Fact Check: The RAISE Act applies to AI companies whose models were trained using more than $100 million in computing resources.

Fact Check: "The RAISE Act applies to AI companies whose models were trained using more than $100 million in computing resources."

What We Know

The Responsible AI Safety and Education (RAISE) Act, recently passed by the New York State Senate, specifically targets large AI companies that have trained their models using more than $100 million in computational resources. According to a statement from New York State Senator Andrew Gounardes, the RAISE Act aims to ensure that these companies develop safety plans to mitigate risks associated with advanced AI technologies (source-1). The legislation mandates that companies publish safety protocols and risk evaluations, focusing on severe risks such as automated crime and bioweapons (source-2).

The act is designed to apply only to the largest AI developers, ensuring that it does not stifle innovation among smaller companies or academic researchers (source-2). This targeted approach is emphasized in multiple reports, confirming that the threshold for applicability is indeed set at $100 million in computing resources used for training AI models (source-3).

Analysis

The claim that the RAISE Act applies to AI companies whose models were trained using more than $100 million in computing resources is supported by multiple credible sources. The official press release from Senator Gounardes explicitly states this requirement, highlighting the focus on large AI companies that pose significant risks due to their advanced technologies (source-1).

Additionally, reports from reputable outlets such as TechCrunch and Yahoo Finance reiterate this point, confirming that the legislation's transparency requirements are specifically aimed at companies that meet this financial threshold for computational resources (source-2, source-4). The targeted nature of the legislation is also noted as a deliberate design choice to avoid hindering innovation among smaller entities in the AI sector (source-2).

The sources cited are reliable, as they come from official government communications and established news organizations that cover technology and policy issues. There is no indication of bias in the reporting, and the information is consistent across multiple outlets.

Conclusion

Verdict: True
The claim that the RAISE Act applies to AI companies whose models were trained using more than $100 million in computing resources is accurate. The legislation explicitly sets this threshold as a criterion for compliance, focusing on the largest AI developers to ensure safety without stifling innovation among smaller firms.

Sources

1. Sen. Gounardes' AI Safety Bill Passes the State Senate
2. New York passes a bill to prevent AI-fueled disasters
3. New York passes RAISE Act to regulate frontier AI safety
4. New York passes a bill to prevent AI-fueled disasters
5. New York Senate progresses bill to prevent AI from ...
6. New York Implements a Bill to Prevent AI-fueled Disasters
7. Major AI labs must comply with New York's transparency ...
8. New York passes a bill to prevent AI-fueled disasters