The Florida Senate
BILL ANALYSIS AND FISCAL IMPACT STATEMENT
(This document is based on the provisions contained in the legislation as of the latest date listed below.)
Prepared By: The Professional Staff of the Committee on Commerce and Tourism
BILL: SB 702
INTRODUCER: Senator Burgess
SUBJECT: Provenance of Digital Content
DATE: March 7, 2025
ANALYST STAFF DIRECTOR REFERENCE ACTION
1. McMillan McKay CM Pre-meeting
2. ATD
3. FP
I. Summary:
SB 702 creates a number of provisions relating to information about the origin and modification
of digital content. The bill defines “provenance data” as information that records the origin of a
piece of visual or audio digital content and the history of modifications to such content which is
in a format that is compliant with widely adopted guidelines or specifications promulgated by an
established standard-setting body. The term includes, but is not limited to, information
identifying whether some or all of the content has been derived through generative artificial
intelligence and, if so, the name of the generative artificial intelligence tool used to generate such
content and the organization that developed such tool.
The bill requires any content regarding an election or purporting to feature a candidate which can
be viewed, heard, or accessed online to carry digital provenance data.
The bill creates a digital content provenance pilot program within the Division of Emergency
Management (DEM), to enhance the security and authenticity of digital content used in
emergency management operations through the inclusion of provenance data.
The bill requires provenance data to be included on any synthetic data wholly created by a
generative artificial intelligence tool. The provider of such tool must make the following
available to the public:
• An application tool; and
• A free provenance reader.
The bill requires a social media platform to retain all available provenance data of visual or audio
digital content provided to or posted on such platform and make such data available to users of
the platform through a conspicuous indicator. Additionally, a capture device sold in Florida is
required to allow an option to include provenance data on any visual or audio digital content
recorded with such device. The manufacturer of a capture device must ensure that such
REVISED: BILL: SB 702 Page 2
provenance data can be read by third-party applications. A violation of s. 501.9741, F.S.,
constitutes an unfair or deceptive act or practice.
The bill takes effect July 1, 2025.
II. Present Situation:
Artificial Intelligence
Generally
Artificial intelligence (AI) is the development of computer systems to perform tasks that
normally require human intelligence, such as learning and decision-making.
1
It enables computer
systems to receive information that is either provided to them by others or gathered by them (e.g.
through camera lenses or other sensors), which they can then process and respond to in some
meaningful way. To a certain extent, AI systems can adapt their behavior by analyzing the
effects of previous actions and working autonomously.
2
Investments in AI have led to many of the transformative advancements that U.S. consumers rely
upon every day,
3
including mapping technologies, voice-assisted smartphones, handwriting
recognition for mail delivery, financial trading, smart logistics, spam filtering, and language
translation. AI advances have also provided significant social benefits in areas such as precision
medicine, environmental sustainability, education, and public welfare.
4
Types of AI
AI may be generally classified in one of three classes based on its capabilities or its
functionalities:
5
• Artificial Narrow AI. Also known as Weak AI, machines using Weak AI can only perform
specific tasks using human-like capabilities. They can do nothing more than what they are
programmed to do. Examples of Artificial Narrow AI include Siri, Alexa, and ChatGPT.
6
• General AI. Also known as Strong AI, and any machine or application using Strong AI in the
future would be able to use what they have learned in the past to accomplish new tasks in
different contexts without the need for additional training by human beings. In other words,
they would be able to learn, perceive, understand, and function completely like a human
beings.
7
1
National Conference of State Legislatures (NCSL), Artificial Intelligence 2023 Legislation, Jan. 12, 2024, available at
https://www.ncsl.org/technology-and-communication/artificial-intelligence-2023-legislation (last visited Mar. 7, 2025).
2
European Parliament, What is artificial intelligence and how is it used?, E.U. News, Jun. 20, 2023, available at
https://www.europarl.europa.eu/topics/en/article/20200827STO85804/what-is-artificial-intelligence-and-how-is-it-used (last
visited Mar. 7, 2025).
3
U.S. Department of State, Artificial Intelligence (AI), available at https://www.state.gov/artificial-intelligence/ (last visited
Mar. 7, 2025).
4
Id.
5
Naveen Joshi, 7 Types of Artificial Intelligence, Jun. 19, 2019, Forbes, available at
https://www.forbes.com/sites/cognitiveworld/2019/06/19/7-types-of-artificial-intelligence/?sh=7b5ddf4d233e (last visited
Mar. 7, 2025).
6
Id.
7
Id. BILL: SB 702 Page 3
• Super AI. Also known as artificial superintelligence, Super AI is strictly theoretical. If ever
realized, machines using Super AI would think, reason, learn, make judgments, and possess
cognitive abilities surpassing those of human beings. Machines possessing Super AI
capabilities would have evolved beyond the point of understanding human sentiments and
experiences to feeling emotions, having needs, and possessing beliefs and desires of their
own.
8
Under the umbrella of Artificial Narrow AI or Weak AI, there are four kinds of AI based upon
functionalities:
9
• Reactive Machine AI. Reactive machines are AI systems with no memory. They are designed
to perform very specific tasks. They can only work with presently available data because
they cannot recollect previous outcomes or decisions. Reactive Machine AI stems from
statistical math and can analyze vast amounts of data to produce a seemingly intelligent
output. Examples of machines and applications that rely upon Reactive Machine AI include
IBM Deep Blue (IBM’s chess-playing supercomputer) and the Netflix recommendation
engine.
10
• Limited Memory AI. In addition to having the capabilities of purely reactive machines,
Limited Memory AI machines and applications are also capable of learning from historical
data to make decisions. Almost all present-day Limited Memory AI applications, including
Generative AI tools (e.g. chatbots and virtual assistants) and self-driving vehicles, are
Limited Memory AI machines and applications.
11
• Theory of Mind AI. Theory of Mind AI is a kind of General AI that exists only in concept. It
is the “next level” of AI systems that researchers are currently developing. Machines and
applications using a Theory of Mind level AI will be able to understand the thoughts and
emotions of other entities. In theory, this will allow them to simulate humanlike relationships
and to contextualize artwork and essays, which today’s Generative AI tools are unable to
do.
12
• Self-Aware AI. Self-Aware AI is a kind of Super AI that exists only in concept. It is strictly
theoretical. If ever achieved, it will have the ability to understand its own internal conditions
and traits along with human emotions and thoughts. It will also have its own set of emotions,
needs, and beliefs.
13
Generative AI
Generative AI is a type of Limited Memory AI technology
14
that can produce high-quality
8
Id.
9
Id.
10
Id.
11
Id.
12
Id. Emotion AI is a kind of Theory of Mind AI that is currently under development. Researchers hope that it will one day
have the ability to analyze voices, images, and other kinds of data to recognize, simulate, monitor, and respond appropriately
to humans on an emotional level. To date, Emotion AI is unable to understand and respond to human feelings. Id.
13
Id.
14
George Lawton, What is generative AI? Everything you need to know, TechTarget, Jan. 2024, available at
https://www.techtarget.com/searchenterpriseai/definition/generative-AI (last visited Mar. 7, 2025). BILL: SB 702 Page 4
content, including text, images, audio, or video, within seconds when prompted by a user.
15
Although it was first introduced in the 1960s, it was not until 2014, with the introduction of
generative adversarial networks, or GANs (a type of machine learning algorithm),
16
that
Generative AI could convincingly create authentic images, videos, and audio of real people.
17
Generative AI systems learn patterns and relationships from massive amounts of data, which
enables them to process and create new content that may be similar, but not identical, to the
underlying training data. Such systems rely upon sophisticated machine learning algorithms and
statistical models to work.
18
In order to generate new content, Generative AI users are required to submit prompts that guide
the generation of new content. Many iterations may be required to produce the intended result
because Generative AI is sensitive to the wording of prompts.
19
Because Generative AI can do so much, it has many potential applications, including in
education, government, medicine, and law. Applications include:
• Writing a speech in a particular tone.
• Summarizing complex research.
• Assessing legal documents.
• Creating images for different applications.
• Composing music.
• Composing poems.
• Designing molecules for new drugs.
• Generating programming codes.
• Translating languages.
• Implementing chatbots.
• Deploying “deepfakes.”
• Improving dubbing for movies.
15
Government Accountability Office (GAO), Science, Technology Assessment, and Analytics, Science & Tech Spotlight:
Generative AI (June 2023), available at https://www.gao.gov/assets/gao-23-106782.pdf; (last visited Mar. 7, 2025). George
Lawton, What is generative AI? Everything you need to know, TechTarget, Jan. 2024, available at
https://www.techtarget.com/searchenterpriseai/definition/generative-AI (last visited Mar. 7, 2025).
16
“A generative adversarial network (GAN) is a deep learning architecture. It trains two neural networks to compete against
each other to generate more authentic new data from a given training dataset. For instance, you can generate new images
from an existing image database or original music from a database of songs. A GAN is called adversarial because it trains
two different networks and pits them against each other. One network generates new data by taking an input data sample and
modifying it as much as possible. The other network tries to predict whether the generated data output belongs in the original
dataset. In other words, the predicting network determines whether the generated data is fake or real. The system generates
newer, improved versions of fake data values until the predicting network can no longer distinguish fake from original.”
Amazon Web Services (AWS), What is a GAN?, available at https://aws.amazon.com/what-is/gan/ (last visited Mar. 7,
2025). GAN can generate images, training data for other models, complete missing information, and generate 3D models
from 2D data. Id.
17
George Lawton, What is generative AI? Everything you need to know, TechTarget, Jan. 2024, available at
https://www.techtarget.com/searchenterpriseai/definition/generative-AI (last visited Mar. 7, 2025).
18
Government Accountability Office (GAO), Science, Technology Assessment, and Analytics, Science & Tech Spotlight:
Generative AI (June 2023), available at https://www.gao.gov/assets/gao-23-106782.pdf (last visited Mar. 7, 2025). Training
data can include opensource information, such as text and images from the internet. Id.
19
Id. BILL: SB 702 Page 5
• Designing physical products and buildings.
20
The U.S. Government Accountability Office has identified several opportunities and challenges
in connection with the proliferation of Generative AI systems.
21
With respect to opportunities,
Generative AI can quicken access to ideas and knowledge by helping people more efficiently
gather new information; help automate a wide variety of administrative and repetitive tasks; and
enhance the productivity of many industries.
22
With respect to challenges, because Generative AI
systems can respond to harmful instructions, they can increase the speed and scale of many real
world harms, such as facilitating the development and proliferation of false information;
facilitating the use of copyrighted, proprietary, or sensitive data, without the owner’s or subject’s
knowledge; reducing privacy for users, including minors, through the retention of personally
identifiable information without consent; and facilitating the storage and use of sensitive
information by foreign adversaries.
23
Regulation
Concerns about the potential misuse or unintended consequences of AI have prompted efforts to
examine and develop standards at the federal and state levels.
24
In the 2024 legislative session, at least 45 states, Puerto Rico, the Virgin Islands and
Washington, D.C., introduced AI bills, and 31 states, Puerto Rico and the Virgin Islands adopted
resolutions or enacted legislation. Examples of those actions include the following:
25
• Colorado enacted comprehensive AI legislation requiring developers and deployers of high-
risk AI systems to use reasonable care to avoid algorithmic discrimination and requires
disclosures to consumers.
• Hawaii required the University of Hawaii to implement a program to develop a wildfire
forecast system to forecast the risk of wildfire statewide using AI.
• Indiana created an AI task force.
• Maryland required the Department of Information Technology to adopt policies and
procedures concerning the development, procurement, deployment, use and assessment of
systems that employ AI by units of state government.
• New Hampshire created the crime of the fraudulent use of deepfakes and establishes a cause
of action.
• Tennessee required the governing boards of public institutions of higher education to
promulgate rules and required local education boards and public charter schools to adopt
policies, regarding the use of AI by students, teachers, faculty and staff for instructional
purposes.
20
Government Accountability Office (GAO), Science, Technology Assessment, and Analytics, Science & Tech Spotlight:
Generative AI (Jun. 2023), available at https://www.gao.gov/assets/gao-23-106782.pdf; (last visited Mar. 7, 2025), George
Lawton, What is generative AI? Everything you need to know, TechTarget, Oct. 2024, available at
https://www.techtarget.com/searchenterpriseai/definition/generative-AI (last visited Mar. 7, 2025).
21
Government Accountability Office (GAO), Science, Technology Assessment, and Analytics, Science & Tech Spotlight:
Generative AI (Jun. 2023), available at https://www.gao.gov/assets/gao-23-106782.pdf (last visited Mar. 7, 2025).
22
Id.
23
Id.
24
NCSL, Artificial Intelligence 2024 Legislation, Sep. 9, 2024, available at https://www.ncsl.org/technology-and-
communication/artificial-intelligence-2024-legislation (last visited Mar. 7, 2025).
25
Id. BILL: SB 702 Page 6
• Utah created the Artificial Intelligence Policy Act.
• West Virginia created a select committee on AI.
26
Government Technology Modernization Council
In 2024, the Legislature established the Government Technology Modernization Council within
the Department of Management Services to study and monitor the development and deployment
of new technologies and provide reports on recommendations for procurement and regulation of
such systems to the Governor, the President of the Senate, and the Speaker of the House of
representatives.
27
By December 31, 2024, and each December 31 thereafter, the council is
required to submit to the President of the Senate and the Speaker of the House of Representatives
any legislative recommendations considered necessary by the council to modernize government
technology, including:
• Recommendations for policies necessary to:
o Accelerate adoption of technologies that will increase productivity of state enterprise
information technology systems, improve customer service levels of government, and
reduce administrative or operating costs;
o Promote the development and deployment of artificial intelligence systems, financial
technology, education technology, or other enterprise management software in Florida;
and
o Protect Floridians from bad actors who use artificial intelligence.
• Any other information the council considers relevant.
28
Florida Cybersecurity Advisory Council
In 2021, the Legislature established the Florida Cybersecurity Advisory Council within the
Department of Management Services to assist state agencies in protecting their information
technology resources from cyber threats and incidents. The council is required to submit to the
President of the Senate and the Speaker of the House of Representatives any legislative
recommendations considered by the council to address cybersecurity beginning June 30, 2022
and each June 30th thereafter.
29
Disclaimer Requirements for Political Advertising
Current law defines “political advertisement” to mean a paid expression in a communications
medium,
30
whether radio, television, newspaper, magazine, periodical, campaign literature, direct
mail, or display or by means other than the spoken word in direct conversation, which expressly
advocates the election or defeat of a candidate or the approval or rejection of an issue.
31
However, the term does not include:
26
Id.
27
Ch. 2024-118, Laws of Fla.
28
Section 282.802, F.S.
29
Department of Management Services, Cybersecurity Advisory Council, available at
https://www.dms.myflorida.com/other_programs/cybersecurity_advisory_council (last visited Mar. 7, 2025).
30
Section 106.011(4), F.S., defines “communications medium” to include broadcasting stations, newspapers, magazines,
outdoor advertising facilities, printers, direct mail, advertising agencies, the Internet, and telephone companies.
31
Section 106.011(15), F.S. BILL: SB 702 Page 7
• A statement by an organization, in existence before the time during which a candidate
qualifies or an issue is placed on the ballot for that election, in support of or in opposition to a
candidate or issue, in that organization’s newsletter, which newsletter is distributed only to
the members of that organization.
• Editorial endorsements by a newspaper, a radio or television station, or any other recognized
news medium.
32
An “electioneering communication” is a text message or communication that is publicly
distributed by a television station, radio station, cable television system, satellite system,
newspaper, magazine, direct mail, or telephone which:
• Refers to or depicts a clearly identified candidate for office without expressly advocating the
election or defeat of a candidate but that is susceptible of no reasonable interpretation other
than an appeal to vote for or against a specific candidate;
• Is made within 30 days before a primary or special primary election or 60 days before any
other election for the office sought by the candidate; and
• Is targeted to the relevant electorate in the geographic area the candidate would represent if
elected.
33
Specified types of communications are exempted from the definition.
34
Political advertisements, electioneering communications, and certain text message and telephone
solicitations must disclose who approved and paid for the advertisement or communication.
35
Voter guides must also include the required disclaimers, as applicable, and be marked “Voter
Guide.”
36
In addition, any advertisement, other than a political advertisement, independent
expenditure,
37
or electioneering communication, on billboards, bumper stickers, radio, or
television, or in a newspaper, a magazine, or a periodical, intended to influence public policy or
the vote of a public official, must clearly designate the sponsor of such advertisement by
including a statement of sponsorship.
38
Generally, penalties for failure to include a required disclaimer apply to a “person” as defined in
s. 106.011(14), F.S. – an individual or a corporation, association, firm, partnership, joint venture,
32
Id.
33
Section 106.011(8)(a), F.S.
34
See s. 106.011(8)(b), F.S.
35
Sections 106.071, 106.143, 106.1439, and 106.147, F.S. A failure to include a required disclaimer for an electioneering
communication or for a political advertisement paid for by independent expenditure constitutes a first-degree misdemeanor
(ss. 106.1439(2) and 106.071(4), F.S.). A willful failure to include a required disclaimer for other political advertisements is
subject to prescribed civil penalties (s. 106.143(11), F.S.). A willful violation of text message and telephone solicitation
requirements constitutes a first-degree misdemeanor (s. 106.147(4), F.S.).
36
Section 106.1436, F.S. The term “voter guide” means direct mail that is either an electioneering communication or a
political advertisement sent for the purpose of advocating for or endorsing particular issues or candidates by recommending
specific electoral choices to the voter or by indicating issue or candidate selections on an unofficial ballot (s. 106.1436(1),
F.S.). In addition to any other penalties provided by law, a person who fails to include the required disclaimer commits a
first-degree misdemeanor (s. 106.1436(4), F.S.).
37
Section 106.011(12)(a), F.S., defines “independent expenditure” to mean an expenditure by a person for the purpose of
expressly advocating the election or defeat of a candidate or the approval or rejection of an issue, which expenditure is not
controlled by, coordinated with, or made upon consultation with, any candidate, political committee, or agent of such
candidate or committee.
38
Section 106.1437, F.S. The section does not specify a penalty for failure to include the required statement. BILL: SB 702 Page 8
joint stock company, club, organization, estate, trust, business trust, syndicate or other
combination of individuals having collective capacity. The term includes a political party,
affiliated party committee, or political committee. Examples of past cases relating to existing
Florida Election Code
39
requirements that apply to a “person:”
• Allowed for an officer of a political committee to be held personally liable for failure to
include a required disclaimer on a political advertisement;
40
• Imposed penalties on a political consulting firm and the president of the firm for failing to
include disclaimers in, and report expenditures for, certain independent political
advertisements;
41
• Imposed penalties on a political committee for exceeding contribution limits;
42
and
• Allowed a corporation to be criminally charged for exceeding a contribution limit.
43
Election-Related Use of Artificial Intelligence
The proliferation of Generative AI use and its outpacing of government regulation has created
concern among policymakers about its potentially negative effect on the electoral process.
Specific concerns include, but are not limited to, voter misinformation by chatbots,
44
phishing
scams on election officials through AI-generated voices, and the use of deepfakes
45
to deceive
voters and damage political rivals. Over time, the use of AI may also erode trust in authentic
information.
46
In 2024, the Legislature enacted a law to regulate the use of AI in political advertising.
47
If a
political advertisement, an electioneering communication, or other miscellaneous advertisement
of a political nature contains images, video, audio, graphics, or other digital content created in
whole or in part with the use of Generative AI, if the generated content appears to depict a real
person performing an action that did not actually occur, and if the generated content was created
with intent to injure a candidate or to deceive regarding a ballot issue, the political
advertisement, electioneering communication, or other miscellaneous advertisement must
prominently state the following disclaimer: “Created in whole or in part with the use of
Generative Artificial Intelligence (AI).”
48
Additionally, the disclaimer must:
• For a printed communication, be stated in bold font with a font size of at least 12 points.
39
Chapters 97-106, F.S., are known as “The Florida Election Code.” Section 97.011, F.S.
40
See Fulton v. Division of Elections, 689 So.2d 1180 (Fla. Dist. Ct. App. 1997).
41
See Guetzloe v. Florida Elections Commission, 927 So.2d 942 (Fla. Dist. Ct. App. 2006).
42
See Florida PBA-PAC v. Division of Elections, 430 So.2d 483 (Fla. Dist. Ct. App. 1983).
43
See Winn-Dixie Stores, Inc., v. State of Florida, 408 So.2d 211 (Fla. 1982).
44
IBM defines “chatbot” to mean a computer program that simulates human conversation with an end user (see What is a
Chatbot?, available at https://www.ibm.com/topics/chatbots (last visited Mar. 7, 2025)).
45
Although exact definitions of “deepfake” vary, all reflect a depiction of something that has not actually occurred. Merriam-
Webster, for example, defines “deepfake” to mean an image or recording that has been convincingly altered and manipulated
to misrepresent someone as doing or saying something that was not actually said or done (see https://www.merriam-
webster.com/dictionary/deepfake, last visited March 7, 2025).
46
National Conference of State Legislatures, Challenges Ahead for Lawmakers Seeking to Legislate AI in Campaigns,
available at https://www.ncsl.org/state-legislatures-news/details/challenges-ahead-for-lawmakers-seeking-to-legislate-ai-in-
campaigns?utm_source=national+conference+of+state+legislatures&utm_term=0_-61bea1f450-
%5blist_email_id%5d&utm_campaign=8fbf8e40e8-canvass-jan-4&utm_medium=email (last visited Mar. 7, 2025).
47
Ch. 2024-126, Laws of Fla.
48
Section 106.145(2), F.S. BILL: SB 702 Page 9
• For a television or video communication, be clearly readable throughout the communication
and occupy at least 4 percent of the vertical picture height.
• For an Internet public communication that includes text or graphic components, be viewable
without the user taking any action and be large enough to be clearly readable.
• For any audio component of a communication, be at least 3 seconds in length and spoken in a
clearly audible and intelligible manner at either the beginning or the end of the audio
component of the communication.
• For a graphic communication, be large enough to be clearly readable but no less than 4
percent of the vertical height of the communication.
49
Florida Elections Commission
The Florida Elections Commission (commission) has jurisdiction
50
to investigate and determine
violations of campaign finance laws and other specified provisions of the Florida Election
Code
51
upon receipt of a report by the Division of Elections
52
or a sworn complaint.
53
Upon a
finding of a violation of one of the laws under its jurisdiction, the commission, or in cases
referred to the Division of Administrative Hearings, an administrative law judge, may impose
fines up to $2,500 per count.
54
Provenance Data
There are various methods to make AI outputs more identifiable and traceable, which include the
following:
• Provenance refers to the origin of data or AI system outputs;
• Authentication is a method of establishing provenance via verifiable assertions about the
origins of the content;
• Watermarking is a method of establishing provenance through “the act of embedding
information, which is typically difficult to remove, into outputs created by AI;” and
• Content labeling refers to informing people as part of the user interface about the source of
the information they are receiving.
55
49
Section 106.145(3), F.S.
50
For the purposes of commission jurisdiction, a violation means the willful performance of an act prohibited by ch. 104 or
106, F.S., or the willful failure to perform an act required by such chapters. Willfulness is a determination of fact. Section
106.25(3), F.S.
51
Section 106.25(2), F.S. The commission is housed within the Department of Legal Affairs, but is not subject to the
department’s control, supervision, or direction. Section 106.24(1), F.S.
52
The Division of Elections is an administrative unit of the Department of State. Section 97.021(9), F.S.
53
Section 106.25(4), F.S.
54
Section 106.265, F.S. The fine may be multiplied by a factor of 3, not to exceed $7,500, after a person commits three
counts of the same category of offense. If applicable, the commission or administrative law judge may instead impose a civil
penalty as provided in s. 104.271 or s. 106.19, F.S.
55
National Telecommunications and Information Administration, AI Output Disclosures: Use, Provenance, Adverse
Incidents, Mar. 27, 2024, available at https://www.ntia.gov/issues/artificial-intelligence/ai-accountability-policy-
report/developing-accountability-inputs-a-deeper-dive/information-flow/ai-output-
disclosures#:~:text=Watermarking%20is%20a%20method%20for,output%20or%20the%20identity%20or (last visited Mar.
7, 2025). BILL: SB 702 Page 10
Data Provenance aims to help people make informed decisions about what data or content can be
confidently trusted. To do this, Data Provenance platforms track metadata while making it
immutable, transparent, and verifiable.
56
The Coalition for Content Provenance and Authenticity (C2PA) is an industry-led organization
that creates technical standards to verify the origin and history of media.
57
The C2PA’s goal is to
provide a mechanism for producers and custodians of any given content to assert, in a verifiable
manner, any information they wish to disclose about the creation of such content, as well as any
actions taken since the content’s creation.
58
In 2020, the “Data & Trust Alliance” was established by a group of CEOs of major companies,
which is based on a shared conviction that the future of business will be powered by the
responsible use of data and AI.
59
The Data & Trust Alliance created the first cross-industry
metadata standards to bring transparency to the origin of datasets used for both traditional data
and AI applications.
60
Division of Emergency Management
Under the leadership of the Governor, the Florida Division of Emergency Management (DEM)
plans for and responds to both natural and man-made disasters.
61
The disasters include floods,
hurricanes, and incidents involving hazardous materials or nuclear power.
62
Additionally, the
DEM prepares and implements a statewide Comprehensive Emergency Management Plan, and
conducts exercises to test state and county emergency response capabilities.
63
Florida Deceptive and Unfair Trade Practices Act
History and Purpose
The Florida Deceptive and Unfair Trade Practices Act (FDUTPA) became law in 1973.
64
The
FDUTPA is a consumer and business protection measure that prohibits unfair methods of
competition, unconscionable acts or practices, and unfair or deceptive acts or practices in trade or
commerce.
65
The FDUTPA is based on federal law, and s. 501.204(2), F.S., provides that it is the
56
Data Trails, What is Data Provenance, Jan. 30, 2024, available at https://www.datatrails.ai/what-is-data-provenance/ (last
visited Mar. 7, 2025).
57
Coalition for Content Provenance and Authenticity, Overview, available at
https://c2pa.org/#:~:text=The%20Coalition%20for%20Content%20Provenance,media%20creation%2C%20publication%20a
nd%20sharing. (last visited Mar. 7, 2025).
58
Coalition for Content Provenance and Authenticity, Guiding Principles, available at https://c2pa.org/principles/ (last
visited Mar. 7, 2025).
59
The Data & Trust Alliance, About D&TA, available at https://dataandtrustalliance.org/about (last visited Mar. 7, 2025).
60
The Data & Trust Alliance, Latest Work, available at https://dataandtrustalliance.org/work (last visited Mar. 7, 2025).
61
Florida Division of Emergency Management, About the Division, available at https://www.floridadisaster.org/dem/about-
the-division/ (last visited Mar. 7, 2025).
62
Id.
63
Id.
64
Ch. 73-124, Laws of Fla.; codified at part II of ch. 501, F.S.
65
See s. 501.202, F.S. Trade or commerce means the advertising, soliciting, providing, offering, or distributing, whether by
sale, rental, or otherwise, of any good or service, or any property, whether tangible or intangible, or any other article,
commodity, or thing of value, wherever situated. “Trade or commerce” shall include the conduct of any trade or commerce,
however denominated, including any nonprofit or not-for-profit person or activity. See s. 501.203(8), F.S. BILL: SB 702 Page 11
intent of the Legislature that due consideration and great weight must be given to the
interpretations of the Federal Trade Commission and the federal courts relating to section 5 of
the Federal Trade Commission Act.
66
The State Attorney or the Department of Legal Affairs may bring actions when it is in the public
interest on behalf of consumers or governmental entities.
67
The Office of the State Attorney may
enforce violations of the FDUTPA if the violations take place in its jurisdiction.
68
The
Department of Legal Affairs has enforcement authority if the violation is multi-jurisdictional, the
state attorney defers in writing, or the state attorney fails to act within 90 days after a written
complaint is filed.
69
Consumers may also file suit through private actions.
70
Remedies under the FDUTPA
The Department of Legal Affairs and the State Attorney, as enforcing authorities, may seek the
following remedies:
• Declaratory judgments.
• Injunctive relief.
• Actual damages on behalf of consumers and businesses.
• Cease and desist orders.
• Civil penalties of up to $10,000 per willful violation.
71
Remedies for private parties are limited to the following:
• A declaratory judgment and an injunction where a person is aggrieved by a FDUTPA
violation.
• Actual damages, attorney fees, and court costs, where a person has suffered a loss due to a
FDUTPA violation.
72
III. Effect of Proposed Changes:
Provenance Data Definition
Section 1 of the bill amends s. 106.145, F.S., to define “provenance data” as information that
records the origin of a piece of visual or audio digital content and the history of modifications to
such content which is in a format that is compliant with widely adopted guidelines or
specifications promulgated by an established standard-setting body. The term includes, but is not
limited to, information identifying whether some or all of the content has been derived through
generative artificial intelligence and, if so, the name of the generative artificial intelligence tool
used to generate such content and the organization that developed such tool.
66
See s 501.204(2), F.S.
67
See ss. 501.203(2), 501.206, and 501.207, F.S.
68
Section 501.203(2), F.S.
69
Id.
70
Section 501.211, F.S.
71
Sections 501.207(1), 501.208, and 501.2075, F.S. Civil Penalties are deposited into general revenue. Section 501.2075,
F.S. Enforcing authorities may also request attorney fees and costs of investigation or litigation. Section 501.2105, F.S.
72
Section 501.211(1) and (2), F.S. BILL: SB 702 Page 12
Election Campaign Content
The bill requires any content regarding an election or purporting to feature a candidate which can
be viewed, heard, or accessed online to carry digital provenance data.
Digital Content Provenance Pilot Program
Section 2 of the bill creates s. 252.353, F.S., to establish a digital content provenance pilot
program within the DEM. The purpose of the pilot program is to enhance the security and
authenticity of digital content used in emergency management operations through the inclusion
of provenance data.
The bill requires the DEM to adopt rules necessary to implement s. 252.353, F.S. Additionally,
by December 1 of each year, the DEM must submit to the President of the Senate and the
Speaker of the House of Representatives a report that includes information concerning the pilot
program and whether any changes should be made to the pilot program which would increase its
effectiveness. In the report submitted by December 1, 2029, the DEM must include a
recommendation of whether the pilot program should be continued, terminated, or expanded.
The pilot program will be repealed on June 30, 2030, unless reviewed and saved from repeal
through reenactment by the Legislature.
Provenance Data of Digital Content
Section 3 of the bill creates s. 501.9741, F.S., and provides the following definitions:
• “Application tool” means a tool or service that enables the user to apply provenance data,
either directly or through the use of third-party technology, to any data that has been
modified to include synthetic content.
• “Capture device” means a device that can record any visual or audio digital content,
including, but not limited to, a camera, a cellular phone with a camera, a microphone, or an
audio or video recorder.
• “Generative artificial intelligence” has the same meaning as in s. 106.145(1), F.S.
73
• “Generative artificial intelligence tool” means a product or feature that uses generative
artificial intelligence to create visual or audio digital content.
• “Provenance data” has the same meaning as in s. 106.145(1), F.S.
74
• “Provenance reader” means a tool or service that allows users to identify the provenance data
of visual or audio digital content.
73
Section 106.145(1), F.S., defines “generative artificial intelligence” as a machine-based system that can, for a given set of
human-defined objectives, emulate the structure and characteristics of input data in order to generate derived synthetic
content including images, videos, audio, text, and other digital content.
74
“Provenance data” is defined as information that records the origin of a piece of visual or audio digital content and the
history of modifications to such content which is in a format that is compliant with widely adopted guidelines or
specifications promulgated by an established standard-setting body. The term includes, but is not limited to, information
identifying whether some or all of the content has been derived through generative artificial intelligence and, if so, the name
of the generative artificial intelligence tool used to generate such content and the organization that developed such tool. BILL: SB 702 Page 13
The bill requires provenance data to be included on any synthetic data wholly created by a
generative artificial intelligence tool. The provider of such an artificial intelligence tool must
make an application tool and a free provenance reader available to the public.
The bill requires a social media platform
75
to retain all available provenance data of visual or
audio digital content provided to or posted on such platform and make such data available to
users of the platform through a conspicuous indicator.
The bill requires any capture device sold in Florida to allow an option to include provenance data
on any visual or audio digital content recorded with such device. Additionally, the manufacturer
of a capture device must ensure that such provenance data can be read by third-party
applications.
The bill provides that a violation of this section constitutes an unfair or deceptive act or practice.
The bill takes effect July 1, 2025.
IV. Constitutional Issues:
A. Municipality/County Mandates Restrictions:
None.
B. Public Records/Open Meetings Issues:
None.
C. Trust Funds Restrictions:
None.
D. State Tax or Fee Increases:
None.
E. Other Constitutional Issues:
The First Amendment of the U.S. Constitution promotes the free exchange of ideas and
information by prohibiting the government from restricting speech because of the
message expressed.
76
Content-based restrictions are presumptively invalid.
77
Among
75
Section 501.2041(1), F.S., defines “social media platform” as any information service, system, Internet search engine, or
access software provider that: (1) provides or enables computer access by multiple users to a computer server, including an
Internet platform or a social media site; (2) operates as a sole proprietorship, partnership, limited liability company,
corporation, association, or other legal entity; (3) does business in Florida; and (4) satisfies at least one of the following
thresholds: (1) has annual gross revenues in excess of $100 million, as adjusted in January of each odd-numbered year to
reflect any increase in the Consumer Price Index; or (2) has at least 100 million monthly individual platform participants
globally.
76
See, e.g., Texas v. Johnson (491 U.S. 397 (1989); State v. T.B.D., 656 So.2d 479 (Fla. 1995).
77
See, e.g., Police Dept. of Chicago v. Mosley, 408 U.S. 92 (1972). BILL: SB 702 Page 14
specific rights, the First Amendment protects the right to associate for expressive or
political activity. The government may infringe upon this right only if it has a compelling
interest unrelated to the suppression of speech and if the interest cannot be achieved
through significantly less restrictive means.
78
This bill does not prohibit or restrict the content of speech generated by AI. Instead, it
requires the use of provenance data that records the origin of a piece of visual or audio
digital content, as well as the history of the modification of such content.
V. Fiscal Impact Statement:
A. Tax/Fee Issues:
None.
B. Private Sector Impact:
Social media platforms will be required to retain all available provenance data of visual
or audio digital content provided to or posted on such platform, as well as make the
provenance data available to users of the platform through a conspicuous indicator.
Sellers of a capture device in Florida will be required to allow an option to include
provenance data on any visual or audio digital content recorded with such device.
Additionally, manufacturers of a capture device must ensure that such provenance data
can be read by third-party applications, and may be implemented differently across
platforms.
C. Government Sector Impact:
The Division of Emergency Management will be responsible for implementing the
Digital Content Provenance Pilot Program.
VI. Technical Deficiencies:
None.
VII. Related Issues:
The bill requires social media platforms to retain all available provenance data of visual or audio
digital content provided to or posted to such platforms and make such data available to users of
the platform through a conspicuous indicator. However, the meaning of “conspicuous indicator”
may not be clear, and may be interpreted differently across platforms.
The definition of “capture device” is broad enough to include many types of consumer electronic
devices.
78
City of Dallas v. Stanglin, 490 U.S. 19 (1989). BILL: SB 702 Page 15
Provenance data must be included on any synthetic data “wholly created” by a generative AI
tool; any synthetic data not “wholly created” by a generative AI would not be required to include
provenance data.
VIII. Statutes Affected:
This bill substantially amends section 106.145 of the Florida Statutes.
The bill creates the following sections of the Florida Statutes: 252.353 and 501.9741.
IX. Additional Information:
A. Committee Substitute – Statement of Changes:
(Summarizing differences between the Committee Substitute and the prior version of the bill.)
None.
B. Amendments:
None.
This Senate Bill Analysis does not reflect the intent or official position of the bill’s introducer or the Florida Senate.