Table of Contents
The U.S. Department of War has threatened to designate artificial intelligence firm Anthropic as a supply chain risk, a move that would effectively ban the company’s technology from military use and force contractors to strip its models from their systems. The escalation follows reports that Anthropic’s "Claude" model was utilized during a classified raid, prompting a clash between the AI company’s strict safety terms and the military's requirement for unrestricted operational use.
Key Points
- Pentagon Blacklist Threat: Defense officials are reviewing ties with Anthropic and considering a "supply chain risk" designation—a penalty typically reserved for foreign adversaries—due to the company's refusal to align with military use-of-force standards.
- Alibaba Releases Qwen 3.5: The Chinese tech giant launched a new 397-billion parameter model that rivals Western competitors like Gemini 3 Pro while significantly undercutting pricing.
- Hollywood vs. ByteDance: Major entertainment guilds and the Motion Picture Association have condemned ByteDance’s new "Seed Dance 2.0" video model for mass copyright infringement.
- Apple’s March Event: Apple has scheduled a multi-city event for March 4, with analysts expecting the debut of M5-powered MacBooks but potential delays for the highly anticipated AI-integrated Siri.
Pentagon Conflict with Anthropic Escalates
The rift between the Department of War and AI developer Anthropic has widened significantly following a Wall Street Journal report that the company’s Claude model was utilized during a raid targeting Venezuelan leadership. While specific operational details remain classified, reports indicate that kinetic fire occurred, violating Anthropic’s terms of service which strictly prohibit the use of its technology for violence, weapons development, or surveillance.
According to reports, an executive at Palantir—which serves Claude to the military—alerted Pentagon officials after Anthropic inquired about the specific usage of their software. The inquiry was perceived by defense officials as a challenge to their operational authority. In response, the Department is now moving to enforce an "all lawful use" standard, rejecting the ability of private tech companies to impose ethical restrictions on military operations.
Chief Pentagon spokesperson Shawn Parnell issued a stark warning regarding the future of the partnership:
"The Department of War's relationship with Anthropic is being reviewed. Our nation requires that our partners be willing to help our war fighters win in any fight. Ultimately, this is about our troops and the safety of the American people."
Senior officials have indicated that Secretary of War Pete Hegseth is close to cutting ties completely. Designating Anthropic as a supply chain risk would force all Department vendors to certify they do not use Anthropic models, a logistical hurdle comparable to the 2019 ban on Huawei telecommunications equipment.
A senior official bluntly described the potential fallout to Axios:
"It will be enormous pain in the ass to disentangle and we are going to make sure they pay a price for forcing our hand like this."
While OpenAI, Google, and xAI also hold Pentagon contracts, defense officials are pressuring all vendors to agree to the removal of restrictions on classified use. Currently, negotiations with major firms regarding these terms remain fluid.
Chinese Tech Advances Spark Market and Legal Fears
Alibaba Challenges Western Pricing Models
In a move that undercuts Western AI pricing structures, Alibaba has released Quen 3.5 Plus. The model features a Mixture-of-Experts (MoE) architecture with 397 billion parameters and a one-million token context window. While Alibaba refrained from comparing the model directly to the latest state-of-the-art U.S. models, benchmark results suggest it performs comparably to Google's Gemini 3 Pro in multimodal reasoning.
The most significant disruption lies in the pricing strategy. Alibaba Cloud is offering the model at approximately $1.20 per million input tokens and $7.20 per million output tokens, reinforcing a trend where high-level intelligence from Chinese laboratories is becoming available at aggressively diminishing price points.
Hollywood Condemns ByteDance Video Model
Concurrently, the entertainment industry has reacted with alarm to the release of "Seed Dance 2.0" by TikTok parent company ByteDance. The new video generation model has demonstrated the ability to create hyper-realistic depictions of actors, such as Tom Cruise and Brad Pitt, without consent. Unlike American AI firms, which have generally complied with requests to restrict the depiction of copyrighted characters, Chinese labs operate under different regulatory pressures regarding foreign IP.
The Motion Picture Association (MPA) issued a strong rebuke of the technology:
"In a single day, the Chinese AI service Seed Dance 2.0 has engaged in unauthorized use of US copyrighted works on a massive scale. By launching a service that operates without meaningful safeguards against infringement, ByteDance is disregarding well-established copyright law that protects the rights of creators and underpins millions of American jobs."
ByteDance acknowledged the criticism in a statement to CNBC, claiming they are taking steps to strengthen safeguards, though industry skepticism remains high regarding the effectiveness of self-regulation by foreign entities.
Apple Targets March for Hardware Refresh
Apple has announced a "special experience" event scheduled for March 4, taking place simultaneously in New York, Shanghai, and London. While the company usually reserves major announcements for its June developer conference, sources indicate this event will focus on a significant hardware refresh.
Bloomberg’s Mark Gurman reports that the event will likely feature new MacBook Pro and MacBook Air models powered by the new M5 chip. The M5 architecture, unveiled in October, boasts a 30% increase in unified memory bandwidth over its predecessor, a critical metric for on-device AI processing.
However, the highly anticipated overhaul of Siri with generative AI capabilities faces potential delays. While CEO Tim Cook previously promised a release this year, internal testing has reportedly revealed latency issues and difficulties in processing complex queries. Consequently, the full release of the new Siri is now expected to align with the iOS update at the Worldwide Developers Conference (WWDC) in June, rather than the March event.
As the Pentagon seeks to solidify its control over AI deployment terms, the outcome of the dispute with Anthropic will likely establish a precedent for how Silicon Valley navigates the tension between corporate ethics policies and national defense contracts.
