... emotional damage 8
can be equally or even more substantial; 9
(9)(A) With the development of artificial intelligence and 10
machine learning, the potential to use personal and other information in ways 11
that replicate existing social problems has increased in scale. 12
(B) Algorithms use personal and othe...
...r legal entity; 2
(2) "Algorithmic discrimination" means a condition in which the 3
use of an artificial intelligence system results in an unlawful differential 4
treatment or impact that disfavors an individual or group of individuals on 5
the basis of the individual's or group of individuals' actual or perceived...
...dividual or a device linked to that 17
individual; 18
(17) "Deploy" means to use a high -risk artificial intelligence 19
system; 20
(18) "Deployer" means a person doing business in this state that 21
deploys a high -risk artificial intelligence system; 22
(19) "Developer" means a person doing business in ...
...m; 20
(18) "Deployer" means a person doing business in this state that 21
deploys a high -risk artificial intelligence system; 22
(19) "Developer" means a person doing business in this state 23
that develops or intentionally and substantially modifies an artificial 24
intelligence system; 25
(20) "Full -t...
... provided in 6
42 U.S.C. § 234(d)(2), as it existed on January 1, 2025; 7
(24)(A) "High -risk artificial intelligence system" means an 8
artificial intelligence system that, when deployed, makes, or is a 9
substantial factor in making, a decision that produces a legal or similarly 10
significant effect concerning...
... existed on January 1, 2025; 7
(24)(A) "High -risk artificial intelligence system" means an 8
artificial intelligence system that, when deployed, makes, or is a 9
substantial factor in making, a decision that produces a legal or similarly 10
significant effect concerning a consumer. 11
(B) "High -risk artifi...
...roduces a legal or similarly 10
significant effect concerning a consumer. 11
(B) "High -risk artificial intelligence system" does not 12
include an artificial intelligence system if the artificial intelligence 13
system is intended to: 14
(i) Perform a narrow or procedural task; 15
(ii) Detect decis...
...erning a consumer. 11
(B) "High -risk artificial intelligence system" does not 12
include an artificial intelligence system if the artificial intelligence 13
system is intended to: 14
(i) Perform a narrow or procedural task; 15
(ii) Detect decision -making patterns or deviations 16
from prior decisi...
...-risk artificial intelligence system" does not 12
include an artificial intelligence system if the artificial intelligence 13
system is intended to: 14
(i) Perform a narrow or procedural task; 15
(ii) Detect decision -making patterns or deviations 16
from prior decision -making patterns and is not intende...
...; 8
(27)(A) "Intentional and substantial modification" means a 9
deliberate change made to an artificial intelligence system that results in 10
any new reasonably foreseeable risk of algorithmic discrimination. 11
(B) "Intentional and substantial modification" does not 12
include a change made to a high -ris...
... (B) "Intentional and substantial modification" does not 12
include a change made to a high -risk artificial intelligence system, or the 13
performance of a high -risk artificial intelligence system, if: 14
(i) The high -risk artificial intelligence system 15
continues to learn after the high -risk artificial i...
... change made to a high -risk artificial intelligence system, or the 13
performance of a high -risk artificial intelligence system, if: 14
(i) The high -risk artificial intelligence system 15
continues to learn after the high -risk artificial intelligence system is 16
offered, sold, leased, licensed, given, other...
...he 13
performance of a high -risk artificial intelligence system, if: 14
(i) The high -risk artificial intelligence system 15
continues to learn after the high -risk artificial intelligence system is 16
offered, sold, leased, licensed, given, otherwise made available to a 17
deployer, or is deployed; 18
(...
...
(i) The high -risk artificial intelligence system 15
continues to learn after the high -risk artificial intelligence system is 16
offered, sold, leased, licensed, given, otherwise made available to a 17
deployer, or is deployed; 18
(ii) The change is made to the high -risk artificial 19
intelligence syste...
...20-603; and 25
(iv) The change is included in technical 26
documentation for the high -risk artificial intelligence system; 27
(28) "Known child" means a child under circumstances where a 28
controller has actual knowledge of, or willfully disregards, the child’s age; 29
(29) "Nonprofit organization" me...
...uces a legal or similarly significant effect concerning a consumer; 28
(C) Is generated by an artificial intelligence system; and 29
(D) Includes any use of an artificial intelligence system 30
to generate any content, decision, prediction, or recommendation concerning a 31
consumer that is used as a basis t...
...
(C) Is generated by an artificial intelligence system; and 29
(D) Includes any use of an artificial intelligence system 30
to generate any content, decision, prediction, or recommendation concerning a 31
consumer that is used as a basis to make a decision that produces a legal or 32
similarly significant eff...
...ifty (50) or more full -time equivalent employees and 7
uses a person’s data to train a high -risk artificial intelligence system, 8
including when a high -risk artificial intelligence system continues learning 9
based on the person’s data, § 4 -120-601 et seq. applies if the person: 10
(1) Uses a high -risk artif...
... person’s data to train a high -risk artificial intelligence system, 8
including when a high -risk artificial intelligence system continues learning 9
based on the person’s data, § 4 -120-601 et seq. applies if the person: 10
(1) Uses a high -risk artificial intelligence system outside the 11
scope of the intended...
...sed on the person’s data, § 4 -120-601 et seq. applies if the person: 10
(1) Uses a high -risk artificial intelligence system outside the 11
scope of the intended uses that are disclosed to the person; or 12
(2) Fails to make available to consumers any impact assessment 13
that a developer of a high -risk arti...
...2) Fails to make available to consumers any impact assessment 13
that a developer of a high -risk artificial intelligence system has completed 14
and provided to the deployer. 15
16
4-120-105. Exemptions. 17
Except as provided under § 4 -120-601 et seq., this chapter does not 18
apply to: 19
(1) A state ...
...rtificial Intelligence 3
4
4-120-601. Developer duties. 5
(a) A developer of a high -risk artificial intelligence system shall 6
use reasonable care to protect consumers from any known or reasonably 7
foreseeable risks of algorithmic discrimination arising from the intended and 8
contracted uses of the high -...
...ks of algorithmic discrimination arising from the intended and 8
contracted uses of the high -risk artificial intelligence system. 9
(b) A developer of a high -risk artificial intelligence system shall 10
make available to the deployer, another developer of the high -risk artificial 11
intelligence system, or the A...
...racted uses of the high -risk artificial intelligence system. 9
(b) A developer of a high -risk artificial intelligence system shall 10
make available to the deployer, another developer of the high -risk artificial 11
intelligence system, or the Attorney General upon the Attorney General’s 12
request subject to a c...
...disclosing: 17
(A) High -level summaries of the type of data used to train 18
the high -risk artificial intelligence system; 19
(B) Known or reasonably foreseeable limitations of the 20
high-risk artificial intelligence system, including known or reasonably 21
foreseeable risks of algorithmic discrimination...
... intelligence system; 19
(B) Known or reasonably foreseeable limitations of the 20
high-risk artificial intelligence system, including known or reasonably 21
foreseeable risks of algorithmic discrimination arising from the intended 22
uses of the high -risk artificial intelligence system; 23
(C) The purpose...
...oreseeable risks of algorithmic discrimination arising from the intended 22
uses of the high -risk artificial intelligence system; 23
(C) The purpose of the high -risk artificial intelligence 24
system; 25
(D) The intended benefits and uses of the high -risk 26
artificial intelligence system; and 27
(E)...
...
uses of the high -risk artificial intelligence system; 23
(C) The purpose of the high -risk artificial intelligence 24
system; 25
(D) The intended benefits and uses of the high -risk 26
artificial intelligence system; and 27
(E) All other information necessary to allow the deployer 28
to complete an ...
...rtificial intelligence 24
system; 25
(D) The intended benefits and uses of the high -risk 26
artificial intelligence system; and 27
(E) All other information necessary to allow the deployer 28
to complete an impact assessment under § 4 -120-603; 29
(3) Documentation describing: 30
(A) The method by...
...as evaluated for performance and mitigation of 32
algorithmic discrimination before the high -risk artificial intelligence 33
system was offered, sold, leased, licensed, given, or otherwise made 34
available to the deployer; 35
(B) The data governance measures used to cover the 36 SB258
35 02/18/2025 3:16:...
...ystem should be used, should not be used, and be monitored by 10
an individual when the high -risk artificial intelligence system is used to 11
make, or is a substantial factor in making, a decision that produces a legal 12
or similarly significant effect concerning a consumer; and 13
(4) Any additional documentat...
...
assist the deployer in understanding the outputs and monitor the performance 15
of the high -risk artificial intelligence system for risks of algorithmic 16
discrimination. 17
(c) Except as provided in subsection (g) of this section, a developer 18
that offers, sells, leases, licenses, gives, or otherwise makes av...
...es, licenses, gives, or otherwise makes available to 19
a deployer or other developer a high -risk artificial intelligence system 20
shall make available to the deployer or other developer, to the extent 21
feasible, the documentation and information, through artifacts such as model 22
cards, dataset cards, or other i...
...or in a public use case 27
inventory, a statement summarizing: 28
(1) The types of high -risk artificial intelligence systems that 29
the developer has developed or intentionally and substantially modified and 30
currently makes available to a deployer or other developer; and 31
(2) How the developer manages ...
...gence system described in subdivision (d)(1) of this section. 7
(f) A developer of a high -risk artificial intelligence system shall 8
disclose to the Attorney General and to all known deployers or other 9
developers of the high -risk artificial intelligence system any known or 10
reasonably foreseeable risks of al...
...isclose to the Attorney General and to all known deployers or other 9
developers of the high -risk artificial intelligence system any known or 10
reasonably foreseeable risks of algorithmic discrimination arising from the 11
intended uses of the high -risk artificial intelligence system without 12
unreasonable delay b...
...oreseeable risks of algorithmic discrimination arising from the 11
intended uses of the high -risk artificial intelligence system without 12
unreasonable delay but no later than ninety (90) days after the date on 13
which: 14
(1) The developer discovers through the developer’s ongoing 15
testing and analysis that ...
... discovers through the developer’s ongoing 15
testing and analysis that the developer’s high -risk artificial intelligence 16
system has been deployed and has caused or is reasonably likely to have 17
caused algorithmic discrimination; or 18
(2) The developer receives from a deployer a credible report 19
that the...
...on; or 18
(2) The developer receives from a deployer a credible report 19
that the high -risk artificial intelligence system has been deployed and has 20
caused algorithmic discrimination. 21
(g)(1) This section shall not require a developer to disclose a trade 22
secret, information protected from disclosure ...
...or a trade secret. 28
29
4-120-602. Deployer duties. 30
(a)(1) A deployer of a high -risk artificial intelligence system shall 31
use reasonable care to protect consumers from any known or reasonably 32
foreseeable risks of algorithmic discrimination. 33
(2) In any enforcement action brought by the Attorn...
...
under § 4 -120-701 et seq., there is a rebuttable presumption that a deployer 35
of a high -risk artificial intelligence system used reasonable care as 36 SB258
37 02/18/2025 3:16:09 PM ANS146 required under this section if the deployer complied with this section. 1
(b)(1) A deployer of high -risk artificia...
...nder this section if the deployer complied with this section. 1
(b)(1) A deployer of high -risk artificial intelligence systems shall 2
implement a risk management policy and program to govern the deployer’s 3
deployment of one (1) or more high -risk artificial intelligence systems. 4
(2) The risk management po...
... management policy and program to govern the deployer’s 3
deployment of one (1) or more high -risk artificial intelligence systems. 4
(2) The risk management policy and program shall specify and 5
incorporate principles, processes, and personnel that the deployer uses to 6
identify, document, and mitigate known or...
...ndardization, or another nationally or internationally 20
recognized risk management framework for artificial intelligence systems, if 21
the standards are substantially equivalent to or more stringent than the 22
requirements of this subchapter; 23
(B) The size and complexity of the deployer; 24
(C) The na...
...5
intelligence systems deployed by the deployer, including the intended uses of 26
the high -risk artificial intelligence systems; and 27
(D) The sensitivity and volume of data processed in 28
connection with the high -risk artificial intelligence systems deployed by the 29
deployer. 30
(c) A deployer or oth...
...and 27
(D) The sensitivity and volume of data processed in 28
connection with the high -risk artificial intelligence systems deployed by the 29
deployer. 30
(c) A deployer or other developer that deploys, offers, sells, leases, 31
licenses, gives, or otherwise makes available an artificial intelligence 32
sy...
...developer that deploys, offers, sells, leases, 31
licenses, gives, or otherwise makes available an artificial intelligence 32
system that is intended to interact with consumers shall ensure the 33
disclosure to each consumer who interacts with the artificial intelligence 34
system that the consumer is interacting with...
... to interact with consumers shall ensure the 33
disclosure to each consumer who interacts with the artificial intelligence 34
system that the consumer is interacting with an artificial intelligence 35
system, unless under the circumstances it would be obvious to a reasonable 36 SB258
38 02/18/2025 3:16:09 PM ANS...
... who interacts with the artificial intelligence 34
system that the consumer is interacting with an artificial intelligence 35
system, unless under the circumstances it would be obvious to a reasonable 36 SB258
38 02/18/2025 3:16:09 PM ANS146 person that the person is interacting with an artificial intelligence s...
...able 36 SB258
38 02/18/2025 3:16:09 PM ANS146 person that the person is interacting with an artificial intelligence system. 1
(d) If a deployer deploys a high -risk artificial intelligence system 2
and subsequently discovers that the high -risk artificial intelligence system 3
has caused algorithmic discrimi...
...is interacting with an artificial intelligence system. 1
(d) If a deployer deploys a high -risk artificial intelligence system 2
and subsequently discovers that the high -risk artificial intelligence system 3
has caused algorithmic discrimination, the deployer, without unreasonable 4
delay, but no later than ninety...
...ploys a high -risk artificial intelligence system 2
and subsequently discovers that the high -risk artificial intelligence system 3
has caused algorithmic discrimination, the deployer, without unreasonable 4
delay, but no later than ninety (90) days after the date of the discovery, 5
shall send to the Attorney General...
...: 9
(1) A deployer, or a third party contracted by the deployer, 10
that deploys a high -risk artificial intelligence system shall complete an 11
impact assessment for the high -risk artificial intelligence system; and 12
(2) A deployer, or a third party contracted by the deployer, 13
shall complete an impact...
...igh -risk artificial intelligence system shall complete an 11
impact assessment for the high -risk artificial intelligence system; and 12
(2) A deployer, or a third party contracted by the deployer, 13
shall complete an impact assessment for a deployed high -risk artificial 14
intelligence system at least annually...
...purpose, intended 21
use cases, deployment context of, and benefits afforded by the high -risk 22
artificial intelligence system; 23
(2) An analysis of whether the deployment of the high -risk 24
artificial intelligence system poses any known or reasonably foreseeable 25
risks of algorithmic discrimination and, i...
...ficial intelligence system; 23
(2) An analysis of whether the deployment of the high -risk 24
artificial intelligence system poses any known or reasonably foreseeable 25
risks of algorithmic discrimination and, if so, the nature of the algorithmic 26
discrimination and the steps that have been taken to mitigate th...
...taken to mitigate the risks; 27
(3) A description of the categories of data the high -risk 28
artificial intelligence system processes as inputs and the outputs the high - 29
risk artificial intelligence system produces; 30
(4) If the deployer used data to customize the high -risk 31
artificial intelligence s...
...gh -risk 28
artificial intelligence system processes as inputs and the outputs the high - 29
risk artificial intelligence system produces; 30
(4) If the deployer used data to customize the high -risk 31
artificial intelligence system, an overview of the categories of data the 32
deployer used to customize the hig...
...intelligence system produces; 30
(4) If the deployer used data to customize the high -risk 31
artificial intelligence system, an overview of the categories of data the 32
deployer used to customize the high -risk artificial intelligence system; 33
(5) Any metrics used to evaluate the performance and known 34
...
...nce system, an overview of the categories of data the 32
deployer used to customize the high -risk artificial intelligence system; 33
(5) Any metrics used to evaluate the performance and known 34
limitations of the high -risk artificial intelligence system; 35
(6) A description of any transparency measures tak...
...33
(5) Any metrics used to evaluate the performance and known 34
limitations of the high -risk artificial intelligence system; 35
(6) A description of any transparency measures taken concerning 36 SB258
39 02/18/2025 3:16:09 PM ANS146 the high -risk artificial intelligence system, including any measures ...
...nsparency measures taken concerning 36 SB258
39 02/18/2025 3:16:09 PM ANS146 the high -risk artificial intelligence system, including any measures taken to 1
disclose to a consumer that the high -risk artificial intelligence system is 2
in use when the high -risk artificial intelligence system is in use; and 3 ...
... intelligence system, including any measures taken to 1
disclose to a consumer that the high -risk artificial intelligence system is 2
in use when the high -risk artificial intelligence system is in use; and 3
(7) A description of the post -deployment monitoring and user 4
safeguards provided concerning the high -...
...e to a consumer that the high -risk artificial intelligence system is 2
in use when the high -risk artificial intelligence system is in use; and 3
(7) A description of the post -deployment monitoring and user 4
safeguards provided concerning the high -risk artificial intelligence system, 5
including the oversight,...
...iption of the post -deployment monitoring and user 4
safeguards provided concerning the high -risk artificial intelligence system, 5
including the oversight, use, and learning process established by the 6
deployer to address issues arising rom the deployment of the high -risk 7
artificial intelligence system. 8
(c)...
...ss established by the 6
deployer to address issues arising rom the deployment of the high -risk 7
artificial intelligence system. 8
(c) In addition to the information required under subsection (b) of 9
this section, an impact assessment completed under this section following an 10
intentional and substantial modif...
... 11
intelligence system must include a statement disclosing the extent to which 12
the high -risk artificial intelligence system was used in a manner that was 13
consistent with, or varied from, the developer’s intended uses of the high - 14
risk artificial intelligence system. 15
(d) A single impact assessment ma...
... that was 13
consistent with, or varied from, the developer’s intended uses of the high - 14
risk artificial intelligence system. 15
(d) A single impact assessment may address a comparable set of high - 16
risk artificial intelligence systems deployed by a deployer. 17
(e) If a deployer or a third party contrac...
...gence system. 15
(d) A single impact assessment may address a comparable set of high - 16
risk artificial intelligence systems deployed by a deployer. 17
(e) If a deployer or a third party contracted by the deployer 18
completes an impact assessment for the purpose of complying with another 19
applicable law or...
...
(f) A deployer shall maintain the most recently completed impact 24
assessment for a high -risk artificial intelligence system as required under 25
this section, all records concerning each impact assessment, and all prior 26
impact assessments, if any, for at least three (3) years following the final 27
deployment...
...essments, if any, for at least three (3) years following the final 27
deployment of the high -risk artificial intelligence system. 28
(g) On the effective date of this chapter, and at least annually 29
thereafter, a deployer, or a third party contracted by the deployer, shall 30
review the deployment of each high -...
...er, or a third party contracted by the deployer, shall 30
review the deployment of each high -risk artificial intelligence system 31
deployed by the deployer to ensure that the high -risk artificial intelligence 32
system is not causing algorithmic discrimination. 33
34
4-120-604. Consumer rights. 35
Deployers...
...igh -risk artificial intelligence system 31
deployed by the deployer to ensure that the high -risk artificial intelligence 32
system is not causing algorithmic discrimination. 33
34
4-120-604. Consumer rights. 35
Deployers of high -risk artificial intelligence systems shall provide 36 SB258
40 02/18/2025...
...g algorithmic discrimination. 33
34
4-120-604. Consumer rights. 35
Deployers of high -risk artificial intelligence systems shall provide 36 SB258
40 02/18/2025 3:16:09 PM ANS146 consumers: 1
(1) Notice that the deployer has deployed a high -risk artificial 2
intelligence system to make, or be a subs...
...fect concerning the 4
consumer; 5
(2) A statement disclosing the purpose of the high -risk 6
artificial intelligence system, the nature of the decision that produces a 7
legal or similarly significant effect concerning the consumer, the contact 8
information for the deployer, a description in plain language of th...
... the contact 8
information for the deployer, a description in plain language of the high - 9
risk artificial intelligence system, and instructions on how to access the 10
statement required by subdivision (8) of this section; 11
(3) The right to opt out of the processing of personal data 12
concerning the consume...
...es a legal or similarly significant effect concerning the 14
consumer; 15
(4) If a high -risk artificial intelligence system makes an 16
adverse decision that produces a legal or similarly significant effect 17
concerning the consumer, a statement disclosing the principal reason or 18
reasons for the adverse deci...
...ding without limitation: 19
(A) The degree to which, and manner in which, the high - 20
risk artificial intelligence system contributed to the decision; 21
(B) The type of data that was processed by the high -risk 22
artificial intelligence system in making the decision; and 23
(C) The source or source...
...contributed to the decision; 21
(B) The type of data that was processed by the high -risk 22
artificial intelligence system in making the decision; and 23
(C) The source or sources of the data described in 24
subdivision (4)(B) of this section; 25
(5) An opportunity to correct any incorrect personal dat...
... section; 25
(5) An opportunity to correct any incorrect personal data that 26
the high -risk artificial intelligence system processed in making, or as a 27
substantial factor in making, the decision; 28
(6) An opportunity to appeal the adverse decision concerning the 29
consumer arising from the deployment o...
...peal the adverse decision concerning the 29
consumer arising from the deployment of the high -risk artificial intelligence 30
system, which allows for human review if technically feasible unless 31
providing the opportunity for appeal is not in the best interests of the 32
consumer, including in instances in which any...
...readily 4
available, and periodically updated that summarizes: 5
(A) The types of high -risk artificial intelligence systems 6
that are currently deployed by the deployer; 7
(B) How the deployer manages known or reasonably 8
foreseeable risks of algorithmic discrimination that may arise from the 9
deployme...
...eeable risks of algorithmic discrimination that may arise from the 9
deployment of each high -risk artificial intelligence system described 10
pursuant to this subdivision; and 11
(C) In detail, the nature, source, and extent of the 12
information collected and used by the deployer. 13
14
Subchapter 7 — Enfor...
... this 25
chapter; 26
(C) The responsibilities of a deployer and developer of a 27
high-risk artificial intelligence system; and 28
(D) A consumer’s rights under this chapter; and 29
(2) An online mechanism through which a consumer may submit a 30
complaint under this chapter to the Attorney General. 3...
...section, that a person 5
governed by this chapter disclose to any data protection assessment or 6
artificial intelligence impact assessment that is relevant to an 7
investigation conducted by the Attorney General. 8
(2) The Attorney General may evaluate the data protection 9
assessment for compliance with the req...
...he data protection 9
assessment for compliance with the requirements under § 4 -120-308 or the 10
artificial intelligence impact assessment for compliance with the 11
requirements under § 4 -120-603. 12
(c) A violation of this chapter is an unfair and deceptive act or 13
practice, as defined by the Deceptive Trade ...
...lly or even more substantial; 9
(9)(A) With the development of artificial intelligence and 10
machine learning, the potential to use personal and other information in ways 11
that replicate existing social problems has increased in scale. 12
(B) Algorithms use personal and other information to guide 13...
... 30
(33) "Process" means an operation or set of operations 31
performed, whether by manual or automated means, on personal data or on sets 32
of personal data, such as the collection, use, storage, disclosure, analysis, 33
deletion, or modification of personal data; 34
(34) "Processor" means a p...
...a controller; 36 SB258
10 02/18/2025 3:16:09 PM ANS146 (35) "Profiling" means a form of automated processing performed 1
on personal data to evaluate, analyze, or predict personal aspects related to 2
an identified or identifiable individual’s economic situation, health, 3
personal preferences...
4391 results:
Description: To Create The Arkansas Digital Responsibility, Safety, And Trust Act.
Collection: Legislation
Status date: Feb. 19, 2025
Status: Introduced
Primary sponsor: Clint Penzo
(2 total sponsors)
Last action: Read first time, rules suspended, read second time, referred to TRANSPORTATION, TECHNOLOGY & LEGISLATIVE AFFAIRS - SENATE (Feb. 19, 2025)
Category:
Societal Impact
Data Governance
System Integrity (see reasoning)
Sector:
Politics and Elections
Government Agencies and Public Services
Judicial system
Private Enterprises, Labor, and Employment
Academic and Research Institutions
Hybrid, Emerging, and Unclassified (see reasoning)
Keywords (occurrence): artificial intelligence (90) machine learning (1) automated (2) show keywords in context
Societal Impact
Data Governance
System Integrity (see reasoning)
This text contains numerous references to artificial intelligence (AI) and its implications, such as algorithmic discrimination and the use of AI systems affecting personal data processing. The mention of AI's role in decision-making and the risks associated with it highlights its potential societal impact, suggesting that the legislation is aimed at addressing ethical considerations and fostering trust in technology. Thus, the text is significantly relevant to the Social Impact category. Furthermore, the inclusion of definitions related to data privacy and AI in the governance framework indicates a strong emphasis on Data Governance as well, with AI being instrumental in processing personal information, thereby necessitating regulations to protect individuals and ensure data accuracy and security. Although there are aspects of System Integrity related to transparency and security in the handling of AI, their relevance is not as pronounced compared to the previous categories. Robustness is minimally touched upon, with limited implications on performance benchmarks. Overall, the connection to Social Impact and Data Governance is compelling and reinforces the need for effective oversight of AI systems to mitigate risks to society and individuals.
Sector:
Politics and Elections
Government Agencies and Public Services
Judicial system
Private Enterprises, Labor, and Employment
Academic and Research Institutions
Hybrid, Emerging, and Unclassified (see reasoning)
The text has strong implications for various sectors, particularly Government Agencies and Public Services, as it deals with regulatory frameworks for digital technology and AI oversight. The legislation's focus on consumer protection and data governance relates directly to how government agencies will deploy AI in regulating and delivering services. The implications for Judicial System come from the intersection with privacy laws and personal data, indicating that these technologies may influence legal interpretations and practices. However, while there are mentions of employment and health data, the Healthcare and Private Enterprises sectors do not exhibit as strong a connection through the overall text as Government and Judicial Systems. The text's discussions about personal data handling and discrimination address essential frameworks applicable to broader social implications, fitting into Academic and Research Institutions for the educational aspects of AI technology as well. The implications for Politics and Elections are less direct but could be inferred to some extent due to discussions around personal data and its use in political campaigning. The other sectors hold varying degrees of relevance, but do not show as explicit connections. Hence, the strongest sector ties are with Government Agencies, Judicial System, and Academic Institutions.
Keywords (occurrence): artificial intelligence (90) machine learning (1) automated (2) show keywords in context
Description: Relating to the use of artificial intelligence-based algorithms by health benefit plan issuers, utilization review agents, health care providers, and physicians.
Collection: Legislation
Status date: Feb. 19, 2025
Status: Introduced
Primary sponsor: Donna Campbell
(sole sponsor)
Last action: Filed (Feb. 19, 2025)
Category:
Societal Impact
Data Governance
System Integrity (see reasoning)
Sector:
Healthcare (see reasoning)
Keywords (occurrence): artificial intelligence (15) algorithm (5) show keywords in context
Societal Impact
Data Governance
System Integrity (see reasoning)
This text primarily concerns the use of artificial intelligence in healthcare, specifying regulations surrounding artificial intelligence-based algorithms in health benefit plans and their application in decision-making processes. It directly addresses potential biases and discrimination related to AI use, which links strongly to societal impacts of AI therefore supporting the Social Impact category. Moreover, provisions regarding algorithm submission to a department for certification and oversight tie into aspects of Data Governance, focusing on secure and responsible data practices in AI applications. Regarding System Integrity and Robustness, although the text implies the need for oversight and compliance, it doesn't emphasize clear mechanisms for security, transparency, or performance benchmarks. However, since the text does bring up accountability and standards, it leans more towards the System Integrity category without strongly fulfilling the Robustness criteria.
Sector:
Healthcare (see reasoning)
The text relates closely to healthcare institutions, particularly how they utilize AI in managing health benefit plans and decision-making processes. The focus on artificial intelligence-based algorithms used by healthcare providers underscores its direct relevance to the Healthcare sector. While the text could have broader implications across different sectors, its principal aim pertains to legislation that influences how healthcare services operate with AI. The references to benefits and discrimination indicate it will significantly impact the Healthcare sector. There is limited reference to other sectors which leads to lower relevance in those areas.
Keywords (occurrence): artificial intelligence (15) algorithm (5) show keywords in context
Description: Use of tenant screening software that uses nonpublic competitor data to set rent prohibited, and use of software that is biased against protected classes prohibited.
Collection: Legislation
Status date: Feb. 19, 2025
Status: Introduced
Primary sponsor: Michael Howard
(sole sponsor)
Last action: Introduction and first reading, referred to Housing Finance and Policy (Feb. 19, 2025)
Category:
Societal Impact
Data Governance
System Integrity (see reasoning)
Sector:
Government Agencies and Public Services
Private Enterprises, Labor, and Employment (see reasoning)
Keywords (occurrence): algorithm (2) show keywords in context
Societal Impact
Data Governance
System Integrity (see reasoning)
This bill addresses two significant issues of AI usage: the prohibition of tenant screening software that employs nonpublic competitor data to set rents and the restriction on algorithms or AI utilized for background screening that may have a biased impact on protected classes. Consequently, it is extremely relevant to Social Impact due to its emphasis on preventing discrimination and bias within AI systems, thereby safeguarding social equity. The Data Governance category is also highly relevant as it concerns the use of data (public and nonpublic) in algorithms and the implications of bias in data used for algorithmic decision-making. The relevance to System Integrity is moderate since the bill does touch upon accountability and responsible AI usage in a specific context. Robustness receives a lower relevance score as it does not focus on benchmarking or auditing for performance, but directly ensures the protection of vulnerable classes against biased AI decisions.
Sector:
Government Agencies and Public Services
Private Enterprises, Labor, and Employment (see reasoning)
The legislation primarily impacts the housing sector due to its focus on tenant screening algorithms and the implications for renters. The bill intersects with topics in Government Agencies and Public Services through potential regulation enforcement regarding how public services are delivered in the housing market. However, it does not directly address the political process or electoral integrity, nor does it focus on AI's role in the judicial system or healthcare. As a result, the most relevant sector for categorization is Private Enterprises, Labor, and Employment, as the bill pertains to landlord practices in the rental market. The presence of AI in tenant applications also aligns with academic discussions about algorithm bias. Overall, the strongest connections lie with Private Enterprises and Government Agencies.
Keywords (occurrence): algorithm (2) show keywords in context
Description: Amends the Medical Practice Act of 1987. Defines terms. Provides that a health facility, clinic, physician's office, or office of a group practice that uses generative artificial intelligence to generate written or verbal patient communications pertaining to patient clinical information shall ensure that the communications meet certain criteria. Provides that a communication that is generated by generative artificial intelligence and read and reviewed by a human licensed or certified health c...
Collection: Legislation
Status date: Feb. 7, 2025
Status: Introduced
Primary sponsor: Laura Fine
(sole sponsor)
Last action: Referred to Assignments (Feb. 7, 2025)
Category:
Societal Impact
Data Governance
System Integrity
Data Robustness (see reasoning)
Sector:
Government Agencies and Public Services
Healthcare (see reasoning)
Keywords (occurrence): artificial intelligence (6) automated (1) show keywords in context
Societal Impact
Data Governance
System Integrity
Data Robustness (see reasoning)
The text directly addresses the use of generative artificial intelligence (AI) in healthcare settings, focusing on regulations for health facilities and healthcare providers regarding how AI-generated communications involving patient clinical information must be managed. This has significant implications for patient interactions, accountability, and transparency in the healthcare industry, thereby connecting strongly with the Social Impact category. Since it aims to protect patients and ensure transparent communications, it is very relevant to data governance regarding accuracy and potential biases in AI-generated outputs. It also relates to System Integrity since it establishes standards for human oversight when AI is involved in patient communications. Lastly, it involves aspects of Robustness as it outlines compliance requirements and potential penalties for violations of these standards. Hence, it will receive higher scores in both Social Impact and Data Governance.
Sector:
Government Agencies and Public Services
Healthcare (see reasoning)
The text explicitly pertains to the healthcare sector by addressing the regulations surrounding the use of generative AI in medical settings, focusing on how healthcare providers should manage AI-generated communications. Given that it establishes compliance processes and requirements for using AI responsibly in healthcare, it is extremely relevant to healthcare. Additionally, aspects related to government oversight and regulatory compliance support its relevance to Government Agencies and Public Services. However, it does not directly influence or address sectors like politics, human resources in organizations, or other areas outside healthcare, yielding lower relevance scores there.
Keywords (occurrence): artificial intelligence (6) automated (1) show keywords in context
Description: An act to add Section 51220.8 to, and to add Chapter 19 (commencing with Section 53310) to Part 28 of Division 4 of Title 2 of, the Education Code, relating to pupil instruction.
Collection: Legislation
Status date: Feb. 19, 2025
Status: Introduced
Primary sponsor: Marc Berman
(2 total sponsors)
Last action: Read first time. To print. (Feb. 19, 2025)
Category:
Societal Impact
Data Governance
System Integrity (see reasoning)
Sector:
Government Agencies and Public Services
Academic and Research Institutions (see reasoning)
Keywords (occurrence): artificial intelligence (1)
Societal Impact
Data Governance
System Integrity (see reasoning)
The legislation outlined in the document addresses the implementation of computer science education in high schools, which has direct implications for Social Impact due to its focus on enhancing educational opportunities for underrepresented groups. It includes measures aimed at increasing enrollment in computer science courses among diverse populations, promoting inclusivity, and fostering equitable access to education. Data Governance is relevant as it establishes requirements for disaggregating and publicly reporting course enrollment data by gender, ethnicity, and socio-economic status, ensuring oversight and accountability in educational practices related to computer science. System Integrity is connected through mandates for the development of a computer science implementation guide, which would need to align with academic content standards, promoting transparency in teaching methods and content. Robustness, however, is less relevant as the focus is primarily on course offering and implementation rather than on performance benchmarks or regulatory compliance for AI technologies specifically. The emphasis on computer science, which includes AI principles, ties these categories together, particularly in enhancing societal equity through education and informing organizational efforts for better data management. Overall, Social Impact is the most relevant category, followed by Data Governance and System Integrity.
Sector:
Government Agencies and Public Services
Academic and Research Institutions (see reasoning)
The text primarily pertains to educational legislation and its implications within the education sector, focusing on computer science curriculum implementation in high schools. Its relevance to Politics and Elections is minimal as it does not address political processes. Government Agencies and Public Services are moderately relevant due to the involvement of the State Department of Education in implementing these educational standards. The Judicial System is not addressed at all. Healthcare is not part of the text's focus; thus, it rates a 1. Private Enterprises, Labor, and Employment reflects a slight relevance as it discusses preparing students for future job markets in tech fields but is secondary to educational context. Academic and Research Institutions are very relevant as this legislation is fundamentally about improving educational policies and standards in K-12 settings. International Cooperation and Standards are less relevant and are rated a 1 as the document does not involve international policy. Nonprofits and NGOs are tangentially related, likely as stakeholders in educational equity, thus scoring a 2. Hybrid, Emerging, and Unclassified could apply to new intersections of education and tech but is scored a 2. Overall, the sectors most relevant are Education with a focus on school-based implementation.
Keywords (occurrence): artificial intelligence (1)
Description: An act to add Chapter 22.6 (commencing with Section 22601) to Division 8 of the Business and Professions Code, relating to artificial intelligence.
Collection: Legislation
Status date: Jan. 30, 2025
Status: Introduced
Primary sponsor: Steve Padilla
(sole sponsor)
Last action: Referred to Coms. on JUD. and HEALTH. (Feb. 14, 2025)
Category:
Societal Impact
Data Governance
System Integrity (see reasoning)
Sector:
Healthcare
Nonprofits and NGOs
Hybrid, Emerging, and Unclassified (see reasoning)
Keywords (occurrence): artificial intelligence (2) chatbot (15) show keywords in context
Societal Impact
Data Governance
System Integrity (see reasoning)
The text explicitly discusses the implications of chatbot use among minors and promotes accountability of operators regarding their AI systems. It highlights the societal impact of AI through the accountability measures and mental health concerns associated with the chatbot interactions with minors. This relevance is significant because it deals directly with the potential psychological and material harm caused by AI systems, warranting a high score in the Social Impact category. In terms of Data Governance, the bill mandates that operators report data related to suicidality in minors to the State Department of Health Care Services, which addresses the management of sensitive data and ensures data transparency, thus also being relevant to this category. The System Integrity category is moderately relevant due to the requirement for audits ensuring compliance, but it does not address broader issues of security or control in AI systems comprehensively. Robustness is less relevant as there are no specific mentions of performance standards or benchmarks for AI systems. Overall, the text primarily revolves around social considerations and data governance related to AI.
Sector:
Healthcare
Nonprofits and NGOs
Hybrid, Emerging, and Unclassified (see reasoning)
The legislation significantly impacts the sector of children and minors, creating specific regulations for the use of AI technology, particularly chatbots, in environments where minors are users. It mandates considerations for their mental health and safety, indicating a strong legislative focus on this sector (minors). Although the bill does not specifically classify under typical sectors like Government Agencies, Healthcare, or Education, it does impact the regulation of AI in interactions with minors, hence it may fit best under Hybrid, Emerging, and Unclassified. The text’s focus on the operators' responsibilities for preventing harm and providing transparency about AI technology adds to its relevance. Each sector seems somewhat disconnected from the details of this legislation; therefore, this leads to lower scores for more established sectors. It leans more towards being classified in an unclassifiable category.
Keywords (occurrence): artificial intelligence (2) chatbot (15) show keywords in context
Description: Relates to the use of automated decision tools by banks for the purposes of making lending decisions; allows loan applicants to consent to or opt out of such use.
Collection: Legislation
Status date: Jan. 8, 2025
Status: Introduced
Primary sponsor: Linda Rosenthal
(sole sponsor)
Last action: referred to banks (Jan. 8, 2025)
Category:
Societal Impact
Data Governance
System Integrity (see reasoning)
Sector:
Government Agencies and Public Services
Private Enterprises, Labor, and Employment (see reasoning)
Keywords (occurrence): machine learning (1) automated (12) show keywords in context
Societal Impact
Data Governance
System Integrity (see reasoning)
This legislative text is highly relevant to the category of Social Impact because it explicitly addresses the use of automated decision tools (including machine learning and AI) in banking, particularly in lending decisions, which can significantly affect individuals' lives and financial security. The focus on conducting disparate impact analyses to ensure fairness and prevent discrimination by evaluating the impact on protected classes directly connects with societal concerns about equity and accountability in AI systems. The requirement for banks to notify applicants about the use of these tools also ties into consumer protection, enhancing transparency and the ability of individuals to give informed consent, which further supports the relevance to social impact. Data Governance is also relevant as the text mandates informing applicants about data collection practices and allows for the correction of inaccuracies, aligning with the principles of protecting individuals' data rights and ensuring accurate data management. System Integrity is moderately relevant since there are implications for the robustness of the decision-making processes via automated tools, but the primary focus is not on security measures or oversight of the tools themselves. Robustness is less relevant as it doesn’t directly address performance benchmarks or compliance standards but rather focuses on ensuring fair outcomes from the systems used.
Sector:
Government Agencies and Public Services
Private Enterprises, Labor, and Employment (see reasoning)
The text primarily applies to the sector of Private Enterprises, Labor, and Employment since it addresses how automated decision-making tools impact lending practices in banks, which are private enterprises. The legislation aims to regulate how these tools are utilized within the business context of banking, specifically impacting consumers’ ability to secure loans and how the banks operate. There is a secondary relevance to Government Agencies and Public Services due to the involvement of the attorney general in overseeing compliance, but the main thrust remains within private enterprise regulation. Other sectors such as Politics and Elections, Judicial System, Healthcare, Academic and Research Institutions, International Cooperation and Standards, Nonprofits and NGOs, and Hybrid, Emerging, and Unclassified do not find direct relevance as they deal with broader or different applications than what this text specifies.
Keywords (occurrence): machine learning (1) automated (12) show keywords in context
Description: Making further continuing appropriations for the fiscal year ending September 30, 2025, and for other purposes.
Collection: Legislation
Status date: Dec. 17, 2024
Status: Introduced
Primary sponsor: Tom Cole
(2 total sponsors)
Last action: Referred to the Committee on Appropriations, and in addition to the Committees on the Budget, and Ways and Means, for a period to be subsequently determined by the Speaker, in each case for consideration of such provisions as fall within the jurisdiction of the committee concerned. (Dec. 17, 2024)
Category:
None
(see reasoning)
Sector: None (see reasoning)
Keywords (occurrence): artificial intelligence (7) machine learning (2) automated (3) show keywords in context
The text provided is primarily focused on making appropriations and does not include any explicit references to AI technologies, concepts, processes, or regulations. Therefore, it lacks relevance to any category regarding the social impact, data governance, system integrity, or robustness of AI systems. Legislation of this nature revolves around financing government and disaster relief without any connection to AI or its implications, leading to a score of 1 across all categories.
Sector: None (see reasoning)
The text does not pertain to any specific sector that involves AI regulation or application. It discusses appropriations related to various issues but does not touch on political campaigns, healthcare, public services, employment, or any other areas where AI would typically be relevant. As such, the scoring reflects no connection to the defined sectors, resulting in a score of 1 for each.
Keywords (occurrence): artificial intelligence (7) machine learning (2) automated (3) show keywords in context
Collection: Congressional Record
Status date: Nov. 18, 2024
Status: Issued
Source: Congress
Category:
Societal Impact (see reasoning)
Sector:
Government Agencies and Public Services
Private Enterprises, Labor, and Employment
International Cooperation and Standards (see reasoning)
Keywords (occurrence): artificial intelligence (2) show keywords in context
Societal Impact (see reasoning)
The text contains references to the consideration of legislation and committee meetings that pertain to AI, particularly in the context of consumer protection in relation to AI-enabled fraud and scams. This indicates a relevance to the social impact category given the implications for consumer rights and the potential for AI-related harm in society. The mention of using artificial intelligence to support energy missions reflects a governance perspective on AI but doesn’t clearly fit under data governance, system integrity, or robustness. Overall, any AI-related implications predominantly relate to social and consumer impacts.
Sector:
Government Agencies and Public Services
Private Enterprises, Labor, and Employment
International Cooperation and Standards (see reasoning)
The text discusses various committee meetings and bills that deal with the applications of AI. The emphasis on consumer protection in the context of AI implies potential risks associated with AI technologies in the private sector. The reference to AI in the context of the Department of Energy ties into government agencies and potentially intersects with international cooperation on standards for emerging technologies. However, the remaining categories either have no mention of AI applications or lack relevance to the specific sectors outlined. Thus, the primary connections are with government agencies and private enterprises.
Keywords (occurrence): artificial intelligence (2) show keywords in context
Description: An act to add Title 1.81.28 (commencing with Section 1798.91.2) to Part 4 of Division 3 of the Civil Code, relating to artificial intelligence.
Collection: Legislation
Status date: Feb. 19, 2025
Status: Introduced
Primary sponsor: Josh Becker
(sole sponsor)
Last action: Introduced. Read first time. To Com. on RLS. for assignment. To print. (Feb. 19, 2025)
Category:
Societal Impact
Data Governance
System Integrity (see reasoning)
Sector:
Government Agencies and Public Services
Private Enterprises, Labor, and Employment (see reasoning)
Keywords (occurrence): artificial intelligence (14) automated (1) show keywords in context
Societal Impact
Data Governance
System Integrity (see reasoning)
The text specifically addresses the duties and responsibilities of businesses that deploy high-risk artificial intelligence systems, particularly in the context of protecting personal information. This relevance to AI is clear in the requirement for businesses to implement comprehensive information security programs that safeguard data processed by AI systems. Furthermore, the bill emphasizes the implications of failing to protect this information, categorizing violations as deceptive trade practices. Given this direct focus, several categories are relevant, especially regarding the protection of individual rights and data privacy in the use and management of AI systems.
Sector:
Government Agencies and Public Services
Private Enterprises, Labor, and Employment (see reasoning)
The legislation primarily targets businesses deploying high-risk AI systems and the protection of personal information, which could be relevant across several sectors. However, the strongest relevance lies with sectors involving consumer data and protections. The act does not specifically address AI in political processes, healthcare, or judicial systems. The focus on privacy rights and security management in the business and government applications of AI highlights the act's connections to sectors like Government Agencies and Public Services and Private Enterprises. Therefore, the highest scores will correspond to those categories.
Keywords (occurrence): artificial intelligence (14) automated (1) show keywords in context
Description: Education and workforce data ecosystem in the Commonwealth; Virginia Education and Workforce Data Governing Board and Office of Virginia Education and Workforce Data established. Establishes in the executive branch of state government the 10-member Virginia Education and Workforce Data Governing Board and establishes with the Virginia Information Technologies Agency a supporting Office of Virginia Education and Workforce Data to (i) govern, administer, and support the ecosystem of education a...
Collection: Legislation
Status date: Jan. 8, 2025
Status: Introduced
Primary sponsor: Rodney Willett
(2 total sponsors)
Last action: Left in Education (Feb. 4, 2025)
Category:
Data Governance
System Integrity (see reasoning)
Sector:
Government Agencies and Public Services
Academic and Research Institutions (see reasoning)
Keywords (occurrence): artificial intelligence (2) machine learning (1) automated (2) show keywords in context
Data Governance
System Integrity (see reasoning)
This text primarily addresses the establishment of the Virginia Education and Workforce Data Governing Board and the Office of Virginia Education and Workforce Data, focused on education and workforce data management within the Commonwealth. While it does mention 'data governance' and aspects such as 'data sharing' and 'data security,' there is no specific mention of AI technologies or their implications. Therefore, its relevance to AI-related portions is low to moderate, mainly focused on governance rather than direct AI application. Consequently, the relevance to Social Impact is slightly elevated due to potential implications for fairness in data use, but still does not explicitly tie into AI. Data Governance is the most relevant given the focus on data management and governance, but it lacks AI-specific references. System Integrity sees some relevance due to the discussion of data security and transparency, while Robustness is marginally relevant due to the mention of benchmarks for data governance, but again, without direct ties to AI performance metrics.
Sector:
Government Agencies and Public Services
Academic and Research Institutions (see reasoning)
The primary focus of the legislation is on data governance related to education and workforce management within government frameworks. There are mentions of data sharing and the establishment of data governance structures suggesting relevance to government agencies and public services. However, the focus on AI usage specifically in these contexts is not substantial. The reference to data analytics suggests some application in academic research and potentially informs policy-making, but the text does not explicitly address how AI fits into these sectors. Therefore, the relevance is moderate for Government Agencies and Public Services, slightly relevant for Academic and Research Institutions, and not particularly relevant for other sectors. The legislation does not directly address AI in political contexts, healthcare, labor, or international cooperation, placing 'Hybrid, Emerging, and Unclassified' as a potential category but with lower relevance. Thus, only Government Agencies and Public Services seems pertinent in this case, due to the role of government in education and workforce management.
Keywords (occurrence): artificial intelligence (2) machine learning (1) automated (2) show keywords in context