Keypoint: Since our last update on US artificial intelligence (AI) legislation impacting the private sector, Utah enacted the first AI private sector bill of 2024, Oklahoma moved closer to passing an AI Bill of Rights, Connecticut’s bill advanced through a committee, and California lawmakers introduced two bills that would establish transparency requirements around generative AI and personal information used to train AI models.
Below is our third update on the status of pending US artificial intelligence (AI) legislation that would affect the private sector.
Table of Contents
1. What’s New
In this third alert, we continue to organize our updates by the following five types of bills: (a) algorithmic discrimination; (b) automated employment decision making; (c) AI Bill of Rights; (d) “working group” bills; and (e) other types of bills that do not fit into those categories but would, if passed, impact private entities. Note that some of the bills blur the lines between the various categories. For a more detailed description of certain bills referenced below, please see our first update from February and second update from earlier this month.
a. Algorithmic Discrimination
This first category of bills seeks to prevent “algorithmic discrimination,” which is generally defined as an automated decision tool’s differential treatment of an individual or group based on their protected class. Note that these bills apply to both AI developers and deployers, which include (but is not limited to) employers that use such tools to make employment decisions. Therefore, this category does overlap with our next category of “automated employment decision making” bills.
Connecticut’s SB 2 was the subject of a public hearing on February 29 and filed with the Legislative Commissioner’s Office on March 14, after receiving a favorable vote out of the General Law Committee. On March 25, the bill was filed with the Office of Legislative Research and Office of Fiscal Analysis. The bill will likely need to pass out of two more committees before reaching a floor vote. It is anticipated that the bill will be amended as it advances.
In Colorado, Senate Majority Leader Robert Rodriguez spoke on a Husch Blackwell panel on March 7 and indicated that the AI bill he is drafting will track portions of the Connecticut bill. The Colorado bill has yet to be introduced.
In Illinois, HB 5322 was assigned to the Executive Committee on March 12.
Other active bills we are tracking in this category that have no new updates since our last post include: (1) California’s AB 2930 (formerly AB 331); (2) Hawaii’s HB 1607 and SB 2524; (3) Illinois’ HB5116; (3) New York’s A8195; (4) Oklahoma’s HB 3835; and (5) Vermont’s H.710 and H.711.
b. Automated Employment Decision Making
Our next category looks at bills that specifically apply to the use of AI in the employment context. These bills seek to regulate AI tools, commonly referred to as “automated employment decision tools” (AEDTs) or “predictive data analytics,” used to make hiring, firing, promotion, and compensation decisions.
In Maryland, hearings were held on March 6 and March 14 for HB 1255 and SB 0957 (the Senate counterpart), respectively. Both bills remain in committee.
In Massachusetts, H.1873 remains with the Joint Committee on Labor and Workforce Development which extended the bill’s reporting deadline to May 8 pending a concurrence.
Turning to New Jersey, A4030 was introduced in the state Assembly on March 7. This bill is identical to Senate bill S1588 and would make it unlawful to sell an AEDT unless the tool (1) has undergone a bias audit; (2) includes an annual bias audit service at no additional cost; and (3) includes a notice that the tool is subject to the requirements of the law. Both bills also require employers to notify subjected candidates if an AEDT will be used. A4030 has been referred to the Assembly Science, Innovation and Technology Committee, while S1588 remains with the Senate Labor Committee.
Staying in New Jersey, S2964 was introduced in the Senate on March 18, and is identical to Assembly bill A3855 (introduced on February 22). Both bills prohibit employers from using AEDTs unless the tool undergoes an annual bias audit. The bills outline the parameters of a bias audit and further require employers to publish the date of their most recent audit and a summary of the audit’s results. Finally, the bills require employers to notify candidates or applicants when an AEDT will be used and provide instructions for how a person may request an alternative selection process or reasonable accommodation.
Other AEDT bills we are tracking that have no new updates include: (1) Illinois’ HB 3773; (2) New York’s A9314, A9315, A7859, S5641A, and S7623A; (3) Rhode Island’s H 7786; and (4) Vermont’s H.114.
c. AI Bill of Rights
Bills in this category seek to establish an AI Bill of Rights, providing state residents with specified rights regarding their use of AI.
Our only update for this category is in Oklahoma, where HB 3453 unanimously passed the House with amendments on March 14. As amended, the bill would grant Oklahoma citizens the following seven rights: (1) the right to know when they are interacting with AI when “consequential information” (undefined) is exchanged; (2) the right to know when contracts or other legally binding documents are generated entirely by AI; (3) the right to know when they are consuming images or text entirely generated by AI; (4) the right to rely on watermarks or other content credentials to verify the authenticity of a creative product; (5) the right to know that reasonable security measures are implemented by companies using a person’s identifiable information in an AI model; (6) the right to consent to any derivative media generated by AI that uses a person’s audio recordings or images; and (7) the right to be free from unlawful discrimination through algorithmic or model bias. If enacted, the law would go into effect on November 1, 2024.
The other bills we are tracking in this category are New York’s A8129 and its companion Senate bill S8209, which both remain in committee.
d. “Working Group” Bills
The next category includes bills that primarily create government commissions or working groups to study the implementation of AI technologies and develop recommendations for future regulation.
The big update in this category is that Utah Governor Spencer Cox signed SB 149 into law on March 13. The law will take effect on May 1, 2024, and you can read more about the law’s requirements in our article here.
The final update in this category comes from Florida, where HB 1459 died in the House before the state legislature adjourned on March 8.
The other “working group” bills we are tracking with no new updates are: (1) Massachusetts’ SB 2539 and (2) Hawaii’s HB2176.
e. “Other” Bills
Our final category encompasses “other” bills that do not neatly fit into the above categories.
There are two new bills in California we are tracking in this “other” category — SB 942 and AB 3204. Titled the “AI Transparency Act,” SB 942 would require certain businesses that provide generative AI systems to create an “AI detection tool” that allows individuals to query whether text, images, videos, or multimedia content was created by a generative AI system. The bill would also require the same businesses to include visible disclosures that identify whether such content was generated by AI. This disclosure must be “machine detectable” and permanent or difficult to remove. Finally, a covered business must register with, and provide a URL to its AI detection tools to, the Department of Technology which is charged with creating a “Generative AI Registry” that lists the names of covered businesses and URLs to their respective AI detection tools. The bill was introduced on January 17 and referred to committee on March 20.
Turning to AB 3204, this bill would apply to “data digesters” which the bill defines as businesses that use “personal information to train artificial intelligence.” It requires data digesters to register with the California Privacy Protection Agency (with a registration fee) and provide the Agency with, among other information, the categories of “personal information” and “sensitive personal information” the business uses to train AI, as those terms are defined in the California Consumer Privacy Act. The bill then requires the Agency to post this information publicly on its website. AB 3204 was introduced on February 16 and referred to committee on March 11.
The remaining bills we are tracking in this “other” category with no new updates include: (1) California’s SB 1047 and (2) Oklahoma’s HB 3577.
2. Bill Tracker Charts and Map
For more information on all the state bills introduced to date, including links to the bills, bill status, last action, hearing dates, and bill sponsor information, please see the following charts:
To access our AI tracker map, click the following link:
3. Recent Husch Articles and Resources on AI-Related Issues
[View source.]
See more
DISCLAIMER: Because of the generality of this update, the information provided herein may not be applicable in all situations and should not be acted upon without specific legal advice based on particular situations. Attorney Advertising.
© Husch Blackwell LLP
Refine your interests »