Anthropic CEO Dario Amodei tsis lees paub meej Pentagon qhov kev thov rau kev nkag mus rau nws cov AI systems rau hnub Thursday, hais nws “tsis tuaj yeem pom zoo rau nws.” Qhov no yog qhov nce zuj zus hauv kev tsis sib haum xeeb ntawm tus thawj coj AI ruaj khov thiab Asmeskas. Lub Tsev Haujlwm Saib Xyuas Kev Tiv Thaiv.
Hauv tsab ntawv tshaj tawm hnub Thursday, Amodei tau hais txog lub tuam txhab txoj haujlwm: “Anthropic nkag siab tias Lub Tsev Haujlwm Saib Xyuas Kev Tiv Thaiv, tsis yog cov tuam txhab ntiag tug, txiav txim siab ua tub rog. Txawm li cas los xij, nyob rau hauv cov xwm txheej nqaim, peb ntseeg tias AI thev naus laus zis tuaj yeem ua rau muaj kev cuam tshuam, es tsis txhob tiv thaiv, kev ywj pheej qhov tseem ceeb. Qee daim ntawv thov kuj tsuas yog dhau ntawm kev nyab xeeb thiab txhim khu kev qha ntawm cov thev naus laus zis tam sim no.”
Qhov kev tsis sib haum xeeb tam sim no nyob rau hauv ob qhov tseem ceeb: huab hwm coj saib xyuas ntawm American pej xeem thiab autonomous riam phom systems ua hauj lwm tsis muaj tib neeg tswj. The Pentagon maintains that it should be able to use Anthropic’s AI model for all lawful purposes, arguing that the technology’s use should not be dictated by a private company’s business considerations.
Amodei’s statement comes on the eve of a deadline set by the Pentagon. Defense Secretary Pete Hegseth has given Anthropic until 5:01 p.m.. Friday to either accept the military’s demands or face the consequences. To pressure the company into compliance, the Pentagon is employing a dual approach: either designating Anthropic a “kev pheej hmoo ntawm cov saw hlau”—a label typically reserved for foreign adversaries—or invoking the Defense Production Act to compel the company to cooperate with military directives on national security grounds. This act grants the president the authority to require companies to prioritize fulfilling national defense orders.
Amodei highlighted the inherent contradiction in these two threats. “Ntawm qhov one tes, they label us a security threat; on the other, they tout our Claude model as critical to national security.”
He emphasized that while the Department of Defense has the legitimate right to choose contractors aligned with its vision, “given the substantial value that Anthropic’s technology provides to our armed forces, we hope they will reconsider this decision.”
e, Anthropic is the only company with a frontier AI system that has been certified as classified-ready for military use. Txawm li cas los xij, reports indicate the Pentagon is also working to accelerate the preparation of Elon Musk’s company, xAI, for similar purposes.