Nvidia and 24 other companies sign open-weights letter as Washington weighs Chinese AI model ban  OpenAI, Anthropic, and Google absent from the list 83%

By Luke James33%

7/24/2026, 11:31:48 AM

BS Summary: This article contains 30 faulty reasoning types, including Availability Heuristic, Confirmation Bias, and Hasty Generalization, with Ambiguity (Equivocation) as the most egregious example at 26.3% saturation with 143 hits. Analysis detected 1,776 faulty-reasoning hits from 544 analyzed words, generating a BS Score of 74% and a BS Rank of 83% (3,927 of 21,887 articles). This article is worse (more manipulative) than 82.10% of the article peer group.

Jensen Huang joined X last month and used his first post Friday to promote Open Weights and American AI Leadership , a three-page policy letter published the same day and co-signed by 25 companies, including Nvidia, Microsoft, Meta, IBM, Dell Technologies, Palantir, and Hugging Face. 
(Image credit: Microsoft) Photonics and high-speed data movement is the next big AI bottleneck 
The data center cooling state of play 
Massive AI data center buildouts are squeezing energy supplies 
Ultra Ethernet: The data center interconnection of tomorrow 
The letter asks Washington to avoid what it calls "premature restrictions on downloadable AI models," and comes just four days after the Trump administration was reported to be reviving a push to ban Chinese models  though the document doesn't directly mention China, Moonshot AI, or DeepSeek. 
Notably missing from the co-signers are OpenAI, Anthropic, and Google. 
The 25 names break down into chipmakers, server vendors, cloud operators, enterprise software firms, security companies, and venture funds: Nvidia, Dell, Microsoft, IBM, Box, ServiceNow, CrowdStrike, Palantir, Telnyx, Replit, Perplexity, Andreessen Horowitz, Y Combinator, and Emergence Capital among them. 
The model developers on the list, Meta, Mistral, Black Forest Labs, Arcee AI, and Reflection, all publish weights already. 
Also present on the list of signatories is the Linux Foundation, which stewards the OpenMDW-1.1 license Nvidia used to release Nemotron 3 Ultra in June, a 550-billion-parameter model that Artificial Analysis scored at 47.7 on its intelligence index against 53.9 for Moonshot's Kimi K2.6. 
For my first post, I’m sharing a letter @NVIDIA signed on why open models matter.AI will transform every industry, power every company, and be built by every country.Open models strengthen safety and cybersecurity, accelerate innovation and diffusion, and enable sovereignty.… pic.twitter.com/t02bi51N4C July 24, 2026 
"The world needs both frontier closed models and frontier open models," Huang wrote in his X post. 
At Nvidia's CES 2026 press Q&A earlier this year, he put a figure on the shift, saying one in every four tokens generated today comes from an open model. 
Weights that anyone can download get served from enterprise clusters, regional clouds, and on-premises racks rather than a handful of hyperscaler API endpoints, and those buyers have no in-house TPU or Trainium program to buy instead. 
The letter's policy section asks for expanded compute access for startups and researchers, alongside public investment in shared datasets and evaluation frameworks. 
The letter goes on to urge policymakers not to treat distillation  the practice of training one model on another's outputs  as misappropriation, arguing that unlawful extraction from closed models should be handled through targeted legal frameworks, rather than broad limits on the technique. 
Treasury Secretary Scott Bessent said on Fox Business earlier this week that the administration would examine Chinese open-source models for intellectual property theft and could sanction the companies behind them, telling the program that officials had found watermarks from U.S. large language models in Chinese systems. 
Huang told Axios two days later that American firms should be allowed to use Chinese models , calling claims of Chinese backdoors a misconception. 
The distillation passage is the only part of the letter that doesn't concern open weights. 
Article reasoning-pattern comparisonThis article: 21.9%Luke James: 4.5%Tom's Hardware: 3.8%Confirmation Bias21.9%This article: 9.6%Luke James: 2.8%Tom's Hardware: 2.3%Anchoring Bias9.6%This article: 22.4%Luke James: 4.5%Tom's Hardware: 3.5%Availability Heuristic22.4%This article: 7.2%Luke James: 0.9%Tom's Hardware: 1.1%Representativeness Heuristic7.2%This article: 0.0%Luke James: 0.2%Tom's Hardware: 0.5%Hindsight Bias0.0%This article: 5.3%Luke James: 1.9%Tom's Hardware: 3.5%Overconfidence Bias5.3%This article: 0.0%Luke James: 6.4%Tom's Hardware: 9.2%Framing Effect0.0%This article: 0.0%Luke James: 0.4%Tom's Hardware: 0.9%Loss Aversion0.0%This article: 6.6%Luke James: 0.9%Tom's Hardware: 0.7%Status Quo Bias6.6%This article: 0.0%Luke James: 0.8%Tom's Hardware: 0.3%Sunk Cost Effect0.0%This article: 16.5%Luke James: 4.8%Tom's Hardware: 6.0%Optimism Bias16.5%This article: 0.0%Luke James: 1.6%Tom's Hardware: 2.0%Pessimism Bias0.0%This article: 16.4%Luke James: 5.8%Tom's Hardware: 6.5%Negativity Bias16.4%This article: 12.5%Luke James: 2.7%Tom's Hardware: 1.3%Self-Serving Bias12.5%This article: 0.0%Luke James: 0.6%Tom's Hardware: 0.6%Fundamental Attribution Error0.0%This article: 0.0%Luke James: 0.0%Tom's Hardware: 0.2%Actor-Observer Bias0.0%This article: 0.0%Luke James: 0.1%Tom's Hardware: 0.6%In-Group Bias0.0%This article: 0.0%Luke James: 0.0%Tom's Hardware: 0.3%Out-Group Homogeneity Bias0.0%This article: 8.1%Luke James: 1.8%Tom's Hardware: 2.8%Halo Effect8.1%This article: 0.0%Luke James: 0.0%Tom's Hardware: 0.0%Horn Effect0.0%This article: 0.0%Luke James: 0.0%Tom's Hardware: 0.0%Dunning-Kruger Effect0.0%This article: 13.1%Luke James: 2.8%Tom's Hardware: 2.0%Recency Bias13.1%This article: 8.3%Luke James: 0.9%Tom's Hardware: 0.6%Primacy Effect8.3%This article: 0.0%Luke James: 0.0%Tom's Hardware: 0.1%Blind-Spot Bias0.0%This article: 0.0%Luke James: 0.0%Tom's Hardware: 0.2%Ad Hominem0.0%This article: 0.0%Luke James: 0.1%Tom's Hardware: 0.2%Straw Man0.0%This article: 17.8%Luke James: 7.4%Tom's Hardware: 5.4%Appeal to Authority17.8%This article: 0.0%Luke James: 1.0%Tom's Hardware: 2.0%False Dilemma0.0%This article: 0.0%Luke James: 0.2%Tom's Hardware: 1.0%Slippery Slope0.0%This article: 0.0%Luke James: 0.1%Tom's Hardware: 0.1%Circular Reasoning0.0%This article: 20.2%Luke James: 4.4%Tom's Hardware: 6.1%Hasty Generalization20.2%This article: 11.4%Luke James: 0.6%Tom's Hardware: 0.3%Red Herring11.4%This article: 4.4%Luke James: 1.1%Tom's Hardware: 1.1%Bandwagon4.4%This article: 8.5%Luke James: 1.2%Tom's Hardware: 3.0%Appeal to Emotion8.5%This article: 8.3%Luke James: 1.1%Tom's Hardware: 0.8%Begging the Question8.3%This article: 10.3%Luke James: 3.0%Tom's Hardware: 3.5%Post Hoc (False Cause)10.3%This article: 4.4%Luke James: 0.1%Tom's Hardware: 0.1%Tu Quoque4.4%This article: 12.5%Luke James: 1.0%Tom's Hardware: 0.8%Burden of Proof12.5%This article: 0.0%Luke James: 0.1%Tom's Hardware: 0.3%Appeal to Nature0.0%This article: 7.2%Luke James: 0.7%Tom's Hardware: 0.6%Composition/Division7.2%This article: 8.1%Luke James: 1.1%Tom's Hardware: 1.6%Anecdotal8.1%This article: 3.5%Luke James: 0.1%Tom's Hardware: 0.0%No True Scotsman3.5%This article: 26.3%Luke James: 2.3%Tom's Hardware: 3.4%Ambiguity (Equivocation)26.3%This article: 0.0%Luke James: 0.0%Tom's Hardware: 0.0%Gambler’s Fallacy0.0%This article: 3.1%Luke James: 0.2%Tom's Hardware: 0.2%Middle Ground3.1%This article: 0.0%Luke James: 0.0%Tom's Hardware: 0.0%Personal Incredulity0.0%This article: 0.0%Luke James: 0.3%Tom's Hardware: 0.2%Special Pleading0.0%This article: 0.0%Luke James: 0.0%Tom's Hardware: 0.1%Genetic Fallacy0.0%This article: 0.0%Luke James: 4.0%Tom's Hardware: 2.4%Unattributed Quote0.0%This article: 8.1%Luke James: 0.8%Tom's Hardware: 0.9%Quote-first Misdirection8.1%This article: 8.3%Luke James: 2.8%Tom's Hardware: 7.1%Biased Writer Voice8.3%This article: 8.1%Luke James: 0.5%Tom's Hardware: 1.3%Indoctrination8.1%This article: 0.0%Luke James: 0.0%Tom's Hardware: 0.1%Politically Left Leaning Bias0.0%This article: 0.0%Luke James: 0.0%Tom's Hardware: 0.0%Politically Right Leaning Bias0.0%This article: 8.3%Luke James: 1.2%Tom's Hardware: 3.3%Attempt to Sell a Product or S…8.3%

544 words analyzed.

Speakers

3speakers38%attributed speech339writer words
Voice mapSelect a segment to jump to its words
Writer's voice • 24 words • 0.0% coverageJensen Huang • 45 words • 100.0% coverageWriter's voice • 14 words • 0.0% coverageWriter's voice • 7 words • 0.0% coverageWriter's voice • 9 words • 0.0% coverageWriter's voice • 8 words • 0.0% coverageWriter's voice • 47 words • 0.0% coverageWriter's voice • 10 words • 0.0% coverageWriter's voice • 39 words • 0.0% coverageWriter's voice • 19 words • 0.0% coverageLinux Foundation • 44 words • 0.0% coverageWriter's voice • 44 words • 100.0% coverageJensen Huang • 17 words • 0.0% coverageJensen Huang • 29 words • 0.0% coverageWriter's voice • 36 words • 0.0% coverageWriter's voice • 22 words • 0.0% coverageWriter's voice • 45 words • 0.0% coverageScott Bessent • 46 words • 0.0% coverageJensen Huang • 24 words • 0.0% coverageWriter's voice • 15 words • 0.0% coverage
Selected voice

Scott Bessent

100%flagged-word coverage
46 attributed words22% of attributed speech100% writer coverage
0%7.5%15.0%Quote-first Misdirection-13.0 ptsWriter: 13.0%Scott Bessent: 0.0%0.0%Indoctrination-13.0 ptsWriter: 13.0%Scott Bessent: 0.0%0.0%

Attribution is sentence-level. Pattern percentages are calculated only from words assigned to that voice.

Loading…
Loading…
Loading…

Analysis

Hover over highlighted words in the article to view the associated bias or fallacy analysis.