China is expanding AI data centers in rural Inner Mongolia, where cheap energy, land and labor, alongside government support, attract major investors including Huawei, Alibaba and ByteDance. Farmers and other residents are discussing the projects and their effects. Chinese AI models are estimated to trail U.S. competitors by about six months but are narrowing the gap; experts argue access to power may matter more than chips in the competition. By contrast, U.S. data center projects face resistance. Former White House AI adviser Ben Buchanan says the United States and other open societies must lead the AI race against closed societies such as China to avoid worst-case outcomes. He helped develop federal AI policy, including GPU export controls to China, and says a January 2023 GPT-4 demonstration helped convince national security adviser Jake Sullivan of AI’s importance.
AI is also prompting debate about human judgment and capability. A November 12, 2026 article warns that AI could speed up military decisions and affect crisis management, strategic deception and information control. It argues political leaders must set ethical limits and never delegate decisions about war to machines; future nuclear command and battle-management systems are expected to rely heavily on AI, and U.S. nuclear-use authorization protocols may need review. In health care, leaders weigh AI’s diagnostic potential against the risk that clinicians lose skills. A Harvard-led study found an advanced AI agent outperformed doctors on demanding diagnosis tests, while OpenAI’s o1-preview handled emergency triage and clinical cases at the level of or beyond well-trained physicians. Some leaders see AI as a tool doctors must learn to use; others stress preserving core skills.
Companies are exploring AI’s role in fashion as agentic commerce lets shoppers interact with brands through chatbots and virtual assistants. Fashion businesses are considering how to protect brand identity: some experts warn AI could diminish human touch and creativity, while others see potential to improve decisions and customer service. For neurodivergent job seekers, experts recommend using AI interview tools for low-pressure practice without masking differences. Advice includes preparing concise answers, requesting interview schedules and interviewer names, and learning to redirect questions about legally protected personal matters. The aim, one expert says, should be finding supportive workplaces—not hiding who candidates are.
On the tools front, Liquid AI has released LiquidCodec, a streaming audio codec for its LFM3-Audio-2B and LFM3-Omni-2B models. The 74-million-parameter codec compresses 24 kHz audio into 12 token streams. On one CPU thread of an Apple M4 Max, it encoded at roughly 361 times real time and decoded at 87 times real time. Liquid AI reports leading ViSQOL, mel distance and TTS UTMOSv2 results across eight test sets, while NanoCodec scored better on PESQ in seven. The model weights are available on Hugging Face, with Transformers support.
Markets show investor caution around AI: Nvidia shares have faced pressure amid concerns about the chipmaker’s leadership and what that could mean for other AI stocks. European markets were mostly higher after Asian markets closed lower, U.S. futures were mixed and oil prices rose. Separately, a trial lawyer makes the case for relying less on the internet and more on paper, handwritten mail and telephones. The argument draws on a lawsuit involving a California construction-supply company that said a computer system adopted in the late 1970s cost it more than $20 million in profits. Pacific Bell disputed the claim, citing warnings about the system’s risks and other causes. After a three-and-a-half-month trial, the jury awarded $3.8 million in compensatory damages and $500,000 in punitive damages.
Key Takeaways
- Inner Mongolia’s low energy, land and labor costs are drawing Huawei, Alibaba and ByteDance to AI data centers; experts say power access could prove more decisive than chips.
- Chinese AI models are estimated to be about six months behind U.S. competitors but are narrowing the gap, while U.S. data center projects face resistance.
- Former U.S. AI adviser Ben Buchanan argues open societies should lead the AI race; he helped shape federal policy, including GPU export controls to China, and says a GPT-4 demonstration helped persuade Jake Sullivan of AI’s importance.
- An analysis of AI and nuclear strategy says AI may accelerate military decision-making, but political leaders must retain control over decisions about war.
- A Harvard-led study found an advanced AI agent outperformed doctors on demanding diagnosis tests; OpenAI’s o1-preview handled some clinical cases and emergency triage at or beyond the level of well-trained physicians.
- Experts recommend AI interview practice for neurodivergent candidates while emphasizing that candidates should not be coached to hide their differences.
- Fashion brands are assessing AI-driven chatbot shopping, balancing potential service and decision-making benefits against concerns about creativity and human connection.
- Liquid AI’s 74-million-parameter LiquidCodec runs with LFM3 models, supports Transformers and is available on Hugging Face; on an Apple M4 Max CPU thread it encoded 24 kHz audio at about 361× real time.
- Nvidia shares face pressure amid leadership concerns as global markets diverge; European markets were mostly higher, Asian markets fell, U.S. futures were mixed and oil rose.
- A lawyer advocating a return to analog tools cites a computer-system lawsuit in which a company claimed more than $20 million in lost profits; the jury awarded $3.8 million in compensatory and $500,000 in punitive damages.
Inner Mongolian Farmers Discuss China’s Expanding AI Data Centers
China is building large data centers in rural Inner Mongolia to support its growing AI industry. Anna Coren traveled to the region and spoke with local residents about the projects. The article focuses on how farmers view the construction and its effects.
China Builds AI Data Centers as US Projects Face Resistance
China is building large data centers in Inner Mongolia to support its technology ambitions. Cheap energy, low land costs, and low labor costs make the region attractive, and the government is encouraging development. Major companies including Huawei, Alibaba, and Byte Dance are investing there. Chinese AI models are about six months behind US competitors but are narrowing the gap, and experts say access to power could matter more than chips in the race for AI leadership.
A Trial Lawyer Argues for Going Analog to Push Back Against AI
A trial lawyer argues that society should reduce its reliance on the internet and return to tools such as paper records, handwritten mail, and telephones. The writer recalls defending Pacific Bell in a lawsuit over a computer system that a California construction supply company adopted in the late 1970s. The company claimed it lost more than $20 million in profits, while Pacific Bell argued that the system’s risks had been explained and that a construction downturn and heavy rain caused the losses. After a three-and-a-half-month trial, the jury awarded $3.8 million in compensatory damages and $500,000 in punitive damages. The writer uses the case to support a broader argument for analog ways of working and living.
Experts Share AI Interview Practice Tips for Neurodivergent Job Seekers
Research cited in the article estimates that 10% to 30% of US college students show neurodivergent traits, and that interviews can screen candidates based on eye contact, small talk, and body language rather than job skills. ResumeCoach spoke with career experts Dahlia Ashford and Chelsea Jay about using AI interview tools to practice in a low-pressure setting without masking who candidates are. Ashford recommends preparing for common questions, requesting the schedule and interviewer list, and practicing concise answers and natural eye contact. She also advises learning how to redirect questions about age, disability, religion, or family status, which are legally off-limits. Jay says AI should not teach candidates to hide their differences, but help them find workplaces that accept and support them.
AI Could Reshape Nuclear Deterrence and Military Decision Making
Published on November 12, 2026, the article examines how artificial intelligence could change nuclear deterrence and military strategy. It says AI may give states an edge in decision-making, strategic deception, crisis management, and control of information, while making conflicts move faster. The author argues that political leaders must set ethical limits and must never hand decisions about war to machines. Future nuclear command, control, and communications systems, along with battle management and command systems, are expected to depend heavily on AI. The article also says the United States will need to revisit nuclear-use authorization protocols, with AI potentially offering decision-makers additional options during a crisis.
Nvidia concerns weigh on AI stocks as global markets turn mixed
European markets were mostly higher after Asian markets ended lower, while U.S. futures were mixed and oil prices rose. Investors are concerned about Nvidia’s leadership troubles and what they could mean for the chipmaker and other AI stocks. Nvidia’s shares have faced pressure in recent days, and some analysts worry about the company’s future. Investors remain cautious as they wait to see how the situation develops.
Liquid AI releases fast streaming audio codec with 74 million parameters
Liquid AI released LiquidCodec, a streaming audio codec used with its LFM3-Audio-2B and LFM3-Omni-2B models. It compresses 24 kHz audio into 12 token streams and has about 74 million parameters. In tests on one CPU thread of an Apple M4 Max, it encoded audio at about 361 times real time and decoded it at about 87 times real time. Liquid AI reported the best ViSQOL, mel distance, and TTS UTMOSv2 scores across eight test sets, though NanoCodec had better PESQ scores on seven of the eight sets. The model weights are available on Hugging Face, and developers can use it with Transformers.
Ben Buchanan makes the case for U.S. leadership in the AI race
Ben Buchanan, President Biden’s former senior AI advisor, discussed AI competition, diplomacy, governance, and cybersecurity in a November 12, 2026 interview. He helped write the first federal AI policy, including export controls on GPUs sent to China, and took part in the 2024 Geneva Summit on AI risks. In a new book co-written with Tantum Collins, “The Bitter Struggle: Superintelligence, Superpowers, and the Fate of the World,” Buchanan argues that open societies such as the United States must win the AI race against closed societies such as China to avoid worst-case outcomes. He also described how seeing GPT-4 demonstrated to national security advisor Jake Sullivan in January 2023 helped convince U.S. policymakers that AI was important.
Fashion brands weigh AI’s impact as shopping habits change
Fashion companies are studying how artificial intelligence could affect their businesses as agentic commerce grows. In this approach, consumers interact with brands through chatbots and virtual assistants. Companies are also focused on protecting their brand identity as the industry changes. Some experts warn that relying on AI could reduce human touch and creativity, while others say it can help businesses make better decisions and improve customer service. The future of fashion remains uncertain, but AI is expected to play a growing role.
Health leaders weigh AI’s benefits against the risk of doctors losing skills
Health leaders are debating whether AI will improve doctors’ work or weaken their skills as it takes on more complex tasks, including diagnosing difficult cases. Duke Health’s Dr. Eric Poon raised concerns about deskilling and how medical schools should prepare future clinicians to work responsibly with AI while keeping essential skills. A Harvard-led study found that an advanced AI agent outperformed doctors on demanding diagnosis tests, and OpenAI’s o1-preview handled emergency triage and clinical cases at the level of or beyond well-trained physicians. The MetroHealth System’s Dr. Nabil Chehade said earlier fears about newer surgical technology changed as the tools advanced, while Northwell Health’s Theo Zanos argued that doctors who use AI may replace those who do not. Zanos also noted that radiologists remain in short supply, despite past predictions that AI would replace them.
Sources
- Inner Mongolian farmers talk about AI data centers powering China's tech boom
- While many Americans resist, China embraces AI data center boom
- A Trial Lawyer’s Case for Going Analog to Defeat AI
- Two Career Experts Explain How AI Practice Can Help Neurodivergent Candidates Navigate Interview Bias
- Nuclear Deterrence and the Significance of AI
- World shares are mixed as Nvidia's troubles weigh on AI stocks
- Liquid AI Releases LiquidCodec, a 74M-Parameter Streaming Audio Codec
- Ben Buchanan Wants the U.S. to Win on AI. But Is It a Race or a Trap?
- Artificial intelligence, new consumers and future of fashion business
- Will AI Make Doctors Better—or Make Them Less Skilled?
Comments
Please log in to post a comment.