,i #aaith ,,congress #ast ,session ,h4 ,r4 #ciai ,,in ,,the ,,house ,,of ,,representatives ,june #aa1 #bjbe,mr4 ,la,hood 7for himself1 ,mr4 ,moolenaar1 ,mr4 ,gottheimer1 and ,mr4 ,krishnamoorthi7 introduced the following bill2 which was referred to the ,permanent ,select ,committee on ,intelligence ,a ,,bill ,to direct the ,director of the ,national ,security ,agency to develop strategies to secure artificial intelligence related technologies4 #a4 ,short title ,this ,act may be cited as the ,advanced ,,ai ,security ,readiness ,act4 #b4 ,,ai security playbook 7a7 ,requirement ,the ,director of the ,national ,security ,agency1 acting through the ,artificial ,intelligence ,security ,center 7or successor office71 shall develop strategies 7in this section referred to as the ,,ai ,security ,playbook7 to defend covered ,,ai technologies from technology theft by threat actors4 7b7 ,elements ,the ,,ai ,security ,playbook under subsection 7a7 shall include the following3 7#a7 ,identification of potential vulnerabilities in advanced ,,ai data centers and among advanced ,,ai developers capable of producing covered ,,ai technologies1 with a focus on cybersecurity risks and other security challenges that are unique to protecting covered ,,ai technologies and critical components of such technologies 7such as threat vectors that do not typically arise1 or are less severe1 in the context of conventional information technology systems74 7#b7 ,identification of components or information that1 if accessed by threat actors1 would meaningfully contribute to progress made by the actor with respect to developing covered ,,ai technologies1 including with respect to-- 7,a7 ,,ai models and key components of such models2 7,b7 core insights relating to the development of advanced ,,ai systems1 including with respect to training such systems1 the inferences made by such systems1 and the engineering of such systems2 and 7,c7 other related information4 7#c7 ,strategies to detect1 prevent1 and respond to cyber threats by threat actors targeting covered ,,ai technologies4 7#d7 ,identification of the levels of security1 if any1 that would require substantial involvement by the ,united ,states ,government in the development or oversight of highly advanced ,,ai systems4 7#e7 ,analysis of how the ,united ,states ,government would be involved to achieve the levels of security identified in paragraph 7#d71 including a description of a hypothetical initiative to build covered ,,ai technology systems in a highly secure governmental environment1 considering1 at a minimum1 cybersecurity protocols1 provisions to protect model weights1 efforts to mitigate insider threats 7including personnel vetting and security clearance adjudication processes71 access control procedures1 counterintelligence and anti-espionage measures1 contingency and emergency response plans1 and other strategies that would be used to reduce threats of technology theft by threat actors4 7c7 ,form ,the ,,ai ,security ,playbook under subsection 7a7 shall include-- 7#a7 detailed methodologies and intelligence assessments1 which may be contained in a classified annex2 and 7#b7 an unclassified portion with general guidelines and best practices suitable for dissemination to relevant individuals1 including in the private sector4 7d7 ,engagement 7#a7 ,in general ,in developing the ,,ai ,security ,playbook under subsection 7a71 the ,director shall-- 7,a7 engage with prominent ,,ai developers and researchers1 as determined by the ,director1 to assess and anticipate the capabilities of highly advanced ,,ai systems relevant to national security1 including by-- 7i7 conducting a comprehensive review of industry documents pertaining to the security of ,,ai systems with respect to preparedness frameworks1 scaling policies1 risk management frameworks1 and other matters2 7ii7 conducting interviews with subject matter experts2 7iii7 hosting roundtable discussions and expert panels2 and 7iv7 visiting facilities used to develop ,,ai2 and 7,b7 to leverage existing expertise and research1 collaborate with a federally funded research and development center that has conducted research on strategies to secure ,,ai models from nation-state actors and other highly resourced actors4 7#b7 ,nonapplicability of ,,faca ,none of the activities described in this subsection shall be construed to establish or use an advisory committee subject to chapter #aj of title #e1 ,united ,states ,code4 7e7 ,reports 7#a7 ,initial report ,not later than #ij days after the date of the enactment of this ,act1 the ,director shall submit to the appropriate congressional committees a report on the ,,ai ,security ,playbook under subsection 7a71 including a summary of progress on the development of ,playbook1 an outline of remaining sections1 and any relevant insights about ,,ai security4 7#b7 ,final report ,not later than #bgj days after the date of enactment of this ,act1 the ,director shall submit to the appropriate congressional committees a report on the ,playbook4 7#c7 ,form ,the report submitted under paragraph 7#b7-- 7,a7 shall include-- 7i7 an unclassified version suitable for dissemination to relevant individuals1 including in the private sector2 and 7ii7 a publicly available version2 and 7,b7 may include a classified annex4 7f7 ,rule of construction ,nothing in subsection 7b77#d7 shall be construed to authorize or require any regulatory or enforcement action by the ,united ,states ,government4 7g7 ,definitions ,in this section3 7#a7 ,the term appropriate congressional committees means the ,permanent ,select ,committee on ,intelligence of the ,house of ,representatives and the ,select ,committee on ,intelligence of the ,senate4 7#b7 ,the terms artificial intelligence and ,,ai have the meaning given the term artificial intelligence in section #bch7;g7 of the ,john ,s4 ,mc,cain ,national ,defense ,authorization ,act for ,fiscal ,year #bjai 7,public ,law #aae-bcb2 #aj ,u4,s4,c4 note prec4 #djfa74 7#c7 ,the term covered ,,ai technologies means advanced ,,ai 7whether developed by the private sector1 the ,united ,states ,government1 or a public-private partnership7 with critical capabilities that the ,director determines would pose a grave national security threat if acquired or stolen by threat actors1 such as ,,ai systems that match or exceed human expert performance in relating to chemical1 biological1 radiological1 and nuclear matters1 cyber offense1 model autonomy1 persuasion1 research and development1 and self-improvement4 7#d7 ,the term technology theft means any unauthorized acquisition1 replication1 or appropriation of covered ,,ai technologies or components of such technologies1 including models1 model weights1 architectures1 or core algorithmic insights1 through any means1 such as cyber attacks1 insider threats1 and side-channel attacks1 or exploitation of public interfaces4 7#e7 ,the term threat actors means nation-state actors and other highly resourced actors capable of technology theft4