Techcrunch Anthropic Proposes AI Safety Bill with Judicial Oversight
Article Content
- •Anthropic's draft bill aims to regulate AI with risk thresholds.
- •The bill allows blocking AI releases based on judicial approval for significant risks.
- •Independent evaluators will assess AI models for potential catastrophic risks.
Anthropic has introduced a draft bill to U.S. lawmakers aimed at regulating AI technologies to mitigate risks. The proposal allows the federal government to block AI model releases if there is an imminent risk of death to over 50 people or property damage exceeding $1 billion, contingent on judicial approval. The risks could arise from biological weapon creation, models evading controls, or cyberattacks. The bill suggests using independent evaluators registered with the U.S. Commerce Department to assess risks, although concerns about potential bias in evaluations remain. This initiative follows a legislative vacuum in U.S. AI policy, which has relied on voluntary industry commitments. Anthropic's spokesperson emphasized the need for transparency and robust security programs from AI developers. The bill reflects ongoing discussions about AI safety amid rising incidents involving autonomous AI systems.
Ask AI about this cluster
Answers cite the sources they use