Anthropic Claude safety features clash with Pentagon contract

Key Takeaways

• The Pentagon has given Anthropic a Friday deadline to remove safety safeguards from its Claude AI for military use, threatening contract termination and a "supply chain risk" designation.
• Anthropic, which holds a $200 million contract with the Pentagon, refuses to allow its AI for autonomous weapons or mass surveillance.
• Elon Musk's xAI has reached an agreement to use its Grok AI in classified systems, potentially as an alternative to Anthropic's Claude.
• The Pentagon is also in talks with OpenAI and Google about using their AI models for classified purposes.
• Senator Bernie Sanders warns that AI could lead to 100 million U.S. job losses in the next decade and exacerbate wealth inequality.
• Apple has updated its SEED app to an AI-powered Sales Coach for its partners, featuring an "Ask" function for product information.
• At least six U.S. states have passed laws restricting AI use in health insurance, reflecting concerns over claims processing and coverage decisions.
• AI integration is increasing laptop costs due to the requirement for advanced hardware like Neural Processing Units (NPUs) and higher memory.
• A New York federal judge ruled that documents created using public AI tools without attorney direction are not protected by attorney-client privilege.
• The Trump administration plans to utilize the Pentagon's Project Maven AI to set pricing benchmarks for critical minerals, aiming to reduce reliance on China.

Pentagon Chief Demands AI Firm Drop Safeguards or Face Contract Loss

Defense Secretary Pete Hegseth is pressuring AI company Anthropic to remove safety features from its Claude software used by the Pentagon. Hegseth has given Anthropic CEO Dario Amodei a Friday deadline to comply. If Anthropic refuses, the Pentagon may cancel its contract, label the company a supply chain risk, or use the Defense Production Act. Anthropic has resisted pressure to remove safeguards related to autonomous weapons and mass surveillance. The Pentagon insists on using the AI for all lawful purposes without restrictions. This dispute highlights ongoing tensions over AI safety in military applications.

Hegseth Gives AI Firm Anthropic Friday Deadline on Safeguards

Defense Secretary Pete Hegseth has set a Friday deadline for AI company Anthropic to remove safeguards on its military-used products. The Pentagon wants Anthropic's AI tools, including Claude AI, available on classified networks without usual restrictions. This comes after reports that the Pentagon was considering this move. Reuters previously reported the Pentagon's interest in making AI tools from OpenAI and Anthropic available with fewer restrictions.

Pentagon Ultimatum: Anthropic Must Drop AI Guardrails by Friday

Defense Secretary Pete Hegseth has given AI company Anthropic until Friday to remove safety guardrails on its Claude AI model for military use. Hegseth threatened to label Anthropic a 'supply chain risk' if it doesn't comply. The Pentagon wants unrestricted access for 'all lawful use,' but Anthropic refuses to allow its AI for mass surveillance or weapons that fire without human control. Anthropic stated they are having productive conversations but have not agreed to remove their core safety concerns.

Hegseth Meets Anthropic CEO, Demands AI Model Changes by Friday

Defense Secretary Pete Hegseth met with Anthropic CEO Dario Amodei, demanding the company remove safeguards from its AI model Claude by Friday. Hegseth also threatened to blacklist the company. The Pentagon has a $200 million contract with Anthropic and wants unrestricted use of Claude. Anthropic has refused to drop concerns about AI-controlled weapons and mass domestic surveillance. Sources say the meeting was cordial, but Anthropic has no plans to change its stance.

Pentagon Warns Anthropic: Remove AI Limits or Lose Contract

Defense Secretary Pete Hegseth has given AI company Anthropic a Friday deadline to allow unrestricted military use of its technology. If Anthropic refuses, it could lose its government contract and face being designated a supply chain risk. Hegseth met with Anthropic CEO Dario Amodei to discuss the issue. Amodei maintained Anthropic's ethical concerns about fully autonomous weapons and domestic surveillance. The Pentagon insists on using the AI for all lawful purposes.

Hegseth Ultimatum: Anthropic Must Allow Unrestricted AI Use by Friday

Defense Secretary Pete Hegseth has given Anthropic CEO Dario Amodei a Friday deadline to allow unrestricted military use of its AI technology. Failure to comply could result in losing its government contract and being labeled a supply chain risk. The meeting was cordial, but Amodei stood firm on Anthropic's refusal to support fully autonomous weapons or domestic surveillance. The Pentagon wants to use the AI model Claude for all lawful purposes without limitations.

Pentagon Threatens Anthropic: Drop AI Safeguards or Face Penalties

Defense Secretary Pete Hegseth has given AI company Anthropic until Friday to remove its safety guardrails or face consequences, including contract termination. Hegseth demanded Anthropic abandon its AI safety measures for military use. Potential penalties include being declared a 'supply chain risk' or using the Defense Production Act. Anthropic confirmed a meeting occurred and stated they are having good-faith conversations about usage policies. Sources suggest Hegseth's demands may be a show of force, as Anthropic's safeguards have not hindered Pentagon operations.

US Pentagon Threatens AI Firm Anthropic with Friday Deadline

U.S. Defense Secretary Pete Hegseth has threatened to remove AI company Anthropic from the Pentagon's supply chain if it doesn't allow unrestricted use of its technology. Hegseth issued this threat during a meeting with Anthropic CEO Dario Amodei. Anthropic stated they are having good-faith conversations about their usage policy. A Pentagon official said Anthropic has until Friday to comply, or the Defense Production Act could be invoked, and the company labeled a supply chain risk. Anthropic's red lines include autonomous weapons and mass domestic surveillance.

Elon Musk's xAI Reaches Deal for Grok AI in Classified Systems

Elon Musk's xAI has reached an agreement with the Pentagon to use its Grok AI chatbot in classified systems. This deal could potentially replace Anthropic's Claude AI, which the Pentagon wants for 'all lawful purposes' without safeguards. Anthropic has refused this demand, citing concerns about mass surveillance and autonomous weapons. The Pentagon is also in talks with OpenAI and Google regarding their AI models for classified use. Replacing Anthropic is expected to be difficult.

Bernie Sanders Criticizes Tech Billionaires' AI Investments

Senator Bernie Sanders criticized tech billionaires like Elon Musk, Mark Zuckerberg, and Jeff Bezos for investing heavily in artificial intelligence. Sanders believes their primary motivation is to increase their own wealth and power, not to benefit humanity. He expressed concern that AI could displace workers and worsen income inequality. Sanders advocates for policies that ensure AI benefits society broadly, rather than concentrating wealth among a few.

US Faces 100 Million Job Losses to AI in Next Decade

The U.S. is projected to lose 100 million jobs over the next decade due to the rapid advancement of artificial intelligence and robotics, according to the AFL-CIO and Senator Bernie Sanders. Sanders warned that this job loss could undermine democracy and concentrate power among billionaires like Elon Musk and Jeff Bezos. He highlighted that AI development is moving much faster than the Industrial Revolution, potentially making millions of workers obsolete. The focus, Sanders argued, should be on who benefits from AI and who is harmed.

Apple Replaces SEED App with AI Sales Coach

Apple has quietly updated its SEED app to an AI-powered Sales Coach app for its sales partners. The update automatically converts SEED, preserving user credentials and access. The new app features an AI-powered 'Ask' function allowing sales teams to quickly get product information and compare features. It also adopts Apple's modern design language. Access to the Sales Coach app is restricted to verified Apple personnel and authorized partners.

States Push Back on AI Use in Health Insurance

States are increasingly enacting laws to limit how health insurers use artificial intelligence, creating political divisions. While President Trump encourages AI integration and opposes state regulations, at least six states have passed laws restricting AI in health insurance. Concerns about AI's use in claims processing and coverage decisions are widespread, with polls showing public skepticism across the political spectrum. Companies like Cigna and UnitedHealth Group are facing scrutiny over their AI practices.

AI Drives Up Laptop Costs and Performance

The increasing adoption of artificial intelligence is raising the cost of laptops as they require more advanced hardware. AI-friendly laptops now include Neural Processing Units (NPUs) and more memory for efficient local AI tasks. While this offers benefits like faster processing and better power management, it increases prices. Organizations need to invest in these AI-capable laptops to leverage new features and maintain productivity. Experts suggest minimum specs include a 40 TOPS NPU and 32GB of memory.

Pentagon's AI Push Creates Opportunities and Risks for Tech Firms

Defense Secretary Pete Hegseth is pushing the Pentagon to accelerate its use of artificial intelligence, seeking partnerships with tech firms to develop equipment like drone swarms and AI agents. This new strategy aims to bypass traditional, slower contractors. While this presents opportunities for AI labs, it also carries risks as the Pentagon demands rapid development and deployment of advanced AI technologies for warfighting.

AI Use in Legal Cases Raises Privilege Concerns

A New York federal judge ruled that documents created using a public AI tool are not protected by attorney-client privilege. In the case United States v. Heppner, the defendant used an AI platform to draft legal documents without his attorney's direction. The court found this compromised confidentiality, a key requirement for privilege. This decision highlights the risks of using consumer-grade AI in legal matters and stresses the importance of secure, attorney-supervised AI tools.

Rotary Club Discusses AI's Impact on Iowa's Economy

The Rotary Club of Des Moines recently discussed the role of artificial intelligence and automation in Iowa's economic future. Experts highlighted the need for workforce upskilling and reskilling to adapt to AI's impact. They emphasized that AI can create new job opportunities while also displacing some roles. The consensus was that embracing technological advancements and investing strategically in automation is crucial for Iowa's competitiveness and prosperity.

Quesma Launches Tool to Test AI and OpenTelemetry Performance

Quesma has released OTelBench, an open-source tool designed to benchmark OpenTelemetry infrastructure and AI performance in Site Reliability Engineering tasks. The tool evaluates the limits of observability systems and the efficiency of AI agents in handling complex SRE challenges. Initial tests show AI agents achieve less than 30% success in sophisticated observability tasks. OTelBench aims to provide data for teams to make informed decisions about adopting AI-driven observability.

Trump Admin to Use Pentagon AI for Critical Minerals Pricing

The Trump administration plans to use a Pentagon AI program, Project Maven, to help set prices for critical minerals. This initiative aims to create a trade bloc excluding China and reduce reliance on its supply chain. By analyzing global mineral markets, the AI will establish pricing benchmarks, enabling the U.S. and allies to negotiate better terms for essential minerals. Project Maven was initially developed for drone footage analysis.

Sources

• Humiliated Pentagon Pete Makes Desperate Last-Ditch Threat
• Hegseth gives Anthropic until Friday to back down on AI safeguards, Axios reports
• Exclusive: Hegseth gives Anthropic until Friday to back down on AI safeguards
• Pete Hegseth meets with Anthropic CEO Dario Amodei over disagreements about AI guardrails for military use
• AP report: Hegseth warns Anthropic to let the military use company’s AI tech as it sees fit
• Hegseth Warns Anthropic to Let the Military Use the Company’s AI Tech as It Sees Fit
• Pete Hegseth Demands Anthropic Drop AI Safety Guardrails
• US threatens Anthropic with deadline in dispute on AI safeguards
• Musk's xAI and Pentagon reach deal to use Grok in classified systems
• Bernie Sanders On AI Investment By Musk, Zuckerberg, Bezos: 'They're In It To Accumulate Even More Wealth And Power For Themselves'
• U.S. faces fast loss of 100 million jobs to Artificial Intelligence
• Apple Quietly Replaces SEED With AI-Powered Sales Coach App
• States want to limit AI in health insurance
• How AI is redefining price and performance in modern laptops
• For AI labs, Pete Hegseth’s Pentagon brings opportunities—and risks
• The Intersection of AI and Attorney-Client Privilege—A Cautionary Tale
• Rotary Club meeting focuses on AI’s role in Iowa’s economic future
• Quesma Releases OTelBench to Evaluate OpenTelemetry Infrastructure and AI Performance
• Exclusive-Trump eyes Pentagon AI program for trade block's minerals pricing, sources say