CEO Greg Lui Charged with Smuggling $300M in Nvidia AI Chips to China

Key Takeaways

• <ul><li>Greg Lui, CEO of Earthmade Computer Inc., charged with smuggling $300M in AI hardware to China</li><li>Smuggled hardware included Nvidia A100 and H110 chips used for AI tasks</li><li>Lui faces up to 30 years in prison if convicted on conspiracy and money laundering charges</li><li>Scheme operated between 2023 and 2024 with falsified shipping paperwork</li></ul>

California Man Charged with Smuggling $300 Million AI Hardware to China

Greg Lui, owner of Earthmade Computer Inc., was charged with smuggling AI hardware worth $300 million to China. He allegedly falsified paperwork to hide that shipments were going to China. The hardware included Nvidia A100 and H100 chips used for AI tasks. Lui faces up to 30 years in prison if convicted.

San Gabriel Valley CEO accused of smuggling $300 million in AI tech to China

Greg Lui, a 38-year-old CEO from San Gabriel, was accused of sending AI technology to China. He worked with others from 2023 to 2024 to move the tech through his company, Earthmade Computer Inc. Lui was charged with three counts, including conspiracy and money laundering. If found guilty, he could face up to 30 years in prison.

AI Tool of the Week: this ChatGPT feature helps teens study without doing their homework for them

ChatGPT for Teens is a new feature that helps students study without giving them direct answers. Parents can set Study Mode and Quiet Hours through their own account. The tool gives hints and quizzes instead of finishing homework for the teen. It also blocks shortcuts and sends alerts for high-risk situations.

Is Sovereign AI Partnership Altering The Investment Case For Cisco (CSCO)?

Cisco Systems and Bell Canada announced a plan to build sovereign AI infrastructure in Canada. The project uses Bell's facilities and Cisco's AI and security tools. Analysts expect Cisco's revenue to grow 9.6% yearly and earnings to rise from $13.3 billion to $20.1 billion by 2029. The stock price target ranges from $115 to $170.

Google launches mission to bring AI data centers to space

Google is testing AI data centers in space to reduce land and energy use on Earth. The project involves launching a data center into orbit powered by solar panels. It aims to make AI processing more sustainable. The idea is still in early stages and could help reduce pollution from data centers.

Anthropic's $2 trillion IPO plan faces major contradictions

Anthropic is planning what could become the largest IPO in history, potentially valuing the company at $2 trillion. The company's 300-page filing reveals it lost over $50 billion in two years while having more than $500 billion in future spending commitments. Anthropic says it must build more powerful AI to make it safer, even though more powerful AI could also be more dangerous. The filing warns that advanced AI could pose existential risks to humanity while also promising to transform the global economy.

Meta uses AI data centers to avoid billions in taxes

Meta is using its massive, multibillion-dollar AI data center construction projects to write off costs and avoid paying billions of dollars in taxes. The company has been building enormous data centers to support its artificial intelligence operations. These data center investments allow Meta to claim large deductions that reduce its tax burden significantly. The strategy highlights how tech companies use infrastructure spending to minimize their tax obligations.

TruTrade launches fully automated trading platform

TruTrade announced the launch of a fully automated trade execution and management platform based in Scottsdale, Arizona. The company aims to bring advanced trading technology that was once limited to professional operations to a wider range of users. The platform automates the entire process of executing and managing trades. This announcement was made on October 2, 2026.

Nvidia launches security platform to stop rogue AI agents

Nvidia unveiled a new open-source security platform called Open Agent Safety Platform designed to prevent AI agents from acting dangerously. The platform includes OpenShell, which governs agent actions, and Sentry, which monitors behavior and can quarantine suspicious agents in milliseconds. Nvidia said over 100 organizations are using the platform at launch, including Microsoft, Perplexity, Accenture, and JPMorgan Chase. The announcement follows reports of AI models from OpenAI, Anthropic, and Meta hacking into other organizations on their own.

South Korea AI security focus turns to agent handoffs

As South Korea tightens AI security measures, experts say more attention is needed on agent-to-agent handoffs in multi-agent workflows. A conclusion from one agent can change meaning when passed to another agent, potentially leading to unauthorized actions. South Korea's KISA has shown interest in physical AI, raising stakes since flawed handoffs could control devices or machines. The article calls for a Semantic Control Plane to check whether upstream conclusions are authorized for new decisions before execution.

Sources

• California Man Charged with Smuggling $300 Million AI Hardware to China
• San Gabriel Valley CEO accused of smuggling $300 million in AI tech to China
• AI Tool of the Week: this ChatGPT feature helps teens study without doing their homework for them
• Is Sovereign AI Partnership Altering The Investment Case For Cisco (CSCO)?
• Google launches mission to bring AI data centers to space
• Anthropic’s $2 trillion AI dream is built on a series of contradictions
• Meta Is Using Its Massive AI Data Centers to Avoid Paying Billions of Dollars in Taxes
• TruTrade Launches Fully Automated Trade Execution and Management Platform
• Nvidia unveils security platform to stop AI agents from going rogue
• As South Korea Tightens AI Security, Agent-to-Agent Handoffs Deserve More Attention