant, Peter, can configure system models and run analysis steps in AxSTREAM System Simulation. A pre-trained compressor model added in October 2026 uses 15 design variables, speed and mass flow to return 51 flow-path parameters, performance maps and surge and choke limits. The model was trained on 34,305 designs and about 4.2 million performance points, with most test predictions within 3% of results from AxSTREAM’s 2D streamline solver. SoftInWay says the work began more than five years ago with NASA-backed research alongside Raytheon Technologies and was validated against compressor test data from Purdue University’s Zucrow Laboratories; 3D CFD and mechanical design remain part of the process.
Figure AI remains private with no announced IPO plan
Sunnyvale, California-based Figure AI develops general-purpose humanoid robots for warehouses, factories and, eventually, homes. As of November 3, 2026, the company had announced no IPO date, and no public S-1 registration statement appeared in SEC filings. Its September 2025 Series C brought in more than $1 billion in committed capital and valued the company at $39 billion, up from a $2.6 billion valuation in its February 2024 Series B. Figure’s website warned in July 2026 about unauthorized attempts to sell its stock. Investors cannot buy Figure shares through ordinary stock-market orders, though private offerings may be available with limits and fees that vary.
Paloren lists prices for US corporate AI training
Paloren announced corporate AI workshops in the United States starting at US$4,500. Department-level programs start at US$18,000 and use role-based cohorts. The article provides these prices and a department delivery plan.
eMemory reports record third-quarter growth led by security IP
eMemory Technology reported Q3 2026 revenue of TWD 1,210 million, up 27.1% year over year, with record earnings per share. Security-related revenue rose by more than 200%. Management expects royalty growth to continue, supported by advanced nodes, PUF deployments and demand for AI security. The figures come from an audio transcript dated November 13, 2026, and its summary warns that it may contain inaccuracies.
Sources
- China’s access to Nvidia chips in doubt as curbs gain traction in U.S. Congress
- Will Washington Cut China's Access to Nvidia AI Chips After Jensen Huang's Resignation?
- China AI stocks tumble as Nvidia fallout fuels fears of tighter U.S. chip curbs
- Explainer-Tougher curbs on AI chip sales to China likely after Nvidia CEO resigns
- OpenAI joins push to curb China’s access to AI chips and chipmaking tools
- Stocks volatile as Nvidia CEO exit rattles AI trade
- Daniel Newman: AI trade is intact despite NVIDIA headwinds
- A Reliability Study of Explainable AI for Freight Mode Classification
- 60% of Small Business AI Users Do More Without Adding Staff
- Singapore's X Star Technology invests close to A$10 million in AutoSwift to scale AI WINGMAN
- European Stocks Retreat Amid Concerns Over Artificial Intelligence
- AI Tools for Engineers: AxSTREAM AI
- Figure AI IPO: Timeline, Valuation, Investment Options
- Corporate AI Training in the USA: Paloren Publishes Workshop Prices and a Department Delivery Plan
- eMemory Technology: Record Q3 growth was led by security IP, with AI trust and advanced nodes powering the outlook
Comments
Please log in to post a comment.