[CLSTR]
< Back to all clusters
[TECHNOLOGY] · United States · 21 sources

started · updated

Part of situation: Anthropic AI safety and regulatory positioning (3 clusters)

Anthropic proposes federal AI safety legislation

Anthropic AI regulation artificial intelligence safety federal legislation AI model governance

Anthropic is proposing a bipartisan federal bill to the US Congress aimed at establishing oversight for advanced artificial intelligence models. The proposal seeks to create a framework for government intervention in cases where AI models pose a catastrophic risk.

Under the proposed legislation, the release or operation of an AI model could be blocked if there is a predictable and immediate risk of more than 50 deaths or serious injuries, or if potential property damage exceeds $1 billion. These interventions would be limited to specific high-stakes scenarios, including threats involving biological weapons, cyberattacks, or instances where a model escapes developer controls.

The bill would require a federal judge in Washington, D.C., to authorize any model block, preventing federal agencies from acting on their own authority. Additionally, the framework suggests that independent evaluators, registered with the US Commerce Department, would assess these risks. While AI labs could fund these evaluators, the proposal mandates that compensation must not be tied to the results of the assessments to ensure impartiality.

Entities

Anthony Cimino · Anthropic · U.S. Commerce Department · US Commerce Department · US Congress · US Department of Commerce · United States Congress · Washington, D.C.

Claims

What the coverage asserts, and how many sources carry each claim.

Sources

Anthropic pide que un juez autorice bloquear un modelo de IA [nacion.com.uy]
NJ wants to lead the way on holding AI firms liable for harm [www.njspotlightnews.org]
Exclusive: bipartisan bill would develop federal guidance for distillation attacks, convene industry on defensive cyber [cybernoz.com]
We are closer to AI regulation than we think [cyberprotection-magazine.com]
European Parliament Divided on How to Govern AI At Home and Abroad [www.europesays.com]
Is this what AI regulation will look like? Break it and then pay a $25 billion court judgment like META? [www.jubakpicks.com]
Anthropic Wants Congress to Pass an AI Bill. [autogpt.net]
Anthropic Wants to Decide When AI Becomes Too Dangerous to Release [insidetelecom.com]
Claude pronašao 493 sigurnosna problema u otvorenom kodu [www.logicno.com]
Claude Duet: A Nova Arma da Anthropic na Corrida dos Agentes de IA! [www.netthings.pt]
Anthropic predlaže blokadu opasnih AI modela [srpskainfo.com]
Bem-vindos à Matrix [www.cartacapital.com.br]
Bartulica: Dok Elon Musk predstavlja leteće automobile, Bruxelles smišlja nove propise [narod.hr]
Rogue AI Must Threaten 50 Lives Before Being Shut Down, Anthropic Suggests [novaramedia.com]
Projeto da Anthropic prevê bloqueio de IA diante de riscos catastróficos [olhardigital.com.br]
‘These … clowns’: Why cyber researchers bristle at Anthropic’s ‘self-replicating code’ claims [san.com]
Defensible AI: Can your governance culture withstand pressure? [businessday.ng]
Anthropic Wants More Government Oversight on Its Own A.I. Models [cheddar.com]
AI And The US Midterms: How Anthropic Is Approaching Political Bias And Election Integrity [www.news18.com]
States Move to Rein in Rogue AI as Federal Guardrails Lag [www.pymnts.com]
Anthropic torna o Claude mais rápido e mais barato [sapo.pt]