Tennessee AI Regulations
Tennessee has not enacted a comprehensive AI law and did not attempt one in 2026. It regulates specific harms instead. The ELVIS Act made Tennessee the first state to protect voice against AI cloning, and it is enforceable by individuals and by their exclusive licensees. SB 1580 bars marketing an AI system as a qualified mental health professional. The Transparency for Deepfakes in Political Advertising Act took effect July 1, 2026. There is no Tennessee duty of care, no algorithmic impact assessment mandate, and no general AI disclosure requirement.
Last reviewed against primary sources: . AI legislation is moving quickly — statutes have been delayed, repealed and replaced mid-cycle. Verify against the linked primary sources before relying on any date here.
Where we work in Tennessee
TrustEdge serves organizations across Tennessee, including these markets.
In-Depth Analysis
Tennessee is a useful counterexample to the assumption that a state either has an AI law or does not. It has several, none of them comprehensive, and the most consequential one is not framed as AI regulation at all.
The ELVIS Act is a right-of-publicity amendment. That framing matters, because right-of-publicity claims come with a plaintiff's bar, contingency economics, and — unusually for state AI law — standing that extends beyond the person depicted to their exclusive licensees. Colorado, Texas, and Utah all route enforcement through an Attorney General who must give notice and an opportunity to cure. Tennessee routes it through whoever holds the rights to a voice.
Be precise about who that reaches. For a record label, an ad agency, a film or game studio, a podcast network, or a company selling voice-cloning or synthetic-presenter software, the ELVIS Act is a live and serious exposure, and the tool-distribution theory means you can be liable without ever publishing a clip yourself. For a hospital, a bank, or a law firm running document summarization, coding assistants, claims triage, or retrieval over internal records, it is close to irrelevant. Those systems do not produce an identifiable person's voice or likeness.
The narrow band where it does reach a regulated-industry buyer is worth naming: cloning a named clinician's voice for patient outreach or after-hours messaging, generating a synthetic video of a named executive for training or marketing, or using a real employee's likeness in AI-generated content after they leave. Written consent — specific, and surviving employment — closes that gap. It is a contracting problem, not an architecture problem.
SB 1580 is narrower still than its headlines. It prohibits a marketing claim. It does not restrict a licensed clinician from using AI tools, does not require disclosure that a patient is interacting with AI, and does not impose any clinical validation duty. If your product markets itself as therapy-adjacent, read your own copy against the Title 33 definition of qualified mental health professional. If it does not, the statute does not apply to you.
What 2026 tells you about direction is arguably more useful than what it enacted. HB 1898 would have imposed frontier-developer transparency, child-safety plans, and 15-day incident reporting to the Attorney General. It passed the House unanimously and then died. The legislature sent the question to TACIR instead, with recommendations due January 31, 2027 — squarely before the 2027 session. Tennessee has deferred, not declined.
Practical Implications for SMBs & Healthcare
For most healthcare, financial services, legal, and government organizations, Tennessee imposes no AI-specific operating obligation today. The honest answer is that your Tennessee exposure is TIPA and your sector regulator, not an AI statute.
That changes if you generate synthetic voice or likeness of identifiable people. Then the ELVIS Act is the governing law, the plaintiff does not need to wait for a regulator, and there is no cure period.
TIPA is the requirement more Tennessee businesses miss. Profiling that produces legal or similarly significant effects carries an opt-out, and profiling that presents a heightened risk requires a documented data protection assessment. HIPAA-covered entities and GLBA financial institutions are exempt at the entity level, which removes a large share of regulated buyers from scope — confirm that you actually qualify rather than assuming it.
If you are planning past this year, watch the TACIR report due January 31, 2027 and the shape of HB 1898. A future Tennessee AI law is more likely to look like incident reporting and published safety plans than like Colorado's impact-assessment model.
Key Provisions
ELVIS Act — voice is now a property right, and simulated voice counts
Public Chapter 588 (2024) amended Tennessee's Protection of Personal Rights statute at Tenn. Code Ann. § 47-25-1101 et seq. to add voice alongside name, photograph, and likeness as a property right. Voice is defined as a sound in a medium readily identifiable and attributable to a particular individual, whether the sound contains that individual's actual voice or a simulation of it. That definition is what brings synthetic audio inside a right-of-publicity statute written in 1984 for the Elvis Presley estate.
Three separate liability theories, only one of which requires an advertisement
First, unauthorized use of a name, photograph, voice, or likeness for advertising or fundraising. Second — and this is the broad one — publishing, performing, distributing, transmitting, or otherwise making available to the public an individual's voice or likeness with knowledge that the use was not authorized. That theory has no commercial-use element. Third, distributing or making available an algorithm, software, tool, or other technology, service, or device whose primary purpose or function is producing a particular individual's photograph, voice, or likeness without authorization. The third theory reaches the vendor, not only the user, and there is no platform safe harbor.
Who can actually sue you under the ELVIS Act
The individual whose voice or likeness was used. Exclusive licensees of that right. A party holding an exclusive personal services contract with a recording artist, or exclusive distribution rights in that artist's sound recordings — meaning a label can sue without holding the publicity right itself. And heirs or the estate, for ten years after death, extendable indefinitely by continued commercial exploitation and terminated by two consecutive years of non-use. This is a well-resourced plaintiff class with existing counsel, which is why the ELVIS Act is the Tennessee statute most likely to produce litigation.
ELVIS Act remedies and exclusions
Injunctive relief, actual damages plus the violator's profits, and impoundment and destruction of infringing material. Treble damages and attorney's fees where the person depicted served in the armed forces. Violation is also a Class A misdemeanor. Excluded: news, public affairs, and sports broadcasts; comment, criticism, scholarship, satire, and parody; fleeting or incidental use; a person's own representation of themselves in an audiovisual work unless it creates a false impression of authenticity; and advertising for any of those uses.
SB 1580 — AI may not be marketed as a mental health professional
Public Chapter 647, signed April 6, 2026, effective July 1, 2026, adding a section at Tenn. Code Ann. § 33-1-205. A person who develops or deploys an AI system may not advertise or represent to the public that the system is, or is able to act as, a qualified mental health professional — a term already defined in Title 33 to cover specified licensed psychiatrists, psychologists, psychological examiners, social workers, and marital and family therapists. The statute regulates the claim, not the tool. A licensed clinician using AI in practice is not restricted by it.
Transparency for Deepfakes in Political Advertising Act
Public Chapter 625, signed March 30, 2026, effective July 1, 2026, amending Title 2, Chapter 19. Political advertisements containing a deepfake that impersonates or depicts a candidate saying or doing something the candidate did not must carry a disclaimer — at the beginning and end for audio, and displayed in large letters for the full duration for video. It covers audio and video only, not still images such as printed mailers. Violation is a Class C misdemeanor, and the depicted candidate is entitled to damages and equitable relief. Broadcasters, cable and satellite operators, and streaming services are not liable for airing paid material they did not create. The Act does not prohibit campaign use of AI; it requires labeling.
Preventing Deepfake Images Act
Public Chapter 466, signed May 9, 2025, effective July 1, 2025. Creates civil and criminal liability for disclosing an intimate digital depiction of an identifiable person without consent, and makes clear that consent to creation is not consent to disclosure. Civil remedies include actual damages, attorney's fees, injunctive relief, and liquidated damages up to $150,000. Criminal exposure ranges from a Class E to a Class C felony depending on the circumstances.
What Tennessee did not do in 2026
HB 1898 / SB 2171, the Artificial Intelligence Public Safety and Child Protection Transparency Act, would have required frontier developers and large chatbot providers to publish child-safety plans, report safety incidents to the Attorney General within 15 days, and face civil penalties up to $50,000 per violation. It passed the House 94-0 on April 16, 2026 and did not become law. The legislature chose to study instead, and did it twice. Public Chapter 1066 (SB 1493, signed May 27, 2026) directs the Tennessee Advisory Commission on Intergovernmental Relations to review AI risks and regulatory approaches and report findings and legislative recommendations by January 31, 2027. Public Chapter 1082 (SB 1700, signed the same day) is the CHAT Act as enacted: introduced as a criminal-penalty bill, stripped of every felony provision and civil remedy before passage, and reduced to a second TACIR study covering AI systems and generative AI chatbots. Amendment 2 removed its report deadline, so unlike Public Chapter 1066 it carries none. Anyone citing "the Tennessee CHAT Act" as an enacted chatbot-safety law is describing the introduced bill, not the statute. Tennessee also has no general chatbot-disclosure law, no AI-in-employment statute, and no AI-in-insurance or utilization-review statute.
State government AI use is governed by policy, not by statute
The Tennessee Artificial Intelligence Advisory Council was created by Public Chapter 988 (2024) — 24 members, co-chaired by the Commissioner of Finance and Administration and the State Chief Information Officer, with annual reports running through a final action plan due December 31, 2028 and a sunset the same day. It is advisory and imposes nothing on private business. Separately, the State operates Enterprise Artificial Intelligence Policy 200-POL-007 governing AI use across executive-branch departments and agencies. If you sell AI into Tennessee state government, that policy is the document your procurement answers will be measured against.
TIPA is the adjacent obligation most Tennessee businesses actually have
The Tennessee Information Protection Act, Public Chapter 408 (2023), took effect July 1, 2025. It gives consumers a right to opt out of profiling in furtherance of decisions that produce legal or similarly significant effects, and requires data protection assessments for targeted advertising, sale of personal data, sensitive data processing, and profiling that presents a heightened risk. It applies to businesses exceeding $25 million in revenue that also meet consumer-volume thresholds, and it exempts HIPAA-covered entities and GLBA-regulated financial institutions at the entity level. Its distinguishing feature is an affirmative defense for maintaining a written privacy program that reasonably conforms to the NIST Privacy Framework — the first state law to build a NIST framework into its enforcement structure.
Bills & Statutes
ELVIS Act (Ensuring Likeness, Voice, and Image Security Act of 2024)
Public Chapter 588; effective July 1, 2024; codified at Tenn. Code Ann. § 47-25-1101 et seq.
Read the primary source(opens in a new tab)Prohibition on representing an AI system as a qualified mental health professional
Public Chapter 647; signed April 6, 2026; effective July 1, 2026
Read the primary source(opens in a new tab)Transparency for Deepfakes in Political Advertising Act
Public Chapter 625; signed March 30, 2026; effective July 1, 2026
Read the primary source(opens in a new tab)Preventing Deepfake Images Act
Public Chapter 466; signed May 9, 2025; effective July 1, 2025
Read the primary source(opens in a new tab)TACIR study of artificial intelligence risks and regulatory approaches
Public Chapter 1066; signed May 27, 2026; report due to the Governor and legislative leadership January 31, 2027
Read the primary source(opens in a new tab)CHAT Act as enacted — second TACIR study, covering AI systems and generative AI chatbots
Public Chapter 1082; signed May 27, 2026. Introduced with criminal penalties, all of which were amended out before passage. Amendment 2 removed the January 31, 2027 report deadline, so the enacted act carries none
Read the primary source(opens in a new tab)Artificial Intelligence Public Safety and Child Protection Transparency Act
Did not become law. Passed the House 94-0 on April 16, 2026; no further action
Read the primary source(opens in a new tab)Tennessee Artificial Intelligence Advisory Council
Public Chapter 988 (2024); advisory only; council sunsets December 31, 2028
Read the primary source(opens in a new tab)State of Tennessee Enterprise Artificial Intelligence Policy
Executive-branch policy governing AI use by state departments and agencies
Read the primary source(opens in a new tab)Tennessee Information Protection Act
Public Chapter 408 (2023); effective July 1, 2025
Read the primary source(opens in a new tab)Applicability & Enforcement
Who It Applies To
The ELVIS Act applies to anyone who uses, publishes, or distributes an identifiable individual's voice or likeness without authorization, and to anyone who supplies a tool whose primary purpose is producing a particular individual's voice or likeness. In practice that is media, entertainment, advertising, agencies, and synthetic-voice and avatar vendors — not clinical or back-office AI. SB 1580 applies to any developer or deployer that markets an AI system as a qualified mental health professional. The political deepfake law applies only to political advertising. TIPA applies to businesses over $25 million in revenue meeting consumer-volume thresholds, and exempts HIPAA-covered entities and GLBA financial institutions at the entity level.
Enforcement
There is no single Tennessee AI regulator. The ELVIS Act is enforced by private civil action brought by the individual, an exclusive licensee, a label holding exclusive personal-services or distribution rights, or the estate — plus Class A misdemeanor prosecution. SB 1580 is enforced as an unfair or deceptive act under the Tennessee Consumer Protection Act of 1977, which carries both Attorney General enforcement and a private right of action. The political deepfake law is enforced by the depicted candidate and by Class C misdemeanor prosecution. TIPA is enforced exclusively by the Attorney General, with a 60-day right to cure and no private right of action.
Penalties
ELVIS Act: injunction, actual damages plus the violator's profits, impoundment and destruction of infringing material, treble damages and attorney's fees for armed-forces members, and Class A misdemeanor. SB 1580: civil penalty up to $5,000 per violation under the Consumer Protection Act, plus private-action damages. Preventing Deepfake Images Act: liquidated damages up to $150,000, attorney's fees, injunctive relief, and Class E to Class C felony exposure. Political deepfake law: damages and equitable relief to the candidate, and Class C misdemeanor.
Recommended Compliance Actions
Steps organizations should consider to prepare for and comply with Tennessee's AI regulations.
- 1
Inventory every place your organization generates or distributes synthetic voice, video, or likeness of an identifiable real person — marketing, training content, patient or customer outreach, IVR, and demo material. That inventory is the entire ELVIS Act question for most non-media businesses.
- 2
Obtain specific written authorization for voice and likeness from employees, clinicians, and executives whose synthetic representations you create, and make the consent survive their departure. Consent is the defense; there is no cure period once a claim is filed.
- 3
If you license a voice-cloning, avatar, or synthetic-presenter vendor, get contractual representations on authorization and rights clearance, plus indemnity. The Act creates liability for distributing the tool as well as for using the output.
- 4
Audit your own marketing copy against Tenn. Code Ann. § 33-1-205 if any product touches mental or behavioral health. The prohibition is on the representation, so this is a review of claims, not of the model.
- 5
Determine whether TIPA applies to you and whether you are exempt at the entity level as a HIPAA-covered entity or GLBA financial institution. If it applies, stand up data protection assessments for profiling that presents a heightened risk and honor the profiling opt-out.
- 6
Consider a written privacy program conforming to the NIST Privacy Framework. Tennessee is the only state that turns that into an affirmative defense.
- 7
If you sell AI into Tennessee state government, read Enterprise AI Policy 200-POL-007 before your next procurement response rather than after.
- 8
Calendar the TACIR report due January 31, 2027 and re-read HB 1898 as the template. Build incident-detection and safety-documentation capability now if you operate consumer-facing chatbots.
Related TrustEdge Services
As of September 14, 2026. We are not lawyers and this is not legal advice. It is the responsibility of consumers of this data that they verify the applicability and current ratified statutes and legal precedence with a qualified attorney licensed in the state or country they are researching.
Need Help with AI Compliance?
Our team helps organizations build compliance-first AI systems that meet current and emerging regulatory requirements.