Nvidia Advances Lobbying While Microsoft Stock Dips

Nvidia recently secured a significant lobbying victory on December 3, 2025, as the U.S. House of Representatives rejected the GAIN AI Act of 2025. This proposed measure would have compelled chipmakers like Nvidia and AMD to prioritize American customers for advanced AI GPUs before selling to countries such as China. Nvidia CEO Jensen Huang met with President Trump and various lawmakers, arguing that such a rule would undermine U.S. competitiveness globally. The White House also supported Nvidia's position against the provision, though some lawmakers still plan to pursue stricter controls through a new bill called the Secure and Feasible Exports Act. Despite this win, China's existing ban on Nvidia hardware continues to limit its market presence there. In response to these dynamics and a broader push for technological independence, Chinese chipmaker Cambricon aims to significantly ramp up its AI chip production in 2026. The company plans to triple its output, delivering half a million AI accelerators, including 300,000 of its advanced Siyuan 590 and 690 chips, with the explicit goal of replacing Nvidia in the Chinese market. However, Cambricon faces challenges such as securing sufficient production capacity at SMIC and dealing with low manufacturing yields, reportedly only 20% for its top chips. Producing high-bandwidth memory chips also remains a difficult hurdle for Chinese companies. The broader AI market is experiencing investor jitters, highlighted by a recent dip in Microsoft's stock. This occurred after a report suggested the company lowered its AI growth targets, despite Microsoft denying any reduction in aggregate sales quotas. This reaction underscores investor anxiety about a potential AI bubble, a sentiment also observed with other major players like Meta, Oracle, and Nvidia. Daniel Newman, CEO of Futurum Group, points out that AI adoption often proceeds slower than investors anticipate, contributing to these concerns. Experts at the Stanford Institute for Economic Policy Research's "AI & the Economy" forum on November 20, 2025, including Google's James Manyika, echoed this, noting that AI adoption in workplaces is still low and concentrated in specific sectors. Amidst these market shifts, companies are pushing new AI initiatives. Amazon, for instance, announced on December 3, 2025, it will offer a free year of credits for its AI coding tool, Kiro, to eligible U.S.-based early-stage startups that have secured venture capital funding. Meanwhile, the practical application and ethical implications of AI are also coming to the forefront. Attorney Tyrone Blackburn recently faced criticism in a civil case against Fat Joe for blaming legal errors, including citing non-existent cases, on AI software. LexisNexis, the company Blackburn claimed to use, denied he was an authorized user of their AI products. Looking ahead to 2026, AI is set to transform enterprise security, shifting security teams towards generalists who leverage AI agents for specialized tasks. This evolution promises more adaptable teams and career mobility for professionals but also introduces new threats, such as rogue insiders using deepfakes for fraud. Offensive security will also advance, with AI-powered tools performing continuous vulnerability testing, moving beyond traditional manual methods. However, Matt Rogers, who helped build the first iPhone, warns the AI industry against solely focusing on chatbots. He advocates for a shift towards AI applications that solve real-world problems in sectors like security, healthcare, energy, food, and waste management, emphasizing the need for a new wave of apps that apply AI to physical systems, much like the App Store revolutionized smartphones.

Key Takeaways

• Nvidia successfully lobbied against the GAIN AI Act of 2025 on December 3, 2025, preventing a mandate to prioritize U.S. customers for advanced AI chips before selling to China.
• China's Cambricon plans to triple its AI chip output in 2026, aiming to deliver 500,000 AI accelerators, including 300,000 Siyuan 590 and 690 chips, to replace Nvidia in the Chinese market.
• Microsoft's stock experienced a dip, reflecting investor anxiety about the AI market and a potential "AI bubble," despite the company denying it lowered aggregate sales quotas.
• Amazon is offering a free year of credits for its AI coding tool, Kiro, to eligible U.S.-based early-stage startups with venture capital funding, announced on December 3, 2025.
• The Stanford Institute for Economic Policy Research forum on November 20, 2025, highlighted that AI adoption in workplaces remains low and concentrated, suggesting a slower spread than anticipated.
• Attorney Tyrone Blackburn blamed AI software for legal errors, including citing non-existent cases, in a civil brief, but LexisNexis denied he was an authorized user of their AI products.
• AI is projected to significantly transform enterprise security in 2026, leading to generalist security teams utilizing AI agents but also introducing new threats like deepfake fraud.
• Nvidia CEO Jensen Huang advocates for a single federal AI regulation in the U.S., arguing that state-by-state rules would slow the industry and create national security concerns.
• Matt Rogers, co-creator of the iPhone, warns the AI industry to shift its focus from chatbots to AI applications that solve real-world problems in sectors like security, healthcare, and waste management.
• Cambricon faces challenges in its AI chip production ramp-up, including securing capacity at SMIC and low manufacturing yields of only 20% for its top chips, alongside difficulties in producing high-bandwidth memory.

Nvidia wins against new chip export rules to China

Nvidia successfully lobbied against a new measure called the GAIN AI Act of 2025. This act would have made companies like Nvidia and AMD prioritize American customers for advanced AI GPUs before selling to countries like China. Nvidia CEO Jensen Huang met with President Trump and lawmakers on Wednesday, December 3, 2025, arguing the rule would hurt US competitiveness. The U.S. House of Representatives rejected the proposal, which was a win for Nvidia. However, China's own ban on Nvidia hardware still limits its market there.

Nvidia CEO Huang discusses chip rules with Trump

Nvidia CEO Jensen Huang met with President Trump on Wednesday, December 3, 2025, to discuss chip export restrictions. Huang also spoke with lawmakers about the GAIN AI Act, which would have required chipmakers like Nvidia and AMD to prioritize US customers for AI chips. He called the decision to exclude this proposal from the defense bill "wise." Huang also criticized state-by-state AI regulations, saying they would slow down the industry and create national security concerns. He believes a single federal AI regulation is the best approach for the United States.

Nvidia wins as Congress rejects chip export bill

Nvidia CEO Jensen Huang achieved a major lobbying victory on December 3, 2025, as Congress decided not to include the GAIN AI Act in the defense bill. This act would have forced chipmakers like Nvidia and AMD to prioritize American customers for advanced AI chips before selling to China. Huang met with President Donald Trump and lawmakers, arguing the legislation would harm global competition. The White House also supported Nvidia's stance against the provision. Despite this, some lawmakers plan to continue pushing for stricter controls with a new bill called the Secure and Feasible Exports Act.

Experts discuss real impact of AI on economy

The Stanford Institute for Economic Policy Research (SIEPR) held a forum called "AI & the Economy" on November 20, 2025. Experts like Fei-Fei Li, known as the "godmother of AI," discussed the true effects of AI, moving past extreme predictions. Speakers including Senator Mark Kelly and Google's James Manyika explored AI's impact on jobs, productivity, energy, and global competition. A key theme was how government policy can help manage AI's risks and benefits. They noted that AI adoption in workplaces is still low and concentrated in specific sectors, suggesting a slower spread than expected.

Amazon offers free AI coding tool Kiro to startups

Amazon plans to boost its AI coding tool, Kiro, by offering a free year of credits to eligible early-stage startups. This move aims to help Kiro compete with other popular AI coding tools. To qualify, startups must be based in the U.S. and have secured venture capital funding from pre-seed to Series B rounds. The offer has some geographic limits and is not available in certain countries like France, Germany, and Italy. This initiative was announced on December 3, 2025.

Lawyer blames AI for errors in Fat Joe case

In the civil case against Fat Joe, attorney Tyrone Blackburn is facing criticism for blaming legal mistakes on artificial intelligence software. Fat Joe's lawyers claimed Blackburn cited non-existent cases in his brief. Blackburn then stated he used a LexisNexis AI feature, but LexisNexis denied this. The company's head of legal confirmed Blackburn was not an authorized user of their AI products and is not responsible for his errors. Blackburn maintains he used LexisNexis and plans to publish his research and invoice.

AI will transform enterprise security in 2026

In 2026, AI will significantly change enterprise security and company structures. Security teams will shift towards generalists who use AI agents for specialized tasks, making them more adaptable. This will benefit CISOs by creating flexible teams and offering career mobility for security professionals. However, AI will also bring new threats, as rogue insiders may use deepfakes to commit fraud, making it harder to trust digital communications. Offensive security will also evolve, with AI-powered tools performing continuous, contextual vulnerability testing, moving beyond traditional manual pen-testing.

China's Cambricon plans to triple AI chip output

Chinese chipmaker Cambricon aims to triple its AI chip output in 2026 to replace Nvidia in China. The company plans to deliver half a million AI accelerators, including 300,000 of its advanced Siyuan 590 and 690 chips. This ramp-up follows China's efforts to reduce reliance on US technology and its discouragement of Nvidia products. However, Cambricon faces challenges like securing production capacity at SMIC and low manufacturing yields of only 20% for its top chips. High-bandwidth memory chips also remain a difficult technology for Chinese companies to produce.

Microsoft stock dip raises investor AI concerns

A recent dip in Microsoft's stock, caused by a report that the company lowered its AI growth targets, highlights investor anxiety about the AI market. Although Microsoft denied lowering aggregate sales quotas, the market reacted quickly, showing nervousness about a potential AI bubble. Other major AI players like Meta, Oracle, and Nvidia have also seen investor jitters despite strong performance. Daniel Newman, CEO of Futurum Group, noted that AI adoption is not always as fast as investors hope, contributing to these concerns. This market reaction reflects a broader unease about the technology's rapid expansion.

iPhone builder warns AI industry about app focus

Matt Rogers, who helped build the first iPhone with Steve Jobs, warns that the AI industry risks repeating an early smartphone mistake. He believes the focus should shift from just chatbots to AI applications that solve real-world problems like security, healthcare, and waste. Rogers emphasizes that while infrastructure investment is crucial, the industry needs a wave of new apps that apply AI to physical systems. He compares this to the success of the App Store, which unlocked countless applications beyond the iPhone's hardware. Rogers, now CEO of Mill, suggests the next big opportunities lie in applying AI to sectors like energy, food, and waste management.

Sources

• Nvidia lobbies White House and wins loosened AI GPU export control to China — U.S. lawmakers reportedly reject GAIN AI Act
• Nvidia CEO Jensen Huang talks chip restrictions with Trump, blasts state-by-state AI regulations
• Nvidia Scores Lobbying Win as Congress Rejects Chip Export Bill
• Cutting through the AI noise
• Amazon hopes to jump start its AI coding tool Kiro by giving it away to startups
• Lawyer Suing Fat Joe Called Out For Blaming Legal Mistakes On AI
• 2026 is the Year AI Attacks Your Enterprise and your Org Chart
• Cambricon Aims to Triple Chip Output to Replace Nvidia in China
• Microsoft dip creates latest excuse for investors to doubt AI trade
• I built the first iPhone with Steve Jobs. The AI industry is at risk of repeating an early smartphone mistake