At a glance

  • Each of the ten state comprehensive privacy laws in the table below gives a resident the right to opt out of the processing of personal data for profiling in furtherance of decisions that produce legal or similarly significant effects, such as decisions on lending, housing, insurance, education, employment, or health care, and requires a documented data protection assessment of profiling that presents a reasonably foreseeable risk of listed harms (in Delaware, for controllers that process the data of at least 100,000 consumers). New Hampshire, and Delaware until January 1, 2027, limit the opt-out to solely automated decisions. None of the ten covers a person acting in a commercial or employment context.
  • Minnesota (since July 31, 2025) and Connecticut (since July 1, 2026) also let a consumer whose data was profiled for such a decision question the result, learn the reason for it, and review the personal data used, in Connecticut each only if feasible. Connecticut requires an impact assessment of that profiling for processing activities created or generated on or after August 1, 2026; Delaware will require one from January 1, 2027 of controllers that process the data of at least 50,000 consumers.
  • From January 1, 2027, a California business that meets the CCPA thresholds and uses automated decisionmaking technology (ADMT), meaning technology that replaces or substantially replaces human decisionmaking, for a significant decision must give a pre-use notice, honor a right to opt out unless an exception such as a human appeal applies, and answer access requests. Since January 1, 2026, a risk assessment has been required before such processing begins.
  • Since January 1, 2026, Texas’s biometric identifier law has not applied to training, processing, or storing biometric identifiers to develop or offer AI models or systems, unless a system is used or deployed for the purpose of uniquely identifying a specific individual. An identifier captured for AI training and later used for another commercial purpose becomes subject to the law’s possession and destruction duties and its penalties.

Which privacy laws regulate AI?

In the United States, the privacy rules that reach AI come mostly from state law: comprehensive consumer privacy laws that regulate profiling, California’s CCPA regulations on automated decisionmaking technology, and biometric privacy laws such as Illinois’s BIPA. The Federal Trade Commission adds enforcement under Section 5 of the FTC Act and COPPA, including settlement orders to delete models and algorithms, not only data.

None of these laws needs the word “AI” to apply. Virginia defines profiling as any form of automated processing performed on personal data to evaluate, analyze, or predict personal aspects related to an identified or identifiable natural person’s economic situation, health, personal preferences, interests, reliability, behavior, location, or movements (Va. Code 59.1-575). A credit model or a tenant screening score that uses personal data fits that definition.

State privacy laws on profiling, as of November 16, 2026 (states verified from statute text; not every state with a privacy law)
StateLawProfiling opt-outAssessment dutyEffectiveEnforcement
VirginiaConsumer Data Protection Act, Va. Code 59.1-575 to 59.1-584Profiling in furtherance of decisions that produce legal or similarly significant effects concerning the consumer (59.1-577(A)(5))Data protection assessment for profiling that presents a reasonably foreseeable risk of unfair or deceptive treatment of, or unlawful disparate impact on, consumers; financial, physical, or reputational injury; an offensive intrusion upon seclusion; or other substantial injury (59.1-580)January 1, 2023Attorney General exclusively, after 30 days’ written notice and a chance to cure; civil penalties of up to $7,500 per violation; no private right of action
ColoradoColorado Privacy Act, C.R.S. 6-1-1301 et seq.; rules at 4 CCR 904-3Profiling in furtherance of decisions that produce legal or similarly significant effects concerning a consumer (6-1-1306(1)(a)(I)(C)); by rule, a controller may decline the opt-out for Human Involved Automated Processing if it gives the consumer listed information (Rule 9.04)Data protection assessment for profiling that presents a reasonably foreseeable risk of unfair or deceptive treatment of, or unlawful disparate impact on, consumers; financial or physical injury; an offensive intrusion; or other substantial injury (6-1-1309); for profiling-based decisions, review and update at least annually (Rule 8.05(C))July 1, 2023Attorney General and district attorneys; a violation is a deceptive trade practice; no private right of action
ConnecticutData Privacy Act, Conn. Gen. Stat. 42-515 to 42-525, as amended by Public Acts 25-113 and 26-64Since July 1, 2026, profiling in furtherance of any automated decision that produces any legal or similarly significant effect (42-518(a)(5)); before that date, solely automated decisions; plus a right to question the result (42-518(a)(6))Data protection assessment for profiling that presents a reasonably foreseeable risk of listed harms; an impact assessment for profiling to make such a decision, for processing activities created or generated on or after August 1, 2026 (42-522)July 1, 2023; amendments July 1, 2026 and October 1, 2026Attorney General exclusively; a violation is an unfair trade practice under 42-110b; no private right of action
TexasData Privacy and Security Act, Bus. & Com. Code ch. 541Profiling in furtherance of a decision that produces a legal or similarly significant effect concerning the consumer (541.051(b)(5)(C))Data protection assessment for profiling that presents a reasonably foreseeable risk of listed harms (541.105)July 1, 2024Attorney General, after 30 days’ written notice and a chance to cure; civil penalties of up to $7,500 per violation; no private right of action
OregonConsumer Privacy Act, ORS 646A.570 to 646A.589Profiling the consumer in furtherance of decisions that produce legal effects or effects of similar significance (646A.574); since January 1, 2026, no such profiling where the controller knows, or willfully disregards, that the consumer is under 16 (646A.578(2)(c))Data protection assessment for profiling that presents a reasonably foreseeable risk of listed harms, kept for at least five years (646A.586)July 1, 2024Attorney General exclusively; civil penalties of up to $7,500 per violation; no private right of action (646A.589)
MontanaConsumer Data Privacy Act, MCA 30-14-2801 et seq.Profiling in furtherance of automated decisions that produce legal or similarly significant effects concerning the consumer (30-14-2808(1)(e))Data protection assessment for profiling that presents a reasonably foreseeable risk of listed harms (30-14-2814)October 1, 2024Attorney General exclusively, using the Consumer Protection Act’s powers; no private right of action
DelawarePersonal Data Privacy Act, 6 Del. C. ch. 12DUntil January 1, 2027, profiling in furtherance of solely-automated decisions; from that date, automated decisions, that produce legal or similarly significant effects (12D-104(a)(6))Controllers that process the data of at least 100,000 consumers (50,000 from January 1, 2027), excluding data processed solely to complete a payment transaction: data protection assessment for profiling that presents a reasonably foreseeable risk of listed harms, for processing activities created or generated on or after July 1, 2025; from January 1, 2027, also an impact assessment for profiling in furtherance of such automated decisions (12D-108)January 1, 2025 (Department of Justice); amendments January 1, 2027Department of Justice; a violation is an unlawful practice under 6 Del. C. 2513; no private right of action
New HampshireRSA 507-HProfiling in furtherance of solely automated decisions that produce legal or similarly significant effects (507-H:4)Data protection assessment for profiling that presents a reasonably foreseeable risk of listed harms (507-H:8)January 1, 2025Attorney General exclusively; a violation is an unfair or deceptive act under RSA 358-A:2; no private right of action
New JerseyP.L.2023, c.266, N.J.S.A. 56:8-166.4 et seq.Profiling in furtherance of decisions that produce legal or similarly significant effects concerning the consumer (56:8-166.10); consent required for that profiling where the controller knows, or willfully disregards, that the consumer is at least 13 but younger than 17 (56:8-166.12)Data protection assessment for profiling that presents a reasonably foreseeable risk of listed harms, available to the Division of Consumer Affairs on request (56:8-166.12)January 15, 2025 (the 365th day after approval on January 16, 2024)Attorney General, with sole and exclusive authority; a violation is an unlawful practice under the Consumer Fraud Act; no private right of action
MinnesotaConsumer Data Privacy Act, Minn. Stat. 325M.10 to 325M.21Profiling in furtherance of automated decisions that produce legal or similarly significant effects (325M.14, subd. 1(f)); plus a right to question the result (subd. 1(g))Data privacy and protection assessment for profiling that presents a reasonably foreseeable risk of listed harms, including disparate impact (325M.18)July 31, 2025 (postsecondary institutions regulated by the Office of Higher Education: July 31, 2029)Attorney General; injunction and civil penalties of up to $7,500 per violation; no private right of action
CaliforniaCCPA regulations, 11 CCR 7150 to 7157 and 7200 to 7222No statutory profiling opt-out; an ADMT opt-out for significant decisions from January 1, 2027, unless an exception such as a human appeal applies (7221)Risk assessment before processing that presents significant risk to consumers’ privacy, including ADMT for a significant decision (7150)Regulations effective January 1, 2026; ADMT duties January 1, 2027California Privacy Protection Agency; Attorney General

Each law applies only to controllers that meet its own applicability thresholds, and each exempts listed entities and data; this table does not list them. Other states’ comprehensive privacy laws were not checked for this page.

What does the right to opt out of profiling cover?

The opt-out reaches profiling used for a defined class of decisions, not profiling in general. Virginia defines decisions that produce legal or similarly significant effects as decisions made by the controller that result in the provision or denial by the controller of financial and lending services, housing, insurance, education enrollment, criminal justice, employment opportunities, health care services, or access to basic necessities, such as food and water (59.1-575). The other nine state laws use nearly identical lists.

The laws split on human involvement. New Hampshire covers profiling in furtherance of “solely automated” decisions. Connecticut used that wording until June 30, 2026, then moved to “any automated decision,” and Delaware will make the same move on January 1, 2027. Virginia, Texas, Oregon, New Jersey, and Colorado’s statute contain no “automated decision” qualifier; Montana and Minnesota say “automated decisions.”

Colorado’s rules sort profiling into three kinds: Solely Automated Processing, with no human review, oversight, involvement, or intervention; Human Reviewed Automated Processing, where a human reviews the processing without meaningful consideration; and Human Involved Automated Processing, where a human engages in a meaningful consideration of available data used in the processing or any output of it and has the authority to change or influence the outcome. A controller must honor opt-outs from the first two kinds (Rule 9.04(B)). It may decline an opt-out from the third, but must then tell the consumer the decision subject to the profiling, the categories of personal data used, a plain-language explanation of the logic and of the role of meaningful human involvement, how profiling is used in the decision, its benefits and potential consequences, and how to correct or delete the data used (Rule 9.04(C)).

The opt-out does not reach employees: each of the ten laws defines a consumer to exclude a person acting in a commercial or employment context. California is different: its ADMT rules treat hiring, allocation or assignment of work, compensation, promotion, demotion, suspension, and termination as significant decisions.

Which states let a consumer question a profiling decision?

Minnesota and Connecticut go further than an opt-out. In Minnesota, if a consumer’s personal data is profiled in furtherance of decisions that produce legal effects or similarly significant effects concerning a consumer, the consumer has the right to question the result of the profiling, to be informed of the reason that the profiling resulted in the decision, and, if feasible, to be informed of what actions the consumer might have taken to secure a different decision and might take to secure a different decision in the future. The consumer may review the personal data used, and if the decision was based on inaccurate personal data, have the data corrected and the decision reevaluated (325M.14, subd. 1(g)).

Connecticut’s version, in effect since July 1, 2026, applies where personal data were processed for profiling in furtherance of any automated decision that produced any legal or similarly significant effect. If feasible, the consumer may question the result, be informed of the reason, and review the personal data processed; correction and reevaluation are available only if the decision concerned housing (42-518(a)(6)). The access right now also covers inferences derived from personal data and whether the data is being processed for profiling to make such a decision, unless a trade secret or another listed limit applies (42-518(a)(1)).

When must a business assess AI profiling?

Apart from California, every law in the table requires a documented assessment of risky profiling, in nearly the same words, though Delaware’s applies only above its consumer threshold. For example, under C.R.S. 6-1-1309, a controller covered by the Colorado Privacy Act must not conduct processing that presents a heightened risk of harm to a consumer, including profiling that presents a reasonably foreseeable risk of unfair or deceptive treatment of, or unlawful disparate impact on, consumers, without conducting and documenting a data protection assessment, which it must make available to the Attorney General on request. Virginia requires an assessment of each listed processing activity, including targeted advertising, the sale of personal data, and the processing of sensitive data (59.1-580(A)).

All ten state laws require the same analysis: identify and weigh the benefits that may flow, directly and indirectly, from the processing to the controller, the consumer, other stakeholders, and the public against the potential risks to the rights of the consumer, as mitigated by safeguards the controller can employ, factoring in de-identified data, consumers’ reasonable expectations, the context of the processing, and the relationship between controller and consumer. An assessment done for another law counts if it has a reasonably comparable scope and effect (Virginia 59.1-580(F)).

What a profiling assessment must contain, by law, as of November 16, 2026
LawTriggerRequired contents beyond the benefit-risk weighingTiming
Colorado, 4 CCR 904-3, Rules 8.05 and 9.06Profiling that presents a reasonably foreseeable risk of the 6-1-1309(2)(a) harms, whether solely automated, human reviewed, or human involvedThe Rule 8.04 elements plus, as applicable: the personal data used; the decision; the benefits of automated over manual processing; why the profiling relates to the controller’s goods and services; the training data and logic, including statistics; for third-party software, its name and evaluations of accuracy and reliability; the outputs and how they are used; human involvement; fairness and disparate impact testing and results; and safeguardsBefore the processing; for profiling-based decisions with legal or similarly significant effects, reviewed at least annually with an updated fairness and disparate impact evaluation; kept for at least three years after the processing ends
Connecticut, 42-522(c)Profiling to make a decision that produces any legal or similarly significant effectImpact assessment, to the extent reasonably known or available: purpose, intended uses, deployment context, and benefits; known or reasonably foreseeable heightened risks and mitigation; input data categories and outputs; data used to customize the profiling; performance metrics and known limitations; transparency measures; post-deployment monitoring and user safeguardsProcessing activities created or generated on or after August 1, 2026
Delaware, 12D-108(a)(2)Profiling in furtherance of automated decisions that produce legal or similarly significant effects, by a controller processing the data of at least 50,000 consumersAn impact assessment with the same seven categories of content as Connecticut’s, conducted and documented on a regular basisFrom January 1, 2027
Minnesota, 325M.18Profiling that presents a reasonably foreseeable risk of listed harms, including disparate impactThe type of personal data, including the extent to which it is sensitive data, and the context; the controller’s written data privacy policies and proceduresRequired since July 31, 2025
California, 11 CCR 7150 to 7157Processing that presents significant risk to consumers’ privacy, including ADMT for a significant decision and certain training usesNine elements in 7152(a), including specific purpose, data categories, ADMT logic and output, benefits, negative impacts, safeguards, and whether to proceedBefore the processing; by December 31, 2027 for covered processing under way before January 1, 2026; attestations and summaries for 2026 and 2027 assessments due April 1, 2028

The AI impact assessment guide compares California’s risk assessment with the EU AI Act’s fundamental rights impact assessment and the GDPR’s data protection impact assessment.

What do California’s ADMT rules add?

The California Privacy Protection Agency’s regulations, approved by the Office of Administrative Law and announced September 23, 2025, took effect January 1, 2026. Article 11 applies to a business that uses ADMT to make a significant decision about a consumer; a business already doing so must comply by January 1, 2027, and one that starts on or after that date must comply from the start (11 CCR 7200).

ADMT means technology that processes personal information and uses computation to replace or substantially replace human decisionmaking. “Substantially replace” means the output is used without human involvement, and human involvement requires a reviewer who knows how to interpret the output, reviews it with other relevant information, and has authority to make or change the decision (11 CCR 7001(e)). Advertising is not a significant decision. Three duties follow: a pre-use notice (7220), a right to opt out with exceptions such as a human appeal to a reviewer with authority to overturn the decision (7221), and a right to access a plain-language explanation of the ADMT’s purpose, logic, and outcome (7222). Trade secrets and information that would compromise security, fraud prevention, or safety need not be disclosed. The pre-use notice must state the specific purpose; a generic purpose such as “to make a significant decision” does not satisfy the rule.

The California AI laws guide covers the rules’ employment provisions.

Do privacy laws treat biometric and neural data as sensitive?

Yes. Each of the ten state laws treats biometric data used for identification as sensitive data, which needs consent. Virginia bars a controller from processing sensitive data concerning a consumer without obtaining the consumer’s consent, and its definition includes the processing of genetic or biometric data for the purpose of uniquely identifying a natural person (59.1-575, 59.1-578(A)(5)); Texas has the same consent rule (541.101(b)(4)). Oregon’s biometric data definition excludes photographs and audio or video recordings, and facial mapping or facial geometry, unless generated or used to identify a specific consumer (646A.570).

Several laws now name neural data:

  • California: sensitive personal information includes a consumer’s neural data, meaning information that is generated by measuring the activity of a consumer’s central or peripheral nervous system, and that is not inferred from nonneural information (11 CCR 7001(bbb)).
  • Colorado: HB 24-1058 added “biological data,” which includes neural data used or intended to be used for identification purposes, to sensitive data.
  • Connecticut: since July 1, 2026, sensitive data includes neural data, defined as any information generated by measuring the activity of an individual’s central nervous system, and genetic or biometric data or information derived from them (42-515).
  • Delaware: from January 1, 2027, sensitive data will include neural data generated by measuring the activity of an individual’s central nervous system, and inferences used to reveal or identify sensitive categories (12D-102).

How do biometric privacy laws apply to AI?

Biometric privacy laws that reach AI, as of November 16, 2026
LawCore dutyAI-relevant limitsEnforcement
Illinois BIPA, 740 ILCS 14Written notice and a written release, which may be an electronic signature, before a private entity collects or otherwise obtains a biometric identifier or biometric information; a public retention and destruction schedule; no selling or profitingSince August 2, 2024, collecting the same identifier from the same person by the same method more than once in violation of Section 15(b) is a single violation, with at most one recovery; repeated disclosure of the same data from the same person to the same recipient using the same method of collection is likewise a single violation of Section 15(d)Private right of action: per violation, the greater of actual damages or liquidated damages of $1,000 (negligent) or $5,000 (intentional or reckless)
Texas, Bus. & Com. Code 503.001Inform the individual and receive consent before capturing a biometric identifier for a commercial purpose; destroy it within a reasonable time, but not later than the first anniversary of the date the purpose expiresSince January 1, 2026, an image on the Internet is not consent unless the individual made it public; the law does not apply to training, processing, or storing biometric identifiers to develop or offer AI models or systems unless a system is used or deployed for the purpose of uniquely identifying a specific individual, or to AI for listed security and fraud purposes; an identifier captured to train an AI system and later used for a commercial purpose not described in those exclusions becomes subject to the section’s possession and destruction provisions and its penalties (503.001(f))Attorney General; civil penalty of not more than $25,000 for each violation
Washington, RCW 19.375No enrolling a biometric identifier in a database for a commercial purpose without first providing notice, obtaining consent, or providing a mechanism to prevent the subsequent use of the identifier for a commercial purpose“Commercial purpose” means selling or disclosing the identifier to a third party to market goods or services unrelated to the initial transaction, and excludes security and law enforcement purposes; “biometric identifier” excludes a physical or digital photograph, video or audio recording, or data generated from them; no notice or consent needed for a security purpose. Separately, since March 31, 2024 (June 30, 2024 for small businesses), the My Health My Data Act (RCW 19.373) has treated biometric data, including imagery of the face and voice recordings from which an identifier template can be extracted, as consumer health data that may be collected only with consent or to the extent necessary to provide a product or service the consumer requested (19.373.030)RCW 19.375: Attorney General only, under the Consumer Protection Act. RCW 19.373: a violation is an unfair or deceptive act under the Consumer Protection Act (19.373.090), and a person injured in business or property may sue (RCW 19.86.090)
Colorado, C.R.S. 6-1-1314 (HB 24-1130)A written policy with a retention schedule, a security incident protocol, and deletion guidelines; before collecting, inform the consumer of the collection, the specific purpose, the retention period, and any disclosure to a processor; obtain consent before collecting biometric data (6-1-1314(4)(e))Applies from July 1, 2025 to a controller that controls or processes any amount of biometric identifiers or biometric data, even below the Privacy Act’s usual thresholds, for that data; an employer may require an employee’s consent as a condition of employment only for listed uses, such as access to secure locations and systems, recording the work day, and safety or security, and may not make consent to other uses a condition of employmentAttorney General and district attorneys, under the Colorado Privacy Act (6-1-1311); no private right of action

BIPA’s exclusions and the 2024 amendment are covered in the Illinois AI laws guide.

What has the FTC done about AI and personal data?

In the United States, the FTC has ordered companies to delete the models and algorithms, not only the data. Each order below settled allegations; none was a court finding of liability.

  • Everalbum (proposed January 11, 2021; finalized May 7, 2021). The order requires deletion of face embeddings derived from photos of users who did not give express consent and of any facial recognition models or algorithms developed with users’ photos or videos.
  • WW International and Kurbo (court order entered March 3, 2022; announced March 4, 2022). In a COPPA case the Department of Justice filed for the FTC, the court-approved order requires deletion of personal information collected from children under 13 through the Kurbo program unless a parent gave verifiable consent, destruction of any models or algorithms developed in whole or in part with children’s personal information collected through the program, and a $1.5 million penalty, and requires a retention schedule under which children’s personal information is deleted no more than one year after the user last tracked food, weight, or activity.
  • Rite Aid (complaint December 19, 2023; stipulated court order entered February 26, 2024). The FTC alleged the retailer deployed facial recognition without reasonable safeguards. The order bars Rite Aid from using facial recognition for security or surveillance purposes for five years and requires deleting the photos and videos of consumers collected for its facial recognition system and any data, models, or algorithms derived in whole or in part from them, and instructing third parties that received them to delete them too.

The AI data governance guide covers the dataset records that show what a deletion order would reach.

How does Colorado’s AI law relate to its privacy law?

They are separate laws. SB 26-189 will regulate ADMT in consequential decisions made on or after January 1, 2027. A federal court order in X.AI LLC v. Weiser bars the Attorney General from initiating enforcement of SB 26-189, including an investigation, for violations occurring on or before 14 days after the court rules on xAI’s preliminary injunction motion; as of November 15, 2026, that motion had not been filed. The order covers alleged violations of SB 24-205 or any legislation replacing or amending it enacted during the 2026 session. The Privacy Act’s profiling duties have applied since July 1, 2023. See the Colorado SB 26-189 guide.

What should a company using AI on personal data do now?

  • Map every model that uses personal data to evaluate, analyze, or predict personal aspects, and flag those that feed decisions with legal or similarly significant effects
  • For each flagged model, decide state by state whether the profiling opt-out applies, including whether human review is meaningful enough to fall outside New Hampshire’s solely automated test or to decline Colorado’s opt-out with the Rule 9.04(C) disclosures
  • Describe the opt-out in the privacy notice and build a decision path that does not rely on the profiling for consumers who opt out
  • For Minnesota and Connecticut residents, prepare to explain the reason for a profiling-based decision, show the personal data used, say in Minnesota, if feasible, what could secure a different decision, and correct and reevaluate where the law requires
  • Complete and date a data protection assessment before profiling that presents a reasonably foreseeable risk of the listed harms; for Colorado, add the Rule 9.06 contents and any Rule 8.05(C) annual review
  • Run the Connecticut impact assessment for processing activities created or generated on or after August 1, 2026; calendar the Delaware changes and California’s ADMT duties for January 1, 2027, and California’s risk assessment submissions for April 1, 2028
  • Get consent before processing sensitive data, including biometric data used for identification and, where the law names it, neural data
  • Before collecting faces or voices, meet each biometric law’s notice, consent or release, retention, and destruction rules

How this fits the other guides: this page is a spoke of the US state AI laws hub. The California and Colorado guides cover each state’s AI-specific law, and the GDPR guide covers EU profiling rules.

This guide is general information as of November 16, 2026, not legal advice; confirm the current text with the state legislature, the Attorney General or privacy agency, or counsel before acting on it.

Common questions

Can consumers opt out of AI profiling under state privacy laws?

Yes, in the states this guide checked. Virginia, Colorado, Connecticut, Texas, Oregon, Montana, Delaware, New Hampshire, New Jersey, and Minnesota give residents a right to opt out of the processing of personal data for profiling in furtherance of decisions that produce legal or similarly significant effects, such as decisions on lending, housing, insurance, education, employment, or health care; New Hampshire, and Delaware until January 1, 2027, limit it to solely automated decisions, and Colorado's rules let a controller decline the opt-out for Human Involved Automated Processing if it gives the consumer listed information. All ten define a consumer to exclude a person acting in a commercial or employment context, and each law applies only to controllers that meet its thresholds, with listed exemptions.

Do state privacy laws require an assessment before using AI to profile people?

Each of the ten state laws in this guide requires a controller (in Delaware, one that processes the data of at least 100,000 consumers) to conduct and document a data protection assessment of profiling that presents a reasonably foreseeable risk of listed harms, such as unfair or deceptive treatment of, or unlawful disparate impact on, consumers, and to make it available to the Attorney General or, in New Jersey, the Division of Consumer Affairs on request. Colorado's rules add required contents and an annual review for profiling used in decisions with legal or similarly significant effects. Connecticut requires an impact assessment of profiling for such decisions for processing activities created or generated on or after August 1, 2026, and Delaware will require one from January 1, 2027 of controllers that process the data of at least 50,000 consumers.

Which states let consumers question an automated profiling decision?

Minnesota, since July 31, 2025, and Connecticut, since July 1, 2026. In Minnesota, a consumer whose personal data is profiled in furtherance of decisions that produce legal or similarly significant effects may question the result, be told the reason, if feasible learn what actions might have secured, or might secure, a different decision, review the data used, and have inaccurate data corrected and the decision reevaluated. In Connecticut, if feasible, the consumer may question the result, be told the reason, and review the data used, with correction and reevaluation for housing decisions.

By industry: all 8 industry pages

Sources

  1. Code of Virginia 59.1-575, Definitions, Virginia General Assembly, accessed November 16, 2026.
  2. Code of Virginia 59.1-577, Personal data rights; consumers, Virginia General Assembly, accessed November 16, 2026.
  3. Code of Virginia 59.1-578, Data controller responsibilities; transparency, Virginia General Assembly, accessed November 16, 2026.
  4. Code of Virginia 59.1-580, Data protection assessments, Virginia General Assembly, accessed November 16, 2026.
  5. Code of Virginia 59.1-584, Enforcement; civil penalty; expenses, Virginia General Assembly, accessed November 16, 2026.
  6. Acts of Assembly 2021 Special Session I, Chapter 35 (HB 2307), Virginia Legislative Information System, accessed November 16, 2026.
  7. SB 21-190, Colorado Privacy Act (signed act), C.R.S. 6-1-1303, 6-1-1306, 6-1-1309, 6-1-1311, Colorado General Assembly, 2021, effective July 1, 2023.
  8. HB 24-1130, Privacy of Biometric Identifiers and Biometric Data (signed act), C.R.S. 6-1-1314, Colorado General Assembly, 2024, effective July 1, 2025.
  9. HB 24-1058, Protect Privacy of Biological Data (signed act), Colorado General Assembly, 2024.
  10. SB 26-189, Session Laws 2026, Chapter 131 (enrolled act), Colorado General Assembly, approved May 14, 2026.
  11. X.AI LLC v. Weiser, No. 1:26-cv-01515 (D. Colo.), docket, minute order ECF No. 24 (April 27, 2026) and ECF No. 25 (October 2, 2026), CourtListener RECAP, accessed November 15, 2026.
  12. 4 CCR 904-3, Colorado Privacy Act Rules, Rules 8.05, 9.02 to 9.06 (PDF), Colorado Department of Law, version effective December 1, 2025, via the Colorado Secretary of State, accessed November 16, 2026.
  13. Conn. Gen. Stat. Chapter 743jj, Data Privacy and Security, 2026 Supplement (42-515, 42-518, 42-522), Connecticut General Assembly, accessed November 16, 2026.
  14. Conn. Gen. Stat. Chapter 743jj, revised to January 1, 2025 (42-525), Connecticut General Assembly, accessed November 16, 2026.
  15. Public Act 25-113 (SB 1295) (PDF), Connecticut General Assembly, approved June 24, 2025.
  16. Public Act 26-64 (SB 4) (PDF), sections 12 and 13 (42-515, 42-518(a)), effective October 1, 2026, Connecticut General Assembly, approved May 27, 2026.
  17. Texas Business and Commerce Code Chapter 541, Consumer Data Protection, Texas Legislature, accessed November 16, 2026.
  18. H.B. No. 4 (enrolled), 88th Legislature, Texas Legislature, 2023.
  19. Texas Business and Commerce Code 503.001, Capture or Use of Biometric Identifier, Texas Legislature, accessed November 16, 2026.
  20. H.B. No. 149 (enrolled), 89th Legislature, Section 2 (amending 503.001) and Section 10, Texas Legislature, 2025, effective January 1, 2026.
  21. ORS Chapter 646A (2025 Edition), 646A.570 to 646A.589, Oregon Legislature, accessed November 16, 2026.
  22. Consumer Privacy, Oregon Department of Justice, accessed November 16, 2026.
  23. Enrolled House Bill 2008 (2025 Or. Laws ch. 251), amending ORS 646A.578 (PDF), Oregon Legislature, 2025 (no effective-date clause).
  24. ORS 171.022, Effective date for Act of Legislative Assembly (January 1 of the year after passage), Oregon Legislature, accessed November 16, 2026.
  25. MCA 30-14-2808, Consumer personal data, opt-out, compliance, appeals, Montana Legislature, Montana Code Annotated 2025.
  26. MCA 30-14-2814, Data protection assessment, Montana Legislature, Montana Code Annotated 2025.
  27. MCA 30-14-2817, Enforcement, Montana Legislature, Montana Code Annotated 2025.
  28. SB 384 memo with the text of the Consumer Data Privacy Act (PDF), Montana Legislative Services, Economic Affairs Interim Committee, May 2024.
  29. Delaware Code Title 6, Chapter 12D, Delaware Personal Data Privacy Act, Delaware General Assembly, accessed November 16, 2026.
  30. Personal Data Privacy Portal: Business, Delaware Department of Justice, accessed November 16, 2026.
  31. RSA Chapter 507-H, Expectation of Privacy, New Hampshire General Court, accessed November 16, 2026.
  32. P.L.2023, c.266 (S332) (PDF), New Jersey Legislature, approved January 16, 2024.
  33. Minn. Stat. 325M.14, Consumer personal data rights, Minnesota Office of the Revisor of Statutes, accessed November 16, 2026.
  34. Minn. Stat. 325M.18, Data privacy policies; data privacy and protection assessments, Minnesota Office of the Revisor of Statutes, accessed November 16, 2026.
  35. Minn. Stat. 325M.20, Attorney General enforcement, Minnesota Office of the Revisor of Statutes, accessed November 16, 2026.
  36. Laws of Minnesota 2024, chapter 121, article 5, section 14, Minnesota Office of the Revisor of Statutes, signed May 24, 2024.
  37. California Finalizes Regulations to Strengthen Consumers’ Privacy, California Privacy Protection Agency, September 23, 2025.
  38. California Consumer Privacy Act Regulations, 11 CCR 7000 et seq., effective January 1, 2026 (PDF), California Privacy Protection Agency, accessed November 16, 2026.
  39. Biometric Information Privacy Act (740 ILCS 14), Illinois General Assembly, accessed November 12, 2026.
  40. Public Act 103-0769 (SB 2979), Illinois General Assembly, approved and effective August 2, 2024.
  41. RCW Chapter 19.375, Biometric Identifiers, Washington State Legislature, accessed November 16, 2026.
  42. RCW Chapter 19.373, Washington My Health My Data Act (19.373.010, 19.373.030, 19.373.090), Washington State Legislature, accessed November 16, 2026.
  43. RCW 19.86.090, Civil action for damages, Washington State Legislature, accessed November 16, 2026.
  44. California Company Settles FTC Allegations It Deceived Consumers about use of Facial Recognition in Photo Storage App, Federal Trade Commission, January 11, 2021.
  45. FTC Finalizes Settlement with Photo App Developer Related to Misuse of Facial Recognition Technology, Federal Trade Commission, May 7, 2021.
  46. FTC Takes Action Against Company Formerly Known as Weight Watchers for Illegally Collecting Kids’ Sensitive Health Data, Federal Trade Commission, March 4, 2022.
  47. United States v. Kurbo, Inc., No. 3:22-cv-00946 (N.D. Cal.), Stipulated Order for Permanent Injunction, Civil Penalty Judgment, and Other Relief, ECF No. 15 (March 3, 2022), CourtListener (RECAP), accessed October 26, 2026.
  48. Rite Aid Banned from Using AI Facial Recognition After FTC Says Retailer Deployed Technology without Reasonable Safeguards, Federal Trade Commission, December 19, 2023.
  49. FTC v. Rite Aid Corp., No. 2:23-cv-05023 (E.D. Pa.), Stipulated Order for Permanent Injunction and Other Relief, ECF No. 19 (entered February 26, 2024), CourtListener (RECAP), accessed October 26, 2026.