New York Governor Kathy Hochul and Democratic lawmakers are pushing for broader AI regulation even before the state's first AI safety law takes effect, citing inaction from the Trump administration. Hochul stated New York will act where Washington refuses, while lawmakers debate rules for data centers and chatbots for minors. Meanwhile, Tennessee lawmakers are weighing proposals to protect children from AI, requiring a study on AI and child safety. Congressional pressure is also mounting, with Republican Speaker Mike Johnson favoring common sense and self-regulation, while Democratic leader Hakeem Jeffries calls for immediate action.
Nvidia founder and CEO Jensen Huang pushed back against calls for AI regulation, saying safety is an engineering problem rather than a legal one. He argues the free market and company choices are enough, and that businesses should slow down if they lack confidence in a product. Huang made the remarks at Salesforce's Dreamforce conference on September 15, 2026.
Microsoft's top AI researcher, Dr. Eric Horvitz, challenged Anthropic's perspective on AI consciousness, calling it based on a flawed understanding of human intelligence. The debate matters as the industry searches for ways to make AI safer. Anthropic, backed by investors including Google and Facebook, holds a different view on these questions.
On the practical side, enterprises are holding onto legacy systems like mainframes longer as AI investment grows. Ensono's report finds 45 percent of firms are scaling AI deployments and 44 percent are targeting high-impact use cases, with many viewing legacy infrastructure as a foundation for AI rather than something to replace. NVIDIA's Resiliency Extension is also helping, enabling fault-tolerant PyTorch FSDP training on Amazon EKS with over 99 percent training efficiency and second-scale recovery from GPU faults.
Key Takeaways
- <ul><li>New York is expanding AI regulation even before its first AI safety law takes effect, with Governor Hochul vowing to act where the federal government won't.</li><li>Tennessee lawmakers are proposing rules focused on child safety around AI, including age verification and parental controls.</li><li>Nvidia CEO Jensen Huang says AI safety is an engineering problem, not a legal one, and opposes new regulation.</li><li>Microsoft's Dr. Eric Horvitz publicly challenged Anthropic's views on AI consciousness, calling them based on flawed assumptions about human intelligence.</li><li>Anthropic is backed by investors including Google and Facebook.</li><li>45 percent of enterprises are scaling AI deployments while many keep legacy systems as a computing foundation for AI.</li><li>NVIDIA's Resiliency Extension enables fault-tolerant training on Amazon EKS with over 99 percent efficiency and second-scale GPU fault recovery.</li><li>Cloudflare now lets website owners block AI training while still allowing search crawlers, with plans to separate controls into Search, Training, and Agent categories.</li><li>A Hong Kong university launched a Cantonese-speaking AI virtual elderly patient built by Votee AI for social worker training.</li><li>Dolby upgraded its OptiView sports platform with AI-driven fan analytics and expanded ad solutions, including deployments with the Seattle Seahawks.</li></ul>}
New York pushes broader AI regulation ahead of first law
New York Governor Kathy Hochul and Democratic lawmakers say the state must expand AI regulation even before its first AI safety law takes effect. They cite a lack of action from the Trump administration and growing fears about powerful AI. Hochul said New York will act where Washington refuses. Lawmakers are also debating rules for data centers and chatbots for minors.
Congress faces pressure to act on AI regulation
Congress is feeling new pressure to regulate artificial intelligence after recent events. Republican Speaker Mike Johnson says AI should be handled with common sense and self-regulation. Democratic leader Hakeem Jeffries calls for immediate action to protect public safety. Experts disagree on whether states or the federal government should lead on AI rules.
Nvidia CEO says AI safety does not need new laws
Nvidia founder and CEO Jensen Huang says AI safety is an engineering problem, not a legal one. He argues the free market and company choices are enough to keep AI products safe. Huang believes companies should slow down if they are not confident in a product. He spoke at Salesforce's Dreamforce conference on September 15, 2026.
Microsoft AI researcher challenges Anthropic on AI consciousness
Microsoft's top AI researcher, Dr. Eric Horvitz, criticized Anthropic's idea of AI consciousness. He argues that Anthropic's view is based on a flawed understanding of human intelligence. The debate comes as the AI industry struggles to find the best way to make AI safer. Anthropic is a major AI company backed by investors like Google and Facebook.
AI virtual patients aim to improve medical training
A Hong Kong university launched a Cantonese-speaking AI virtual elderly patient to train social workers. The tool was built by Votee AI and the Nethersole School of Nursing at CUHK. It lets students practice real-time counseling scenarios and won a gold medal in Geneva. The system runs on university servers to protect patient data and support local language needs.
Dolby expands AI sports platform with fan analytics and ads
Dolby Laboratories upgraded its Dolby OptiView live sports platform with AI-driven fan engagement analytics, Dolby Vision support, and expanded ad solutions. The updates include deployments with organizations such as the Seattle Seahawks. The changes tie Dolby's technology more closely to how games are produced, personalized, and monetized.
Tennessee lawmakers weigh new AI rules to protect children
Tennessee lawmakers are considering new proposals to keep children safe from artificial intelligence. A bill requires the Tennessee Advisory Commission on Intergovernmental Regulations to study AI and child safety and produce a report. Experts and officials discussed risks, parental controls, age verification, and possible legal obligations for AI companies at a recent hearing.
Companies hold onto legacy IT as AI investment grows
Enterprises are keeping legacy systems such as mainframes longer as AI investment rises. Ensono's report says 45 percent of firms are scaling AI deployments and 44 percent are targeting high-impact use cases. Many organizations now view legacy systems as a key data and computing foundation for AI rather than replacing them outright.
NVRx enables fault tolerant training on Amazon EKS
A new post explains how to integrate NVIDIA Resiliency Extension (NVRx) into PyTorch FSDP training on Amazon EKS. The setup overlaps checkpoint I/O with training and can recover from GPU faults in seconds. H100 benchmarks at 2 to 8 nodes showed over 99 percent training efficiency and second-scale recovery.
Cloudflare adds separate controls for AI training and search
Cloudflare introduced a "Disallow AI Training" setting that lets website owners block AI training while still allowing crawlers for search indexing. The feature targets mixed-use crawlers that support both search and AI data collection. Cloudflare also plans to separate its controls into Search, Training, and Agent categories.
Why AI Customer Service Frustrates Customers and How Companies Can Improve It
Automated customer service systems often frustrate people by repeating irrelevant information and failing to solve real problems. The author describes being stuck on hold for 35 minutes and unable to reach a human for help with a credit card breach. She contrasts this with stores like a Northampton bra shop where expert staff provide personalized help. Companies like Fidelity show that good service is possible by using bots only to verify identity and then connecting customers quickly to humans. The article argues that businesses must respect customers and avoid treating everyone the same.
Sources
- New York wants to go bigger on AI regulation — before its first law even takes effect
- Congress is under pressure to act on AI — here's what that could look like
- We don't need AI regulation — leave safety to us, Nvidia's Jensen Huang says
- Exclusive: Microsoft AI chief blasts Anthropic's notion of AI consciousness
- Can AI patients improve medical training?
- How Investors May Respond To Dolby (DLB) AI Sports Platform Update
- Tennessee lawmakers to consider new AI rules to protect children
- Enterprises are sweating legacy IT assets as AI investment grows
- Fault tolerant distributed training on Amazon EKS using NVRx
- Cloudflare separates AI training opt-outs from search crawling for publishers
- Why AI Makes Customer Service Worse, And What Companies Can Do To Fix It
Comments
Please log in to post a comment.