Anthropic Proposes AI Safety Law To Block Dangerous Models Posing Risks To Lives
· Free Press Journal
Artificial intelligence company Anthropic has submitted a draft bill to US lawmakers proposing stricter safety regulations for advanced AI systems, including powers to prevent the deployment of models considered capable of causing catastrophic harm.
According to a Politico report, the proposed legislation would allow federal authorities to block an AI model's release if it poses an imminent threat of killing or injuring more than 50 people or causing damage exceeding $1 billion.
Visit extonnews.click for more information.
However, these powers would apply only to specific risks involving biological weapons, cyberattacks or AI systems operating beyond human control.
The proposal comes amid increasing concerns about the potential dangers associated with rapidly advancing artificial intelligence technologies.
'Trust The Model The Least': Microsoft CEO Satya Nadella Warns Of Superintelligence Risks, Calls For Stronger AI Security & Human OversightMandatory AI Safety Testing, Independent Reviews
Anthropic's proposed framework would require AI developers to conduct safety assessments before deploying advanced models.
Companies would also have to disclose their testing findings publicly and permit independent evaluators to conduct regular reviews.
An Anthropic spokesperson said the draft legislation reflected recommendations contained in the company's Advanced AI Framework.
The company has advocated these measures among federal and state lawmakers since June.
According to the spokesperson, the proposal would give the federal government authority to prevent or discourage the deployment of particularly dangerous AI models.
Anthropic believes Congress should introduce safeguards that address emerging technological threats while establishing clearer accountability for AI developers.
However, the company declined to disclose how widely it had circulated the draft or whether lawmakers had responded positively.
Cybersecurity Threats Intensify Calls For Regulation
The proposal follows growing concern about AI-related security incidents, including reported breaches and hacking activities involving artificial intelligence systems.
Such developments have intensified discussions among policymakers and technology companies about the need for stronger safeguards.
The draft specifically identifies biological weapons, uncontrolled AI behaviour and cyberattacks as potential triggers for federal intervention.
These risks would need to meet the proposed severity threshold before authorities could restrict a model's deployment.
Despite mounting calls for additional oversight, President Donald Trump's administration has resisted proposals for broader AI regulation.
Anthropic's initiative therefore highlights differing approaches to balancing technological innovation with public safety.
The legislation remains a company-backed proposal rather than an enacted law.
Its prospects will depend on whether US lawmakers decide to introduce and advance the proposed measures through Congress.