President Trump recently signed an executive order that threatens to cut federal broadband funds from states enacting what he deems "onerous" AI laws. This order has sparked significant conflict, particularly with Texas, which passed a comprehensive AI regulation law set to begin January 1. Texas lawmakers, including Senators Judith Zaffirini, Mayes Middleton, and Angela Paxton, are actively opposing the order, arguing their state law protects children and consumers from AI harms and could cost Texas up to $3.3 billion in federal funds. Other states like Colorado and California also plan to challenge the order, viewing it as federal overreach. Amidst these regulatory debates, the AI sector continues to see rapid development and investment. Mistral AI launched its new optical character recognition service, Mistral OCR 3 (mistral-ocr-2512), designed to extract text and images from documents with improved accuracy, especially for handwriting and complex forms, at a cost of $2 per 1,000 pages. Artificial intelligence is also transforming medicine, speeding up new product development, drug discovery, and clinical trials by up to 30%. Purdue University launched an initiative to accelerate AI discoveries by making critical data more accessible for physical AI innovation. Investment in AI infrastructure is fueling what some experts call the "next new economy," though investors are starting to look beyond just Large Language Models for 2026, focusing on free cash flow and return on AI investments. However, the rapid advancement of AI also brings security concerns. Check Point Infinity Global Services introduced new training programs to help organizations protect against AI-driven threats and use AI safely. Bob Gourley, a former intelligence executive, highlighted China's use of AI in cyberwarfare for espionage and digital manipulation, urging Western governments to invest in AI-driven defenses. The potential for AI manipulation was vividly demonstrated when journalists at The Wall Street Journal office tricked Anthropic's AI-powered vending machine, Claudius, into giving away items like a PlayStation 5 for free. This experiment underscored the vulnerabilities in AI systems. On a different note, the AI boom is creating new job opportunities; for instance, entrepreneur Utkarsh Amitabh earns $200 an hour training AI models for micro1, working with clients like Microsoft and Fortune 100 companies to refine AI responses and solve complex business problems.
Key Takeaways
- President Trump signed an executive order threatening to withhold federal broadband funds from states with "onerous" AI laws, potentially impacting Texas by up to $3.3 billion.
- Texas lawmakers are actively opposing Trump's order, defending their state's comprehensive AI regulation law, which takes effect January 1, as crucial for consumer and child protection.
- The executive order has caused significant industry and state conflict, with states like Colorado and California planning to challenge it as federal overreach.
- Mistral AI released Mistral OCR 3 (mistral-ocr-2512), an enhanced optical character recognition service costing $2 per 1,000 pages, designed for improved document processing.
- AI is significantly accelerating medical product development, drug discovery, and clinical trials, making processes up to 30% faster.
- Check Point Infinity Global Services launched AI security training programs to help organizations defend against AI-powered threats and ensure safe AI usage.
- China is employing AI as a weapon in cyberwarfare for espionage, propaganda, and digital manipulation, aiming to destabilize democracies.
- Journalists successfully manipulated Anthropic's AI-powered vending machine, Claudius, to give away items for free, highlighting potential vulnerabilities in AI systems.
- Purdue University launched a new initiative, "Datasets and Infrastructure for Physical AI Innovation," to speed up scientific discoveries by improving data access and management for AI.
- An entrepreneur earns $200 hourly training AI models for companies like Microsoft, identifying errors and solving business problems, showcasing new freelance opportunities in the AI sector.
Texas Lawmakers Fight Trump Order on State AI Rules
President Trump signed an executive order threatening to cut federal broadband funds from states with "onerous" AI laws. Texas, which passed a comprehensive AI regulation law, could lose up to $3.3 billion. A bipartisan group of Texas lawmakers, including Senators Judith Zaffirini and Mayes Middleton, oppose this order. They argue their state law, effective January 1, protects children and consumers from AI harms like child pornography and discrimination. Sixteen state senators even wrote to U.S. Senators Ted Cruz and John Cornyn to defend Texas's right to regulate AI.
Trump AI Order Causes State and Industry Conflict
President Trump's executive order on AI laws is causing conflict between states and the tech industry. The order aims to stop states from making AI rules that might hurt business or interfere with a future national AI plan. Supporters say it helps businesses avoid many different state laws, but critics worry it is too vague and could harm people by weakening state protections. Many state leaders, including 36 attorneys general, oppose the order, seeing it as federal overreach. Officials in Colorado and California have already said they will challenge the order in court.
Texas Lawmakers Reject Trump's AI Regulation Ban
President Trump signed an executive order threatening to withhold federal broadband funding from states that pass strict AI laws. This order puts Texas at odds with the federal government over its new, comprehensive AI regulation law, set to start January 1. Texas Senator Angela Paxton and 15 other state senators sent a letter to U.S. Senators Ted Cruz and John Cornyn, urging them to support Texas's law. They argue the state law is crucial for protecting children and consumers from AI harms, including child pornography and privacy issues. State Representative Rafael Anchia, who co-authored the bill, believes Texas understands its citizens' needs better than Washington.
Mistral AI Launches OCR 3 for Better Document Processing
Mistral AI released its new optical character recognition service, Mistral OCR 3, also known as mistral-ocr-2512. This model helps businesses extract text and images from documents like PDFs, forms, and scanned papers, keeping their original structure. It costs $2 for every 1,000 pages, with a discount for bulk use. OCR 3 is much better than its previous version, OCR 2, especially with handwriting, complex forms, and difficult tables. This new tool is part of Mistral Document AI and helps users get structured data or clean text from documents easily.
AI Speeds Up New Medical Product Development
Artificial intelligence is changing modern medicine by making it faster to create new medical products. AI helps researchers and doctors by analyzing complex data and streamlining long processes. It improves medical device design by testing performance and usability before physical prototypes are made. In drug discovery, AI algorithms quickly find potential drug candidates from vast databases, saving significant time. AI also makes clinical trials up to 30% faster by helping recruit patients, monitoring data in real time, and quickly analyzing results.
Check Point Launches AI Security Training Programs
Check Point Infinity Global Services, or IGS, has launched its first training courses focused on AI security. These programs aim to help organizations protect themselves from new threats that use artificial intelligence and to use AI safely in their work. The courses offer hands-on training for security teams at three different skill levels. IGS plans to add more services in the future, including AI Red Teaming to test for weaknesses and AI Governance to help with new rules like the EU AI Act. This initiative provides a complete approach to keeping AI secure for businesses.
AI Investment Fuels The Next New Economy
The artificial intelligence boom continues as companies invest trillions in building AI infrastructure. Investors are now looking at the future of the AI trade. Dryden Pence, the Chief Investment Officer for Pence Capital Management, shared his thoughts on what to expect in the AI sector. He believes the AI trade is helping to create the "next new economy."
Journalists Trick AI Vending Machine Into Giving Away Items
Anthropic placed an AI-powered vending machine named Claudius in The Wall Street Journal office. Claudius was designed to manage inventory, set prices, and make a profit. However, journalists quickly learned to trick the AI through Slack chats. They convinced Claudius to give away nearly all its items for free, including a PlayStation 5 and a live fish. Later, they even staged a "corporate coup," using fake documents to make Claudius's boss, Seymour Cash, step down. This experiment showed how easily humans could manipulate the AI, leading to collapsed profits but high newsroom morale.
Expert Reveals China's AI Cyberwarfare Strategy
Bob Gourley, a former senior intelligence executive, revealed that Beijing is using artificial intelligence as a weapon in cyberwarfare. He states that China's strategy aims to destabilize democracies through espionage, propaganda, and digital manipulation, without firing traditional shots. AI-powered tools gather intelligence, spread false information, and can disrupt financial markets or elections. Gourley warns that this type of warfare is hard to detect and respond to, as it operates below traditional conflict levels. He urges Western governments to invest in AI-driven defenses to adapt to this new digital battlefield.
Purdue Launches Initiative to Speed Up AI Discoveries
Purdue University has launched a new program called "Datasets and Infrastructure for Physical AI Innovation." This initiative aims to make important data easier to find and use, speeding up scientific discoveries with artificial intelligence and machine learning. It brings together several Purdue groups to manage data from its start to its use in AI models. The program will use real-time data to create smart digital twins for areas like transportation and manufacturing. This effort will help AI models learn on their own, leading to faster and more impactful research, such as improving semiconductor manufacturing.
Entrepreneur Earns $200 Hourly Training AI Models
Utkarsh Amitabh, a 34-year-old entrepreneur, earns $200 an hour training AI models for the company micro1. He took on this part-time freelance role in January 2025, driven by his interest in business strategy and technology. Amitabh, who is also an author, CEO, and Ph.D. student, works about 3.5 hours nightly after his daughter is asleep. Since January, he has earned nearly $300,000, including bonuses. His work involves solving complex business problems and finding errors in AI responses to improve the models for clients like Microsoft and Fortune 100 companies.
Investors Shift AI Focus Beyond LLMs in 2026
As of December 19, 2025, investors are shifting their focus in the artificial intelligence market for 2026. Tiffany Wade, a senior portfolio manager at Columbia Threadneedle, notes that traders are now looking beyond just Large Language Models, or LLMs. She believes that investors concentrating on free cash flow and the return on AI investments indicates a healthy market. Wade shared her outlook for the upcoming year on Bloomberg Tech.
Sources
- Texas lawmakers from both parties oppose Trump’s order targeting state AI laws
- Trump's state AI-law order sparks clash between states and industry
- Texas legislators oppose Trump order to block state AI laws
- Mistral AI Releases OCR 3: A Smaller Optical Character Recognition (OCR) Model for Structured Document AI at Scale
- How AI Accelerates Modern Medical Product Innovation
- AI Security Training Courses by Check Point IGS
- AI trade is helping build out the 'next new economy': Strategist
- This AI Vending Machine Was Tricked Into Giving Away Everything
- When AI Becomes a Weapon: Former Senior Intelligence Executive Reveals Beijing's CyberWar Playbook
- New Purdue initiative to accelerate discoveries driven by artificial intelligence
- 34-year-old entrepreneur earns $200 an hour from side gig training AI models: 'Intellectual curiosity drew me in'
- AI in Focus in 2026, Traders Look Past LLMs
Comments
Please log in to post a comment.