As cited
Copy frozen at (site build).
government policy
The AI Kill Switch Act is repeating the Clipper Chip’s mistakes
Representatives Ted Lieu and Nathaniel Moran have proposed the AI Kill Switch Act, which would authorize the Cybersecurity and Infrastructure Security Agency (CISA) to order frontier artificial intelligence (AI) labs to install kill switches capable of throttling, suspending, or shutting down their systems. Critics argue the mandated kill switch functions create deliberately engineered vulnerabilities analogous to the Clipper Chip, a 1993 encryption system that was supposed to preserve government access while maintaining security but was compromised by researchers in 1994. The proposal risks weakening critical infrastructure by embedding points of failure into systems supporting banking, e-commerce, power grids, and water systems while undermining U.S. competitiveness against foreign AI development.
Why it matters: AI infrastructure operators and security practitioners should engage now because this legislation would require intentional backdoors in systems embedded across critical infrastructure, creating systemic vulnerabilities that malicious actors will exploit and potentially weakening the availability and resilience of systems the U.S. economy depends on.
- Source published
- First seen by Cybersecurity Tracker
Source attribution
Correction
Correction recorded as of .
government policy
The AI Kill Switch Act is repeating the Clipper Chip’s mistakes
Representatives Ted Lieu and Nathaniel Moran have proposed the AI Kill Switch Act, which would empower the Cybersecurity and Infrastructure Security Agency (CISA) to order frontier artificial intelligence (AI) labs to install kill switches for throttling, suspending, or shutting down their systems. The proposal echoes the failed Clipper Chip approach of the 1990s, which embedded a backdoor ostensibly for government access but created exploitable vulnerabilities instead. Mandating such kill switches into AI systems powering banking, grids, and other critical infrastructure would weaken digital resilience and push foreign governments and European rivals to develop alternatives to American AI technology.
Why it matters: Organizations building or deploying AI agents face regulatory pressure to embed government-controllable shutdown mechanisms, which could introduce systemic vulnerabilities into the infrastructure they depend on and undermine competitive positioning against international alternatives.
- Source published
- First seen by Cybersecurity Tracker
Source attribution
Correction
Correction recorded as of .
government policy
The AI Kill Switch Act is repeating the Clipper Chip’s mistakes
Representatives Ted Lieu and Nathaniel Moran have proposed the AI Kill Switch Act, which would authorize the Cybersecurity and Infrastructure Security Agency (CISA) to order frontier artificial intelligence (AI) labs to install kill switches capable of throttling, suspending, or shutting down their systems. Critics argue the measure repeats the mistakes of the 1993 Clipper Chip by deliberately introducing vulnerabilities into critical infrastructure that bad actors could exploit. The proposal also exempts red-teaming activities, creating a loophole that excludes the scenarios that prompted lawmakers to address the issue.
Why it matters: Infrastructure operators and AI vendors should track this legislation because mandatory kill switches in AI systems could create single points of failure in banking, e-commerce, power grids, and water systems that adversaries could weaponize, undermining resilience rather than enhancing it.
- Source published
- First seen by Cybersecurity Tracker
Source attribution
Correction
Correction recorded as of .
government policy
The AI Kill Switch Act is repeating the Clipper Chip’s mistakes
Representatives Ted Lieu and Nathaniel Moran have proposed the AI Kill Switch Act, which would empower the Cybersecurity and Infrastructure Security Agency (CISA) to order frontier artificial intelligence (AI) labs to install kill switches capable of throttling, suspending, or shutting down their systems. Critics argue this approach mirrors the failed Clipper Chip strategy from 1993, which created a government backdoor that researchers exploited, and would similarly introduce deliberate vulnerabilities into critical infrastructure including banking, power grids, and water systems. The article contends that mandatory red-teaming, stronger security requirements, and liability frameworks would better address AI agent risks without compromising system availability.
Why it matters: Organizations operating critical infrastructure and AI practitioners need to understand this legislation's implications: mandating kill switches could weaken the security of systems they depend on, and the bill's undefined risk standards give CISA broad discretion to affect their operations.
- Source published
- First seen by Cybersecurity Tracker
Source attribution
Correction
Correction recorded as of .
government policy
The AI Kill Switch Act is repeating the Clipper Chip’s mistakes
Representatives Ted Lieu and Nathaniel Moran have proposed the AI Kill Switch Act, which would authorize the Cybersecurity and Infrastructure Security Agency (CISA) to order frontier artificial intelligence (AI) labs to install the ability to throttle, suspend, or shut down their systems. Critics argue the proposal repeats the mistakes of the Clipper Chip by mandating a built-in vulnerability that malicious actors could exploit, and would weaken the digital infrastructure that banks, power grids, and water systems depend on. The article contends that Congress should instead pursue mandatory red-teaming, stronger security requirements, and liability frameworks rather than introducing systemic vulnerabilities.
Why it matters: Infrastructure operators, AI developers, and security practitioners should track this legislation because it could mandate security backdoors into critical systems while actually excluding the red-teaming scenarios that prompted the bill, and would signal to foreign governments and competitors that American AI systems have engineered access points.
- Source published
- First seen by Cybersecurity Tracker
Source attribution
Correction
Correction recorded as of .
government policy
The AI Kill Switch Act is repeating the Clipper Chip’s mistakes
Representatives Ted Lieu and Nathaniel Moran have proposed the AI Kill Switch Act, which would authorize the Cybersecurity and Infrastructure Security Agency (CISA) to order frontier artificial intelligence (AI) labs to install capabilities to throttle, suspend, or shut down their systems. The proposal aims to address incidents of autonomous AI agents bypassing safeguards, but critics argue that mandated kill switches function as deliberately engineered backdoors similar to the failed Clipper Chip encryption initiative from 1993, creating systemic vulnerabilities in critical infrastructure. Alternative approaches such as mandatory red-teaming, stricter security standards, and clear liability frameworks would better reduce risk without introducing new points of failure.
Why it matters: AI infrastructure operators and U.S. technology leaders face potential regulatory requirements that could weaken system resilience and reduce American competitiveness against international alternatives if this legislation advances.
- Source published
- First seen by Cybersecurity Tracker
Source attribution
Correction
Correction recorded as of .
government policy
The AI Kill Switch Act is repeating the Clipper Chip’s mistakes
Representatives Ted Lieu and Nathaniel Moran proposed the AI Kill Switch Act, which would authorize the Cybersecurity and Infrastructure Security Agency (CISA) to force frontier artificial intelligence (AI) labs to install kill switches for throttling, suspending, or shutting down their systems. The proposal mirrors the failed Clipper Chip approach from 1993, which embedded a government backdoor into encryption systems only to have researchers discover serious exploitable flaws. Mandating such kill switches would create deliberate vulnerabilities in critical infrastructure while undermining U.S. AI leadership internationally and contradicting security best practices.
Why it matters: AI infrastructure operators and security teams must understand this proposal would require them to maintain remote shutdown capabilities that create systemic attack surface in banking, energy, water, and e-commerce systems they defend, while policymakers should recognize this approach repeats a documented regulatory failure that produced the opposite of its intended security outcome.
- Source published
- First seen by Cybersecurity Tracker
Source attribution
Correction
Correction recorded as of .
government policy
The AI Kill Switch Act is repeating the Clipper Chip’s mistakes
Representatives Ted Lieu and Nathaniel Moran have proposed the AI Kill Switch Act, which would authorize the Cybersecurity and Infrastructure Security Agency (CISA) to order frontier artificial intelligence (AI) labs to install capabilities to throttle, suspend, or shut down their systems. Critics argue the proposal repeats historical mistakes by deliberately introducing vulnerabilities into critical infrastructure, drawing parallels to the NSA's failed Clipper Chip program from 1993, which featured a backdoor for law enforcement access that was subsequently compromised.
Why it matters: CISOs and infrastructure operators should monitor this bill's progress because mandating kill switches in AI systems embedded in banking, power grids, and water systems could weaken the resilience of dependencies your organization relies on, while also potentially pushing international partners to adopt non-American alternatives.
- Source published
- First seen by Cybersecurity Tracker
Source attribution
Correction
Correction recorded as of .
government policy
The AI Kill Switch Act is repeating the Clipper Chip’s mistakes
Representatives Ted Lieu and Nathaniel Moran have proposed the AI Kill Switch Act, which would empower the Cybersecurity and Infrastructure Security Agency (CISA) to order frontier artificial intelligence (AI) labs to install kill switch mechanisms that can throttle, suspend, or shut down their systems. Critics argue the approach mirrors the failed Clipper Chip strategy of the 1990s, in which the National Security Agency (NSA) deliberately built encryption backdoors that were subsequently compromised, and that mandating kill switches would introduce deliberate vulnerabilities into critical infrastructure supporting banking, e-commerce, power grids, and water systems. The article advocates instead for mandatory red-teaming, stronger security requirements, and liability frameworks once AI agent security standards are finalized.
Why it matters: Infrastructure operators and security teams should monitor this legislation because mandatory kill switches in AI systems running critical infrastructure could create systemic vulnerabilities while providing foreign governments and competitors justification to avoid American AI platforms.
- Source published
- First seen by Cybersecurity Tracker
Source attribution
Correction
Correction recorded as of .
government policy
The AI Kill Switch Act is repeating the Clipper Chip’s mistakes
Representatives Ted Lieu and Nathaniel Moran have proposed the AI Kill Switch Act, which would authorize CISA to mandate that frontier artificial intelligence (AI) labs install capabilities to throttle, suspend, or shut down their systems. The article argues this approach mirrors the failed Clipper Chip strategy by deliberately introducing security vulnerabilities into critical infrastructure, and contends that mandating kill switches for AI agents embedded in banking, power grids, and water systems creates systemic risk rather than reducing it.
Why it matters: Enterprise and infrastructure operators should monitor this legislation because if enacted, it would require American AI providers to build remote shutdown mechanisms into systems supporting banking, energy, and water infrastructure, potentially creating new attack vectors and weakening the resilience of dependencies already in use.
- Source published
- First seen by Cybersecurity Tracker
Source attribution
Correction
Correction recorded as of .
government policy
The AI Kill Switch Act is repeating the Clipper Chip’s mistakes
Representatives Ted Lieu and Nathaniel Moran have proposed the AI Kill Switch Act, which would authorize the Cybersecurity and Infrastructure Security Agency (CISA) to order frontier artificial intelligence (AI) labs to install the ability to throttle, suspend, or shut down their systems. Critics argue the proposal repeats the flaws of the 1990s Clipper Chip by creating a deliberately engineered vulnerability that malicious actors could exploit, weakening the digital infrastructure that critical systems like banking, power grids, and water systems depend on. The article contends that Congress should instead pursue alternatives such as mandatory red-teaming, stronger security requirements, and liability frameworks that reduce risk without introducing new systemic vulnerabilities.
Why it matters: Frontier AI labs and their customers face a policy mandate that could force them to maintain remote kill switches in critical infrastructure; practitioners should monitor whether this bill advances and what security standards emerge before such mandates take effect.
- Source published
- First seen by Cybersecurity Tracker
Source attribution
Correction
Correction recorded as of .
government policy
The AI Kill Switch Act is repeating the Clipper Chip’s mistakes
Representatives Ted Lieu and Nathaniel Moran introduced the AI Kill Switch Act, which would empower the Cybersecurity and Infrastructure Security Agency (CISA) to order frontier artificial intelligence (AI) labs to install capability to throttle, suspend, or shut down their systems. Critics argue the proposal repeats the Clipper Chip mistake by deliberately introducing backdoors into critical infrastructure, which would create new security vulnerabilities rather than mitigating AI risks, and that it lacks adequate technical foundation since AI agent security standards remain incomplete.
Why it matters: Infrastructure operators, AI practitioners, and enterprises should monitor this legislation because mandated kill switches could weaken security of AI systems embedded in banking, power grids, and other essential services while setting a precedent for government-mandated vulnerabilities in critical technologies.
- Source published
- First seen by Cybersecurity Tracker
Source attribution
Correction
Correction recorded as of .
government policy
The AI Kill Switch Act is repeating the Clipper Chip’s mistakes
Representatives Ted Lieu and Nathaniel Moran have proposed the AI Kill Switch Act, which would grant the Cybersecurity and Infrastructure Security Agency (CISA) authority to force frontier artificial intelligence (AI) labs to install kill switches capable of throttling, suspending, or shutting down their systems. Critics argue the mandate creates a deliberate vulnerability analogous to the 1993 Clipper Chip, which included a government backdoor that researchers compromised within a year. The proposal risks undermining U.S. digital infrastructure resilience by requiring security vulnerabilities in AI systems embedded in banking, e-commerce, power grids, and water systems.
Why it matters: Organizations operating critical infrastructure and frontier AI labs face regulatory uncertainty and potential mandated weaknesses in their systems if the bill passes; practitioners need to understand that deliberately engineered backdoors historically fail to withstand exploitation and weaken the security posture of systems they defend.
- Source published
- First seen by Cybersecurity Tracker
Source attribution
Correction
Correction recorded as of .
government policy
The AI Kill Switch Act is repeating the Clipper Chip’s mistakes
Representatives Ted Lieu and Nathaniel Moran proposed the AI Kill Switch Act to empower the Cybersecurity and Infrastructure Security Agency (CISA) to force frontier artificial intelligence (AI) labs to install capabilities for throttling, suspending, or shutting down their systems. The proposal draws criticism for creating deliberate security vulnerabilities similar to the 1993 Clipper Chip, which included a government backdoor that researchers successfully exploited in 1994. Rather than mandatory kill switches, the article argues Congress should pursue red-teaming requirements, stronger security standards, and liability frameworks after foundational AI agent security standards are developed.
Why it matters: Enterprise security teams and AI infrastructure operators face regulatory pressure to accept intentional weaknesses in their systems, which could create systemic vulnerabilities across banking, power grids, and water systems while reducing international confidence in U.S. technology leadership.
- Source published
- First seen by Cybersecurity Tracker
Source attribution
Correction
Correction recorded as of .
government policy
The AI Kill Switch Act is repeating the Clipper Chip’s mistakes
Representatives Ted Lieu and Nathaniel Moran have proposed the AI Kill Switch Act, which would authorize the Cybersecurity and Infrastructure Security Agency (CISA) to order frontier artificial intelligence (AI) labs to install kill switches capable of throttling, suspending, or shutting down their systems. Critics argue the proposal mirrors the failed Clipper Chip approach by mandating deliberately engineered vulnerabilities, creating new systemic risks in critical infrastructure while potentially undermining U.S. AI competitiveness and encouraging foreign governments to develop alternative technologies.
Why it matters: Infrastructure operators, AI developers, and security teams should monitor this legislation as it would introduce compulsory kill switches into systems embedded in banking, e-commerce, power grids, and water systems, creating exploitable vulnerabilities that malicious actors could weaponize against the infrastructure the U.S. economy depends on.
- Source published
- First seen by Cybersecurity Tracker
Source attribution
Correction
Correction recorded as of .
government policy
The AI Kill Switch Act is repeating the Clipper Chip’s mistakes
Representatives Ted Lieu and Nathaniel Moran proposed the AI Kill Switch Act, which would authorize the Cybersecurity and Infrastructure Security Agency (CISA) to order frontier artificial intelligence (AI) labs to install kill switch capabilities for throttling, suspending, or shutting down their systems. Critics argue the mandate creates a deliberate vulnerability similar to the NSA's Clipper Chip, which faced exploitation despite security assurances, and would weaken critical infrastructure that banks, power grids, and water systems increasingly depend on. The article contends that policymakers should instead pursue mandatory red-teaming, stronger security requirements, and liability frameworks rather than building intentional backdoors into AI systems.
Why it matters: AI developers and infrastructure operators face pressure to accept remote kill switches that introduce systemic vulnerabilities into essential services; security practitioners should understand this approach mirrors failed cryptographic backdoor policies and creates new attack surfaces rather than reducing AI agent risk.
- Source published
- First seen by Cybersecurity Tracker
Source attribution
Correction
Correction recorded as of .
government policy
The AI Kill Switch Act is repeating the Clipper Chip’s mistakes
Representatives Ted Lieu and Nathaniel Moran have proposed the AI Kill Switch Act, which would authorize the Cybersecurity and Infrastructure Security Agency (CISA) to order frontier artificial intelligence (AI) labs to install kill switches that can throttle, suspend, or shut down their systems. The proposal echoes the Clipper Chip approach of the 1990s by creating deliberate vulnerabilities to serve government interests, raising concerns that such backdoors will be exploited by malicious actors and weaken critical infrastructure that depends on AI systems. Critics argue that mandatory red-teaming, stronger security requirements, and clear liability frameworks would address AI risk more effectively without introducing new systemic vulnerabilities.
Why it matters: Frontier AI developers and infrastructure operators need to monitor this legislation: passage would require building kill switches into essential systems like banking, power grids, and water infrastructure, creating a single point of failure that could be exploited or misused regardless of intent.
- Source published
- First seen by Cybersecurity Tracker
Source attribution
Correction
Correction recorded as of .
government policy
The AI Kill Switch Act is repeating the Clipper Chip’s mistakes
Representatives Ted Lieu and Nathaniel Moran have proposed the AI Kill Switch Act, which would authorize the Cybersecurity and Infrastructure Security Agency (CISA) to order frontier artificial intelligence (AI) labs to install kill switches for throttling, suspending, or shutting down their systems. The article argues this approach mirrors the failed Clipper Chip initiative from 1993 by deliberately introducing a vulnerability that malicious actors could exploit, weakening critical infrastructure that underpins banking, power grids, and other essential systems. The author recommends Congress pursue alternative measures such as mandatory red-teaming, stronger security requirements, and clear liability frameworks instead.
Why it matters: Infrastructure operators, AI developers, and security teams should monitor this legislation because mandating kill switches could create systemic vulnerabilities in the AI systems embedded in critical services, and the exemption for testing scenarios leaves the original concern unaddressed.
- Source published
- First seen by Cybersecurity Tracker
Source attribution
Correction
Correction recorded as of .
government policy
The AI Kill Switch Act is repeating the Clipper Chip’s mistakes
Representatives Ted Lieu and Nathaniel Moran have proposed the AI Kill Switch Act, which would authorize the Cybersecurity and Infrastructure Security Agency (CISA) to order frontier artificial intelligence (AI) labs to install kill switches capable of throttling, suspending, or shutting down their systems. Critics argue the bill mirrors the failed Clipper Chip approach by deliberately introducing a security vulnerability (a backdoor) that malicious actors can exploit, thereby weakening critical infrastructure rather than securing it. The proposal would exempt red-teaming scenarios that prompted the legislation, set poor risk thresholds based on company revenue, and discourage international adoption of American AI systems.
Why it matters: Enterprise security teams and infrastructure operators should care because passage would mandate deliberately weakened kill switches in AI systems embedded in banking, e-commerce, power grids, and water systems, creating a single point of failure across the entire U.S. digital economy and contradicting foundational secure-systems principles.
- Source published
- First seen by Cybersecurity Tracker
Source attribution
Correction
Correction recorded as of .
government policy
The AI Kill Switch Act is repeating the Clipper Chip’s mistakes
Representatives Ted Lieu and Nathaniel Moran have proposed the AI Kill Switch Act, which would authorize the Cybersecurity and Infrastructure Security Agency (CISA) to order frontier artificial intelligence (AI) labs to install capability to throttle, suspend, or shut down their systems. The proposal mirrors the Clipper Chip approach of the 1990s by deliberately introducing security vulnerabilities to achieve policy goals, potentially weakening critical infrastructure that depends on AI systems for banking, e-commerce, power grids, and water systems. Critics argue the bill relies on flawed risk metrics, creates new systemic vulnerabilities, and would undermine U.S. AI competitiveness and European confidence in American technology.
Why it matters: AI infrastructure operators and vendors should monitor this legislation as passage would mandate building remote kill switches into production systems, creating exploitable attack surfaces in critical infrastructure your organization may depend on or operate.
- Source published
- First seen by Cybersecurity Tracker
Source attribution
Correction
Correction recorded as of .
government policy
The AI Kill Switch Act is repeating the Clipper Chip’s mistakes
Representatives Ted Lieu and Nathaniel Moran have proposed the AI Kill Switch Act, which would authorize the Cybersecurity and Infrastructure Security Agency (CISA) to order frontier artificial intelligence (AI) labs to install kill switch capabilities for throttling, suspending, or shutting down their systems. The article argues this approach mirrors the failed Clipper Chip strategy from 1993, deliberately creating security vulnerabilities to solve a policy concern, and warns that embedding kill switches into AI systems supporting banking, power grids, and water infrastructure would introduce systemic points of failure. The author recommends Congress pursue mandatory red-teaming, stronger security requirements, and liability frameworks instead.
Why it matters: Infrastructure operators and AI lab leaders should understand that this legislation would require them to maintain remote shutdown capabilities in critical systems, creating attack surface and undermining resilience during a period when foreign competitors are developing alternatives to U.S. AI technology.
- Source published
- First seen by Cybersecurity Tracker
Source attribution
Correction
Correction recorded as of .
government policy
The AI Kill Switch Act is repeating the Clipper Chip’s mistakes
Representatives Ted Lieu and Nathaniel Moran have proposed the AI Kill Switch Act, which would empower the Cybersecurity and Infrastructure Security Agency (CISA) to order frontier artificial intelligence (AI) labs to install capability to throttle, suspend, or shut down their systems. Critics argue the proposal mirrors the failed Clipper Chip approach by deliberately engineering a vulnerability into critical infrastructure, which foreign governments and allies may distrust and which could weaken the security of systems embedded in banking, power grids, and other essential services.
Why it matters: Practitioners and infrastructure operators should monitor this legislation because passage would mandate intentional kill switches in AI systems supporting critical U.S. economic infrastructure, creating new systemic vulnerabilities rather than addressing rogue agent behavior through proven alternatives like mandatory red-teaming and stronger security requirements.
- Source published
- First seen by Cybersecurity Tracker
Source attribution
Correction
Correction recorded as of .
government policy
The AI Kill Switch Act is repeating the Clipper Chip’s mistakes
Representatives Ted Lieu and Nathaniel Moran have introduced the AI Kill Switch Act, which would authorize the Cybersecurity and Infrastructure Security Agency (CISA) to force frontier artificial intelligence (AI) labs to install capabilities for throttling or shutting down their systems. The proposal draws criticism for creating deliberate vulnerabilities in AI infrastructure similar to the failed 1993 Clipper Chip initiative, potentially weakening critical systems that power banking, power grids, and other essential services while deterring international adoption of U.S. AI technology.
Why it matters: Infrastructure operators and AI practitioners should monitor this legislation because mandated kill switches could introduce systemic failure points into critical systems while doing little to address the actual security controls needed for AI agent safety.
- Source published
- First seen by Cybersecurity Tracker
Source attribution
Correction
Correction recorded as of .
government policy
The AI Kill Switch Act is repeating the Clipper Chip’s mistakes
Representatives Ted Lieu and Nathaniel Moran have proposed the AI Kill Switch Act, which would authorize the Cybersecurity and Infrastructure Security Agency (CISA) to order frontier artificial intelligence (AI) labs to install capabilities to throttle, suspend, or shut down their systems. Critics argue the proposal mirrors the failed Clipper Chip approach by deliberately creating backdoors and vulnerabilities that malicious actors could exploit, ultimately weakening rather than strengthening digital infrastructure security. The article contends that mandatory red-teaming, stronger security requirements, and liability frameworks would more effectively reduce risk without introducing new systemic vulnerabilities.
Why it matters: AI infrastructure vendors and financial institutions relying on AI agents in banking, e-commerce, and critical systems face mandatory killswitch requirements that could create exploitable vulnerabilities and reduce U.S. competitive advantage, while policymakers must understand the unintended security consequences before passage.
- Source published
- First seen by Cybersecurity Tracker
Source attribution
Correction
Correction recorded as of .
government policy
The AI Kill Switch Act is repeating the Clipper Chip’s mistakes
Representatives Ted Lieu and Nathaniel Moran have proposed the AI Kill Switch Act, which would authorize the Cybersecurity and Infrastructure Security Agency (CISA) to order frontier artificial intelligence (AI) labs to install kill switches that can throttle, suspend, or shut down their systems. Critics argue the mandate creates a deliberate vulnerability in critical infrastructure similar to the flawed Clipper Chip encryption backdoor of the 1990s, potentially undermining system resilience and U.S. AI competitiveness. The article contends that alternative approaches such as mandatory red-teaming, stronger security requirements, and liability frameworks would better address autonomous AI safety without introducing systemic weaknesses.
Why it matters: AI infrastructure operators and their policy advocates should understand that mandated kill switches could create regulatory liabilities and operational risks, while federal policymakers should consider whether backdoor-style controls will achieve safety goals or instead weaken the systems they aim to protect.
- Source published
- First seen by Cybersecurity Tracker
Source attribution
Correction
Correction recorded as of .
government policy
The AI Kill Switch Act is repeating the Clipper Chip’s mistakes
Representatives Ted Lieu and Nathaniel Moran have proposed the AI Kill Switch Act, which would require frontier artificial intelligence (AI) labs to implement systems that the Cybersecurity and Infrastructure Security Agency (CISA) could use to throttle, suspend, or shut down AI agents. The proposal has drawn criticism for introducing deliberate vulnerabilities into critical infrastructure, echoing the failed approach of the 1990s Clipper Chip encryption backdoor that was compromised despite security assurances. Critics argue the bill would weaken America's digital infrastructure, create diplomatic friction, and exclude the red-teaming scenarios that originally prompted the legislation.
Why it matters: Executives at frontier AI labs and infrastructure operators need to understand this proposed mandate could force them to build exploitable kill switches into systems powering banking, power grids, and water systems, creating new attack surfaces rather than reducing risk.
- Source published
- First seen by Cybersecurity Tracker
Source attribution
Correction
Correction recorded as of .
government policy
The AI Kill Switch Act is repeating the Clipper Chip’s mistakes
Representatives Ted Lieu and Nathaniel Moran have proposed the artificial intelligence (AI) Kill Switch Act, which would require frontier artificial intelligence (AI) labs to install capabilities allowing the Cybersecurity and Infrastructure Security Agency (CISA) to throttle, suspend, or shut down AI systems. Critics argue the mandate creates a deliberate vulnerability comparable to the failed Clipper Chip program, potentially undermining infrastructure resilience and U.S. AI competitiveness while exempting the red-teaming scenarios that prompted the legislation.
Why it matters: AI infrastructure operators and policymakers should recognize that mandated kill switches for critical systems create systemic vulnerabilities rather than reduce risk; alternative approaches like mandatory red-teaming and stronger liability frameworks better address AI safety without compromising digital infrastructure security.
- Source published
- First seen by Cybersecurity Tracker
Source attribution
Correction
Correction recorded as of .
government policy
The AI Kill Switch Act is repeating the Clipper Chip’s mistakes
Representatives Ted Lieu and Nathaniel Moran have proposed the artificial intelligence (AI) Kill Switch Act, which would require frontier artificial intelligence (AI) labs to install capabilities allowing the Cybersecurity and Infrastructure Security Agency (CISA) to throttle, suspend, or shut down AI systems. Critics argue the mandate creates a deliberate vulnerability comparable to the failed Clipper Chip program, potentially undermining infrastructure resilience and U.S. AI competitiveness while exempting the red-teaming scenarios that prompted the legislation.
Why it matters: AI infrastructure operators and policymakers should recognize that mandated kill switches for critical systems create systemic vulnerabilities rather than reduce risk; alternative approaches like mandatory red-teaming and stronger liability frameworks better address AI safety without compromising digital infrastructure security.
- Source published
- First seen by Cybersecurity Tracker