AMENDED IN ASSEMBLY AUGUST 22, 2024

AMENDED IN ASSEMBLY AUGUST 19, 2024

AMENDED IN ASSEMBLY JULY 3, 2024


AMENDED IN ASSEMBLY JUNE 20, 2024

AMENDED IN ASSEMBLY JUNE 5, 2024

AMENDED IN SENATE MAY 16, 2024


AMENDED IN SENATE APRIL 30, 2024

AMENDED IN SENATE APRIL 16, 2024

AMENDED IN SENATE APRIL 8, 2024


AMENDED IN SENATE MARCH 20, 2024

# SENATE BILL No. 1047


**Introduced by Senator Wiener**

**(Coauthors: Senators Roth, Rubio, and Stern)**


February 7, 2024

An act to add Chapter 22.6 (commencing with Section 22602) to
Division 8 of the Business and Professions Code, and to add Sections
11547.6 and 11547.6.1 to the Government Code, relating to artificial
intelligence.


legislative counsel’s digest


SB 1047, as amended, Wiener. Safe and Secure Innovation for
Frontier Artificial Intelligence Models Act.

Existing law requires the Secretary of Government Operations to
develop a coordinated plan to, among other things, investigate the

89


-----

**SB 1047** **2**

feasibility of, and obstacles to, developing standards and technologies
for state departments to determine digital content provenance. For the
purpose of informing that coordinated plan, existing law requires the
secretary to evaluate, among other things, the impact of the proliferation
of deepfakes, defined to mean audio or visual content that has been
generated or manipulated by artificial intelligence that would falsely
appear to be authentic or truthful and that features depictions of people
appearing to say or do things they did not say or do without their
consent, on state government, California-based businesses, and residents
of the state.

This bill would enact the Safe and Secure Innovation for Frontier
Artificial Intelligence Models Act to, among other things, require that
a developer, before beginning to initially train a covered model, as
defined, comply with various requirements, including implementing
the capability to promptly enact a full shutdown, as defined, and
implement a written and separate safety and security protocol, as
specified. The bill would require a developer to retain an unredacted
copy of the safety and security protocol for as long as the covered model
is made available for commercial, public, or foreseeably public use plus
5 years, including records and dates of any updates or revisions and
would require a developer to grant to the Attorney General access to
the unredacted safety and security protocol. The bill would prohibit a
developer from using a covered model or covered model derivative for
a purpose not exclusively related to the training or reasonable evaluation
of the covered model or compliance with state or federal law or making
a covered model or a covered model derivative available for commercial
or public, or foreseeably public, use, if there is an unreasonable risk
that the covered model or covered model derivative will cause or
materially enable a critical harm, as defined. The bill would require a
developer, beginning January 1, 2026, to annually retain a third-party
auditor to perform an independent audit of compliance with those
provisions, as prescribed. The bill would require the auditor to produce
an audit report, as prescribed, and would require a developer to retain
an unredacted copy of the audit report for as long as the covered model
is made available for commercial, public, or foreseeably public use plus
5 years. The bill would require a developer to grant to the Attorney
General access to the unredacted auditor’s report upon request. The bill
would exempt from disclosure under the California Public Records Act
the safety and security protocol and the auditor’s report described above.


89


-----

**SB 1047**


This bill would require a developer of a covered model to submit to
the Attorney General a statement of compliance with these provisions,
as specified. The bill would also require a developer of a covered model
to report each artificial intelligence safety incident affecting the covered
model, or any covered model derivative controlled by the developer to
the Attorney General, as prescribed.

This bill would require a person that operates a computing cluster,
as defined, to implement written policies and procedures to do certain
things when a customer utilizes compute resources that would be
sufficient to train a covered model, including assess whether a
prospective customer intends to utilize the computing cluster to train a
covered model.

This bill would authorize the Attorney General to bring a civil action,
as provided. The bill would also provide for whistleblower protections,
including by prohibiting a developer of a covered model or a contractor
or subcontractor of the developer from preventing an employee from
disclosing information, or retaliating against an employee for disclosing
information, to the Attorney General or Labor Commissioner if the
employee has reasonable cause to believe the information indicates the
developer is out of compliance with certain requirements or that the
covered model poses an unreasonable risk of critical harm.

This bill would create the Board of Frontier Models within the
Government Operations Agency, independent of the Department of
Technology, and provide for the board’s membership. The bill would
require the Government Operations Agency to, on or before January 1,
2027, and annually thereafter, issue regulations to, among other things,
update the definition of a “covered model,” as provided, and would
require the regulations to be approved by the board before taking effect.

This bill would establish in the Government Operations Agency a
consortium required to develop a framework for the creation of a public
cloud computing cluster to be known as “CalCompute” that advances
the development and deployment of artificial intelligence that is safe,
ethical, equitable, and sustainable by, among other things, fostering
research and innovation that benefits the public, as prescribed. The bill
would on or before January 1, 2026, require the Government Operations
Agency to submit a report from the consortium to the Legislature with
that framework. The bill would make those provisions operable only
upon an appropriation in a budget act for its purposes.

Existing constitutional provisions require that a statute that limits the
right of access to the meetings of public bodies or the writings of public

89


-----

**SB 1047** **4**

officials and agencies be adopted with findings demonstrating the
interest protected by the limitation and the need for protecting that
interest.

This bill would make legislative findings to that effect.


Vote:  majority.  Appropriation:  no. Fiscal committee:  yes.​
State-mandated local program:  no.​

_The people of the State of California do enact as follows:_


line 1 SECTION 1. This act shall be known, and may be cited, as the

line 2 Safe and Secure Innovation for Frontier Artificial Intelligence
line 3 Models Act.
line 4 SEC. 2. The Legislature finds and declares all of the following:

line 5 (a) California is leading the world in artificial intelligence

line 6 innovation and research, through companies large and small, as
line 7 well as through our remarkable public and private universities.
line 8 (b) Artificial intelligence, including new advances in generative

line 9 artificial intelligence, has the potential to catalyze innovation and

line 10 the rapid development of a wide range of benefits for Californians
line 11 and the California economy, including advances in medicine,
line 12 wildfire forecasting and prevention, and climate science, and to
line 13 push the bounds of human creativity and capacity.
line 14 (c) If not properly subject to human controls, future development

line 15 in artificial intelligence may also have the potential to be used to
line 16 create novel threats to public safety and security, including by
line 17 enabling the creation and the proliferation of weapons of mass
line 18 destruction, such as biological, chemical, and nuclear weapons,
line 19 as well as weapons with cyber-offensive capabilities.
line 20 (d) The state government has an essential role to play in ensuring

line 21 that California recognizes the benefits of this technology while
line 22 avoiding the most severe risks, as well as to ensure that artificial
line 23 intelligence innovation and access to compute is accessible to
line 24 academic researchers and startups, in addition to large companies.
line 25 SEC. 3. Chapter 22.6 (commencing with Section 22602) is

line 26 added to Division 8 of the Business and Professions Code, to read:
line 27
line 28 Chapter 22.6. Safe and Secure Innovation for Frontier

line 29 Artificial Intelligence Models

line 30
line 31 22602. As used in this chapter:

89


-----

**SB 1047**


line 1 (a) “Advanced persistent threat” means an adversary with

line 2 sophisticated levels of expertise and significant resources that
line 3 allow it, through the use of multiple different attack vectors,
line 4 including, but not limited to, cyber, physical, and deception, to
line 5 generate opportunities to achieve its objectives that are typically
line 6 to establish and extend its presence within the information
line 7 technology infrastructure of organizations for purposes of
line 8 exfiltrating information or to undermine or impede critical aspects
line 9 of a mission, program, or organization or place itself in a position

line 10 to do so in the future.
line 11 (b) “Artificial intelligence” means an engineered or

line 12 machine-based system that varies in its level of autonomy and that
line 13 can, for explicit or implicit objectives, infer from the input it
line 14 receives how to generate outputs that can influence physical or
line 15 virtual environments.
line 16 (c) “Artificial intelligence safety incident” means an incident

line 17 that demonstrably increases the risk of a critical harm occurring
line 18 by means of any of the following:
line 19 (1) A covered model or covered model derivative autonomously

line 20 engaging in behavior other than at the request of a user.
line 21 (2) Theft, misappropriation, malicious use, inadvertent release,

line 22 unauthorized access, or escape of the model weights of a covered
line 23 model or covered model derivative.
line 24 (3) The critical failure of technical or administrative controls,

line 25 including controls limiting the ability to modify a covered model
line 26 or covered model derivative.
line 27 (4) Unauthorized use of a covered model or covered model

line 28 derivative to cause or materially enable critical harm.
line 29 (d) “Computing cluster” means a set of machines transitively

line 30 connected by data center networking of over 100 gigabits per
line 31 second that has a theoretical maximum computing capacity of at
line 32 least 10^20 integer or floating-point operations per second and
line 33 can be used for training artificial intelligence.
line 34 (e) (1) “Covered model” means either of the following:

line 35 (A) Before January 1, 2027, “covered model” means either of

line 36 the following:
line 37 (i) An artificial intelligence model trained using a quantity of

line 38 computing power greater than 10^26 integer or floating-point
line 39 operations, the cost of which exceeds one hundred million dollars
line 40 ($100,000,000) when calculated using the average market prices

89


-----

**SB 1047**


line 1 of cloud compute at the start of training as reasonably assessed by
line 2 the developer.
line 3 (ii) An artificial intelligence model created by fine-tuning a

line 4 covered model using a quantity of computing power equal to or
line 5 greater than three times 10^25 integer or floating-point operations,
line 6 the cost of which, as reasonably assessed by the developer, exceeds
line 7 ten million dollars ($10,000,000) if calculated using the average
line 8 market price of cloud compute at the start of fine-tuning.
line 9 (B) (i) Except as provided in clause (ii), on and after January

line 10 1, 2027, “covered model” means any of the following:
line 11 (I) An artificial intelligence model trained using a quantity of

line 12 computing power determined by the Government Operations
line 13 Agency pursuant to Section 11547.6 of the Government Code, the
line 14 cost of which exceeds one hundred million dollars ($100,000,000)
line 15 when calculated using the average market price of cloud compute
line 16 at the start of training as reasonably assessed by the developer.
line 17 (II) An artificial intelligence model created by fine-tuning a

line 18 covered model using a quantity of computing power that exceeds
line 19 a threshold determined by the Government Operations Agency,
line 20 the cost of which, as reasonably assessed by the developer, exceeds
line 21 ten million dollars ($10,000,000) if calculated using the average
line 22 market price of cloud compute at the start of fine-tuning.
line 23 (ii) If the Government Operations Agency does not adopt a

line 24 regulation governing subclauses (I) and (II) of clause (i) before
line 25 January 1, 2027, the definition of “covered model” in subparagraph
line 26 (A) shall be operative until the regulation is adopted.
line 27 (2) On and after January 1, 2026, the dollar amount in this

line 28 subdivision shall be adjusted annually for inflation to the nearest
line 29 one hundred dollars ($100) based on the change in the annual
line 30 California Consumer Price Index for All Urban Consumers
line 31 published by the Department of Industrial Relations for the most
line 32 recent annual period ending on December 31 preceding the
line 33 adjustment.
line 34 (f) “Covered model derivative” means any of the following:

line 35 (1) An unmodified copy of a covered model.

line 36 (2) A copy of a covered model that has been subjected to

line 37 post-training modifications unrelated to fine-tuning.
line 38 (3) (A) (i) Before January 1, 2027, a copy of a covered model

line 39 that has been fine-tuned using a quantity of computing power not
line 40 exceeding three times 10^25 integer or floating point operations,

89


-----

**SB 1047**


line 1 the cost of which, as reasonably assessed by the developer, exceeds
line 2 ten million dollars ($10,000,000) if calculated using the average
line 3 market price of cloud compute at the start of fine-tuning.
line 4 (ii) On and after January 1, 2027, a copy of a covered model

line 5 that has been fine-tuned using a quantity of computing power not
line 6 exceeding a threshold determined by the Government Operations
line 7 Agency, the cost of which, as reasonably assessed by the developer,
line 8 exceeds ten million dollars ($10,000,000) if calculated using the
line 9 average market price of cloud compute at the start of fine-tuning.

line 10 (B) If the Government Operations Agency does not adopt a

line 11 regulation governing clause (ii) of subparagraph (A) by January
line 12 1, 2027, the quantity of computing power specified in clause (i)
line 13 of subparagraph (A) shall continue to apply until the regulation is
line 14 adopted.
line 15 (4) A copy of a covered model that has been combined with

line 16 other software.
line 17 (g) (1) “Critical harm” means any of the following harms caused

line 18 or materially enabled by a covered model or covered model
line 19 derivative:
line 20 (A) The creation or use of a chemical, biological, radiological,

line 21 or nuclear weapon in a manner that results in mass casualties.
line 22 (B) Mass casualties or at least five hundred million dollars

line 23 ($500,000,000) of damage resulting from cyberattacks on critical
line 24 infrastructure by a model conducting, or providing precise
line 25 instructions for conducting, a cyberattack or series of cyberattacks
line 26 on critical infrastructure.
line 27 (C) Mass casualties or at least five hundred million dollars

line 28 ($500,000,000) of damage resulting from an artificial intelligence
line 29 model engaging in conduct that does both of the following:
line 30 (i) Acts with limited human oversight, intervention, or

line 31 supervision.
line 32 (ii) Results in death, great bodily injury, property damage, or

line 33 property loss, and would, if committed by a human, constitute a
line 34 crime specified in the Penal Code that requires intent, recklessness,
line 35 or gross negligence, or the solicitation or aiding and abetting of
line 36 such a crime.
line 37 (D) Other grave harms to public safety and security that are of

line 38 comparable severity to the harms described in subparagraphs (A)
line 39 to (C), inclusive.
line 40 (2) “Critical harm” does not include any of the following:

89


-----

**SB 1047**


line 1 (A) Harms caused or materially enabled by information that a

line 2 covered model or covered model derivative outputs if the
line 3 information is otherwise reasonably publicly accessible by an
line 4 ordinary person from sources other than a covered model or
line 5 covered model derivative.
line 6 (B) Harms caused or materially enabled by a covered model

line 7 combined with other software, including other models, if the
line 8 covered model did not materially contribute to the other software’s
line 9 ability to cause or materially enable the harm.

line 10 (C) Harms that are not caused or materially enabled by the

line 11 developer’s creation, storage, use, or release of a covered model
line 12 or covered model derivative.
line 13 (3) On and after January 1, 2026, the dollar amounts in this

line 14 subdivision shall be adjusted annually for inflation to the nearest
line 15 one hundred dollars ($100) based on the change in the annual
line 16 California Consumer Price Index for All Urban Consumers
line 17 published by the Department of Industrial Relations for the most
line 18 recent annual period ending on December 31 preceding the
line 19 adjustment.
line 20 (h) “Critical infrastructure” means assets, systems, and networks,

line 21 whether physical or virtual, the incapacitation or destruction of
line 22 which would have a debilitating effect on physical security,
line 23 economic security, public health, or safety in the state.
line 24 (i) “Developer” means a person that performs the initial training

line 25 of a covered model either by training a model using a sufficient
line 26 quantity of computing power and cost, or by fine-tuning an existing
line 27 covered model or covered model derivative using a quantity of
line 28 computing power and cost greater than the amount specified in
line 29 subdivision (e).
line 30 (j) “Fine-tuning” means adjusting the model weights of a trained

line 31 covered model or covered model derivative by exposing it to
line 32 additional data.
line 33 (k) “Full shutdown” means the cessation of operation of all of

line 34 the following:
line 35 (1) The training of a covered model.

line 36 (2) A covered model controlled by a developer.

line 37 (3) All covered model derivatives controlled by a developer.

line 38 (l) “Model weight” means a numerical parameter in an artificial

line 39 intelligence model that is adjusted through training and that helps
line 40 determine how inputs are transformed into outputs.

89


-----

**SB 1047**


line 1 (m) “Person” means an individual, proprietorship, firm,

line 2 partnership, joint venture, syndicate, business trust, company,
line 3 corporation, limited liability company, association, committee, or
line 4 any other nongovernmental organization or group of persons acting
line 5 in concert.
line 6 (n) “Post-training modification” means modifying the

line 7 capabilities of a covered model or covered model derivative by
line 8 any means, including, but not limited to, fine-tuning, providing
line 9 the model with access to tools or data, removing safeguards against

line 10 hazardous misuse or misbehavior of the model, or combining the
line 11 model with, or integrating it into, other software.
line 12 (o) “Safety and security protocol” means documented technical

line 13 and organizational protocols that meet both of the following
line 14 criteria:
line 15 (1) The protocols are used to manage the risks of developing

line 16 and operating covered models and covered model derivatives
line 17 across their life cycle, including risks posed by causing or enabling
line 18 or potentially causing or enabling the creation of covered model
line 19 derivatives.
line 20 (2) The protocols specify that compliance with the protocols is

line 21 required in order to train, operate, possess, and provide external
line 22 access to the developer’s covered model and covered model
line 23 derivatives.
line 24 22603. (a) Before beginning to initially train a covered model,

line 25 the developer shall do all of the following:
line 26 (1) Implement reasonable administrative, technical, and physical

line 27 cybersecurity protections to prevent unauthorized access to, misuse
line 28 of, or unsafe post-training modifications of, the covered model
line 29 and all covered model derivatives controlled by the developer that
line 30 are appropriate in light of the risks associated with the covered
line 31 model, including from advanced persistent threats or other
line 32 sophisticated actors.
line 33 (2) (A) Implement the capability to promptly enact a full

line 34 shutdown.
line 35 (B) When enacting a full shutdown, the developer shall take

line 36 into account, as appropriate, the risk that a shutdown of the covered
line 37 model, or particular covered model derivatives, could cause
line 38 disruptions to critical infrastructure.
line 39 (3) Implement a written and separate safety and security protocol

line 40 that does all of the following:

89


-----

**SB 1047** **10**


line 1 (A) Specifies protections and procedures that, if successfully

line 2 implemented, would successfully comply with the developer’s
line 3 duty to take reasonable care to avoid producing a covered model
line 4 or covered model derivative that poses an unreasonable risk of
line 5 causing or materially enabling a critical harm.
line 6 (B) States compliance requirements in an objective manner and

line 7 with sufficient detail and specificity to allow the developer or a
line 8 third party to readily ascertain whether the requirements of the
line 9 safety and security protocol have been followed.

line 10 (C) Identifies a testing procedure, which takes safeguards into

line 11 account as appropriate, that takes reasonable care to evaluate if
line 12 both of the following are true:
line 13 (i) A covered model poses an unreasonable risk of causing or

line 14 enabling a critical harm.
line 15 (ii) Covered model derivatives pose an unreasonable risk of

line 16 causing or enabling a critical harm.
line 17 (D) Describes in detail how the testing procedure assesses the

line 18 risks associated with post-training modifications.
line 19 (E) Describes in detail how the testing procedure addresses the

line 20 possibility that a covered model or covered model derivative can
line 21 be used to make post-training modifications or create another
line 22 covered model in a manner that may cause or materially enable a
line 23 critical harm.
line 24 (F) Describes in detail how the developer will fulfill their

line 25 obligations under this chapter.
line 26 (G) Describes in detail how the developer intends to implement

line 27 the safeguards and requirements referenced in this section.
line 28 (H) Describes in detail the conditions under which a developer

line 29 would enact a full shutdown.
line 30 (I) Describes in detail the procedure by which the safety and

line 31 security protocol may be modified.
line 32 (4) Ensure that the safety and security protocol is implemented

line 33 as written, including by designating senior personnel to be
line 34 responsible for ensuring compliance by employees and contractors
line 35 working on a covered model, or any covered model derivatives
line 36 controlled by the developer, monitoring and reporting on
line 37 implementation.
line 38 (5) Retain an unredacted copy of the safety and security protocol

line 39 for as long as the covered model is made available for commercial,

89


-----

**11** **SB 1047**


line 1 public, or foreseeably public use plus five years, including records
line 2 and dates of any updates or revisions.
line 3 (6) Conduct an annual review of the safety and security protocol

line 4 to account for any changes to the capabilities of the covered model
line 5 and industry best practices and, if necessary, make modifications
line 6 to the policy.
line 7 (7) (A) (i) Conspicuously publish a copy of the redacted safety

line 8 and security protocol and transmit a copy of the redacted safety
line 9 and security protocol to the Attorney General.

line 10 (ii) A redaction in the safety and security protocol may be made

line 11 only if the redaction is reasonably necessary to protect any of the
line 12 following:
line 13 (I) Public safety.

line 14 (II) Trade secrets, as defined in Section 3426.1 of the Civil

line 15 Code.
line 16 (III) Confidential information pursuant to state and federal law.

line 17 (B) The developer shall grant to the Attorney General access

line 18 to the unredacted safety and security protocol upon request.
line 19 (C) A safety and security protocol disclosed to the Attorney

line 20 General pursuant to this paragraph is exempt from the California
line 21 Public Records Act (Division 10 (commencing with Section
line 22 7920.000) of Title 1 of the Government Code).
line 23 (D) If the safety and security protocol is materially modified,

line 24 conspicuously publish and transmit to the Attorney General an
line 25 updated redacted copy within 30 days of the modification.
line 26 (8) Take reasonable care to implement other appropriate

line 27 measures to prevent covered models and covered model derivatives
line 28 from posing unreasonable risks of causing or materially enabling
line 29 critical harms.
line 30 (b) Before using a covered model or covered model derivative

line 31 for a purpose not exclusively related to the training or reasonable
line 32 evaluation of the covered model or compliance with state or federal
line 33 law or before making a covered model or covered model derivative
line 34 available for commercial or public, or foreseeably public, use, the
line 35 developer of a covered model shall do all of the following:
line 36 (1) Assess whether the covered model is reasonably capable of

line 37 causing or materially enabling a critical harm.
line 38 (2) Record, as and when reasonably possible, and retain for as

line 39 long as the covered model is made available for commercial,
line 40 public, or foreseeably public use plus five years information on

89


-----

**SB 1047** **12**


line 1 the specific tests and test results used in the assessment pursuant
line 2 to paragraph (1) that provides sufficient detail for third parties to
line 3 replicate the testing procedure.
line 4 (3) Take reasonable care to implement appropriate safeguards

line 5 to prevent the covered model and covered model derivatives from
line 6 causing or materially enabling a critical harm.
line 7 (4) Take reasonable care to ensure, to the extent reasonably

line 8 possible, that the covered model’s actions and the actions of
line 9 covered model derivatives, as well as critical harms resulting from

line 10 their actions, can be accurately and reliably attributed to them.
line 11 (c) A developer shall not use a covered model or covered model

line 12 derivative for a purpose not exclusively related to the training or
line 13 reasonable evaluation of the covered model or compliance with
line 14 state or federal law or make a covered model or a covered model
line 15 derivative available for commercial or public, or foreseeably public,
line 16 use, if there is an unreasonable risk that the covered model or
line 17 covered model derivative will cause or materially enable a critical
line 18 harm.
line 19 (d) A developer of a covered model shall annually reevaluate

line 20 the procedures, policies, protections, capabilities, and safeguards
line 21 implemented pursuant to this section.
line 22 (e) (1) Beginning January 1, 2026, a developer of a covered

line 23 model shall annually retain a third-party auditor that conducts
line 24 audits consistent with best practices for auditors to perform an
line 25 independent audit of compliance with the requirements of this
line 26 section.
line 27 (2) An auditor shall conduct audits consistent with regulations

line 28 issued by the Government Operations Agency pursuant to
line 29 subdivision (d) of Section 11547.6 of the Government Code.
line 30 (3) The auditor shall be granted access to unredacted materials

line 31 as necessary to comply with the auditor’s obligations under this
line 32 subdivision.
line 33 (4) The auditor shall produce an audit report including all of

line 34 the following:
line 35 (A) A detailed assessment of the developer’s steps to comply

line 36 with the requirements of this section.
line 37 (B) If applicable, any identified instances of noncompliance

line 38 with the requirements of this section, and any recommendations
line 39 for how the developer can improve its policies and processes for
line 40 ensuring compliance with the requirements of this section.

89


-----

**13** **SB 1047**


line 1 (C) A detailed assessment of the developer’s internal controls,

line 2 including its designation and empowerment of senior personnel
line 3 responsible for ensuring compliance by the developer, its
line 4 employees, and its contractors.
line 5 (D) The signature of the lead auditor certifying the results of

line 6 the auditor.
line 7 (5) The developer shall retain an unredacted copy of the audit

line 8 report for as long as the covered model is made available for
line 9 commercial, public, or foreseeably public use plus five years.

line 10 (6) (A) (i) The developer shall conspicuously publish a redacted

line 11 copy of the auditor’s report and transmit to the Attorney General
line 12 a copy of the redacted auditor’s report.
line 13 (ii) A redaction in the auditor’s report may be made only if the

line 14 redaction is reasonably necessary to protect any of the following:
line 15 (I) Public safety.

line 16 (II) Trade secrets, as defined in Section 3426.1 of the Civil

line 17 Code.
line 18 (III) Confidential information pursuant to state and federal law.

line 19 (B) The developer shall grant to the Attorney General access

line 20 to the unredacted auditor’s report upon request.
line 21 (C) An auditor’s report disclosed to the Attorney General

line 22 pursuant to this paragraph is exempt from the California Public
line 23 Records Act (Division 10 (commencing with Section 7920.000)
line 24 of Title 1 of the Government Code).
line 25 (7) An auditor shall not knowingly make a material

line 26 misrepresentation in the auditor’s report.
line 27 (f) (1) (A) A developer of a covered model shall annually

line 28 submit to the Attorney General a statement of compliance with
line 29 the requirements of this section signed by the chief technology
line 30 officer, or a more senior corporate officer, that meets the
line 31 requirements of paragraph (2).
line 32 (B) This paragraph applies if the covered model or any covered

line 33 model derivatives controlled by the developer remain in
line 34 commercial or public use or remain available for commercial or
line 35 public use.
line 36 (2) In a statement submitted pursuant to paragraph (1), a

line 37 developer shall specify or provide, at a minimum, all of the
line 38 following:
line 39 (A) An assessment of the nature and magnitude of critical harms

line 40 that the covered model or covered model derivatives may

89


-----

**SB 1047** **14**


line 1 reasonably cause or materially enable and the outcome of the
line 2 assessment required by paragraph (1) of subdivision (b).
line 3 (B) An assessment of the risk that compliance with the safety

line 4 and security protocol may be insufficient to prevent the covered
line 5 model or covered model derivatives from causing or materially
line 6 enabling critical harms.
line 7 (C) A description of the process used by the signing officer to

line 8 verify compliance with the requirements of this section, including
line 9 a description of the materials reviewed by the signing officer, a

line 10 description of testing or other evaluation performed to support the
line 11 statement and the contact information of any third parties relied
line 12 upon to validate compliance.
line 13 (g) A developer of a covered model shall report each artificial

line 14 intelligence safety incident affecting the covered model, or any
line 15 covered model derivatives controlled by the developer, to the
line 16 Attorney General within 72 hours of the developer learning of the
line 17 artificial intelligence safety incident or within 72 hours of the
line 18 developer learning facts sufficient to establish a reasonable belief
line 19 that an artificial intelligence safety incident has occurred.
line 20 (h) (1) A developer shall submit to the Attorney General a

line 21 statement described by subdivision (f) no more than 30 days after
line 22 using a covered model or covered model derivative for a purpose
line 23 not exclusively related to the training or reasonable evaluation of
line 24 the covered model or compliance with state or federal law or
line 25 making a covered model or covered model derivative available
line 26 for commercial or public, or foreseeably public, use for the first
line 27 time.
line 28 (2) This subdivision does not apply with respect to a covered

line 29 model derivative if the developer submitted a statement described
line 30 by subdivision (f) for the applicable covered model from which
line 31 the covered model derivative is derived.
line 32 (i) In fulfilling its obligations under this chapter, a developer

line 33 shall consider industry best practices and applicable guidance from
line 34 the U.S. Artificial Intelligence Safety Institute, National Institute
line 35 of Standards and Technology, the Government Operations Agency,
line 36 and other reputable standard-setting organizations.
line 37 (j) (1) This section shall not apply to products or services to

line 38 the extent that the requirements would strictly conflict with the
line 39 terms of a contract with a federal government entity and a
line 40 developer of a covered model.

89


-----

**15** **SB 1047**


line 1 (2) This section applies to the development, use, or commercial

line 2 or public release of a covered model or covered model derivative
line 3 for any use that is not the subject of a contract with a federal
line 4 government entity, even if that covered model or covered model
line 5 derivative has already been developed, trained, or used by a federal
line 6 government entity.
line 7 22604. (a) A person that operates a computing cluster shall

line 8 implement written policies and procedures to do all of the following
line 9 when a customer utilizes compute resources that would be

line 10 sufficient to train a covered model:
line 11 (1) Obtain the prospective customer’s basic identifying

line 12 information and business purpose for utilizing the computing
line 13 cluster, including all of the following:
line 14 (A) The identity of the prospective customer.

line 15 (B) The means and source of payment, including any associated

line 16 financial institution, credit card number, account number, customer
line 17 identifier, transaction identifiers, or virtual currency wallet or
line 18 wallet address identifier.
line 19 (C) The email address and telephonic contact information used

line 20 to verify the prospective customer’s identity.
line 21 (2) Assess whether the prospective customer intends to utilize

line 22 the computing cluster to train a covered model.
line 23 (3) If a customer repeatedly utilizes computer resources that

line 24 would be sufficient to train a covered model, validate the
line 25 information initially collected pursuant to paragraph (1) and
line 26 conduct the assessment required pursuant to paragraph (2) prior
line 27 to each utilization.
line 28 (4) Retain a customer’s Internet Protocol addresses used for

line 29 access or administration and the date and time of each access or
line 30 administrative action.
line 31 (5) Maintain for seven years and provide to the Attorney

line 32 General, upon request, appropriate records of actions taken under
line 33 this section, including policies and procedures put into effect.
line 34 (6) Implement the capability to promptly enact a full shutdown

line 35 of any resources being used to train or operate models under the
line 36 customer’s control.
line 37 (b) A person that operates a computing cluster shall consider

line 38 industry best practices and applicable guidance from the U.S.
line 39 Artificial Intelligence Safety Institute, National Institute of

89


-----

**SB 1047** **16**


line 1 Standards and Technology, and other reputable standard-setting
line 2 organizations.
line 3 (c) In complying with the requirements of this section, a person

line 4 that operates a computing cluster may impose reasonable
line 5 requirements on customers to prevent the collection or retention
line 6 of personal information that the person that operates a computing
line 7 cluster would not otherwise collect or retain, including a
line 8 requirement that a corporate customer submit corporate contact
line 9 information rather than information that would identify a specific

line 10 individual.
line 11 22606. (a) The Attorney General may bring a civil action for

line 12 a violation of this chapter and to recover all of the following:
line 13 (1) For a violation that causes death or bodily harm to another

line 14 human, harm to property, theft or misappropriation of property,
line 15 or that constitutes an imminent risk or threat to public safety that
line 16 occurs on or after January 1, 2026, a civil penalty in an amount
line 17 not exceeding 10 percent of the cost of the quantity of computing
line 18 power used to train the covered model to be calculated using
line 19 average market prices of cloud compute at the time of training for
line 20 a first violation and in an amount not exceeding 30 percent of that
line 21 value for any subsequent violation.
line 22 (2) For a violation of Section 22607 that would constitute a

line 23 violation of the Labor Code, a civil penalty specified in subdivision
line 24 (f) of Section 1102.5 of the Labor Code.
line 25 (3) For a person that operates a computing cluster for a violation

line 26 of Section 22604, for an auditor for a violation of paragraph (6)
line 27 of subdivision (e) of Section 22603, or for an auditor who
line 28 intentionally or with reckless disregard violates a provision of
line 29 subdivision (e) of Section 22603 other than paragraph (6) or
line 30 regulations issued by the Government Operations Agency pursuant
line 31 to Section 11547.6 of the Government Code, a civil penalty in an
line 32 amount not exceeding fifty thousand dollars ($50,000) for a first
line 33 violation of Section 22604, not exceeding one hundred thousand
line 34 dollars ($100,000) for any subsequent violation, and not exceeding
line 35 ten million dollars ($10,000,000) in the aggregate for related
line 36 violations.
line 37 (4) Injunctive or declaratory relief.

line 38 (5) (A) Monetary damages.

line 39 (B) Punitive damages pursuant to subdivision (a) of Section

line 40 3294 of the Civil Code.

89


-----

**17** **SB 1047**


line 1 (6) Attorney’s fees and costs.

line 2 (7) Any other relief that the court deems appropriate.

line 3 (b) In determining whether the developer exercised reasonable

line 4 care as required in Section 22603, all of the following
line 5 considerations are relevant but not conclusive:
line 6 (1) The quality of a developer’s safety and security protocol.

line 7 (2) The extent to which the developer faithfully implemented

line 8 and followed its safety and security protocol.
line 9 (3) Whether, in quality and implementation, the developer’s

line 10 safety and security protocol was inferior, comparable, or superior
line 11 to those of developers of comparably powerful models.
line 12 (4) The quality and rigor of the developer’s investigation,

line 13 documentation, evaluation, and management of risks of critical
line 14 harm posed by its model.
line 15 (c) (1) A provision within a contract or agreement that seeks

line 16 to waive, preclude, or burden the enforcement of a liability arising
line 17 from a violation of this chapter, or to shift that liability to any
line 18 person or entity in exchange for their use or access of, or right to
line 19 use or access, a developer’s products or services, including by
line 20 means of a contract of adhesion, is void as a matter of public
line 21 policy.
line 22 (2) A court shall disregard corporate formalities and impose

line 23 joint and several liability on affiliated entities for purposes of
line 24 effectuating the intent of this section to the maximum extent
line 25 allowed by law if the court concludes that both of the following
line 26 are true:
line 27 (A) The affiliated entities, in the development of the corporate

line 28 structure among the affiliated entities, took steps to purposely and
line 29 unreasonably limit or avoid liability.
line 30 (B) As the result of the steps described in subparagraph (A),

line 31 the corporate structure of the developer or affiliated entities would
line 32 frustrate recovery of penalties, damages, or injunctive relief under
line 33 this section.
line 34 (d) Penalties collected pursuant to this section by the Attorney

line 35 General shall be deposited into the Public Rights Law Enforcement
line 36 Special Fund established pursuant to Section 12530 of the
line 37 Government Code.
line 38 (e) This section does not limit the application of other laws.

line 39 22607. (a) A developer of a covered model or a contractor or

line 40 subcontractor of the developer shall not do any of the following:

89


-----

**SB 1047** **18**


line 1 (1) Prevent an employee from disclosing information to the

line 2 Attorney General or the Labor Commissioner, including through
line 3 terms and conditions of employment or seeking to enforce terms
line 4 and conditions of employment if the employee has reasonable
line 5 cause to believe the information indicates either of the following:
line 6 (A) The developer is out of compliance with the requirements

line 7 of Section 22603.
line 8 (B) An artificial intelligence model, including a model that is

line 9 not a covered model or a covered model derivative, poses an

line 10 unreasonable risk of causing or materially enabling critical harm,
line 11 even if the employer is not out of compliance with any law.
line 12 (2) Retaliate against an employee for disclosing information to

line 13 the Attorney General or the Labor Commissioner pursuant to
line 14 paragraph (1).
line 15 (3) Make false or materially misleading statements related to

line 16 its safety and security protocol in a manner that violates Part 2
line 17 (commencing with Section 16600) of Division 7 or any other
line 18 provision of state law.
line 19 (b) An employee harmed by a violation of this subdivision may

line 20 petition a court for appropriate temporary or preliminary injunctive
line 21 relief as provided in Sections 1102.61 and 1102.62 of the Labor
line 22 Code.
line 23 (c) (1) The Attorney General or Labor Commissioner may

line 24 publicly release or provide to the Governor any complaint, or a
line 25 summary of that complaint, pursuant to this section if the Attorney
line 26 General or the Labor Commissioner concludes that doing so will
line 27 serve the public interest.
line 28 (2) If the Attorney General or the Labor Commissioner publicly

line 29 releases a complaint, or a summary of a complaint, pursuant to
line 30 paragraph (1), the Attorney General or the Labor Commissioner
line 31 shall redact from the complaint any information that is confidential
line 32 or otherwise exempt from public disclosure pursuant to the
line 33 California Public Records Act (Division 10 (commencing with
line 34 Section 7920.000) of Title 1 of the Government Code) and any
line 35 information that the Attorney General or the Labor Commissioner
line 36 determines would likely pose an unreasonable risk to public safety
line 37 if it were disclosed to the public.
line 38 (d) A developer shall provide a clear notice to all employees

line 39 working on covered models and covered model derivatives of their
line 40 rights and responsibilities under this section, including the right

89


-----

**19** **SB 1047**


line 1 of employees of contractors and subcontractors to use the
line 2 developer’s internal process for making protected disclosures
line 3 pursuant to subdivision (e). A developer is presumed to be in
line 4 compliance with the requirements of this subdivision if the
line 5 developer does either of the following:
line 6 (1) At all times post and display within all workplaces

line 7 maintained by the developer a notice to all employees of their
line 8 rights and responsibilities under this section, ensure that all new
line 9 employees receive equivalent notice, and ensure that employees

line 10 who work remotely periodically receive an equivalent notice.
line 11 (2) No less frequently than once every year, provides written

line 12 notice to all employees of their rights and responsibilities under
line 13 this chapter and ensures that the notice is received and
line 14 acknowledged by all of those employees.
line 15 (e) (1) (A) A developer shall provide a reasonable internal

line 16 process through which an employee may anonymously disclose
line 17 information to the developer if the employee believes in good faith
line 18 that the information indicates that the developer has violated any
line 19 provision of Section 22603 or any other law, or has made false or
line 20 materially misleading statements related to its safety and security
line 21 protocol, or failed to disclose known risks to employees, including,
line 22 at a minimum, a monthly update to the person who made the
line 23 disclosure regarding the status of the developer’s investigation of
line 24 the disclosure and the actions taken by the developer in response
line 25 to the disclosure.
line 26 (B) The process required by this paragraph shall apply to

line 27 employees of the developer’s contractors and subcontractors
line 28 working on covered models and covered model derivatives and
line 29 allow those employees to disclose the same information to the
line 30 developer that an employee of the developer may disclose and
line 31 provide the same anonymity and protections against retaliation to
line 32 the employees of the contractor or subcontractor that apply to
line 33 disclosures by employees of the developer.
line 34 (2) The disclosures and responses of the process required by

line 35 this subdivision shall be maintained for a minimum of seven years
line 36 from the date when the disclosure or response is created. Each
line 37 disclosure and response shall be shared with officers and directors
line 38 of the developer whose acts or omissions are not implicated by
line 39 the disclosure or response no less frequently than once per quarter.
line 40 In the case of a report or disclosure regarding alleged misconduct

89


-----

**SB 1047** **20**


line 1 by a contractor or subcontractor, the developer shall notify the
line 2 officers and directors of the contractor or subcontractor whose acts
line 3 or omissions are not implicated by the disclosure or response about
line 4 the status of their investigation no less frequently than once per
line 5 quarter.
line 6 (f) This section does not limit protections provided to employees

line 7 by Section 1102.5 of the Labor Code, Section 12964.5 of the
line 8 Government Code, or other law.
line 9 (g) As used in this section:

line 10 (1) “Employee” has the same meaning as defined in Section

line 11 1132.4 of the Labor Code and includes both of the following:
line 12 (A) Contractors or subcontractors and unpaid advisors involved

line 13 with assessing, managing, or addressing the risk of critical harm
line 14 from covered models and covered model derivatives.
line 15 (B) Corporate officers.

line 16 (2) “Contractor or subcontractor” has the same meaning as in

line 17 Section 1777.1 of the Labor Code.
line 18 22608. The duties and obligations imposed by this chapter are

line 19 cumulative with any other duties or obligations imposed under
line 20 other law and shall not be construed to relieve any party from any
line 21 duties or obligations imposed under other law and do not limit any
line 22 rights or remedies under existing law.
line 23 22609. This chapter does not apply to the extent that it is

line 24 preempted by federal law.
line 25 SEC. 4. Section 11547.6 is added to the Government Code, to

line 26 read:
line 27 11547.6. (a) As used in this section, “critical harm” has the

line 28 same meaning as defined in Section 22602 of the Business and
line 29 Professions Code.
line 30 (b) There is hereby established the Board of Frontier Models.

line 31 The board shall be housed in the Government Operations Agency
line 32 and shall be independent of the Department of Technology. The
line 33 Governor may appoint an executive officer of the board, subject
line 34 to Senate confirmation, who shall hold the office at the pleasure
line 35 of the Governor. The executive officer shall be the administrative
line 36 head of the board and shall exercise all duties and functions
line 37 necessary to ensure that the responsibilities of the board are
line 38 successfully discharged.
line 39 (c) (1) Commencing January 1, 2026, the Board of Frontier

line 40 Models shall be composed of nine members, as follows:

89


-----

**21** **SB 1047**


line 1 (1)

line 2 _(A) A member of the open-source community appointed by the_

line 3 Governor and subject to Senate confirmation.
line 4 (2)

line 5 _(B) A member of the artificial intelligence industry appointed_

line 6 by the Governor and subject to Senate confirmation.
line 7 (3)

line 8 _(C) An expert in chemical, biological, radiological, or nuclear_

line 9 weapons appointed by the Governor and subject to Senate

line 10 confirmation.
line 11 (4)

line 12 _(D) An expert in artificial intelligence safety appointed by the_

line 13 Governor and subject to Senate confirmation.
line 14 (5)

line 15 _(E) An expert in cybersecurity of critical infrastructure appointed_

line 16 by the Governor and subject to Senate confirmation.
line 17 (6)

line 18 _(F) Two members who are academics with expertise in artificial_

line 19 _intelligence appointed by the Speaker of the Assembly._
line 20 (7)

line 21 _(G) Two members appointed by the Senate Rules Committee._

line 22 _(2) A member of the Board of Frontier Models shall meet all of_

line 23 _the following criteria:_
line 24 _(A) A member shall be free of direct and indirect external_

line 25 _influence and shall not seek or take instructions from another._
line 26 _(B) A member shall not take an action or engage in an_

line 27 _occupation, whether gainful or not, that is incompatible with the_
line 28 _member’s duties._
line 29 _(C) A member shall not, either at the time of the member’s_

line 30 _appointment or during the member’s term, have a financial interest_
line 31 _in an entity that is subject to regulation by the board._
line 32 _(3) A member of the board shall serve at the pleasure of the_

line 33 _member’s appointing authority but shall serve for no longer than_
line 34 _eight consecutive years._
line 35 (d) (1) On or before January 1, 2027, and annually thereafter,

line 36 the Government Operations Agency shall issue regulations to
line 37 update both of the following thresholds in the definition of a
line 38 “covered model” to ensure that it accurately reflects technological
line 39 developments, scientific literature, and widely accepted national
line 40 and international standards and applies to artificial intelligence

89


-----

**SB 1047** **22**


line 1 models that pose significant risk of causing or materially enabling
line 2 critical harms.
line 3 (2) The updated definition shall contain both of the following:

line 4 (A) The initial compute threshold that an artificial intelligence

line 5 model shall exceed to be considered a covered model.
line 6 (B) The fine-tuning compute threshold that an artificial

line 7 intelligence model shall meet to be considered a covered model.
line 8 (3) In developing regulations pursuant to this subdivision, the

line 9 Government Operations Agency shall take into account all of the

line 10 following:
line 11 (A) The quantity of computing power used to train covered

line 12 models that have been identified as being reasonably likely to
line 13 cause or materially enable a critical harm.
line 14 (B) Similar thresholds used in federal law, guidance, or

line 15 regulations for the management of artificial intelligence models
line 16 with reasonable risks of causing or enabling critical harms.
line 17 (C) Input from stakeholders, including academics, industry, the

line 18 open-source community, and government entities.
line 19 (e) (1) On or before January 1, 2027, and annually thereafter,

line 20 the Government Operations Agency shall issue regulations to
line 21 establish binding auditing requirements applicable to audits
line 22 conducted pursuant to subdivision (e) of Section 22603 of the
line 23 Business and Professions Code to ensure the integrity,
line 24 independence, efficiency, and effectiveness of the auditing process.
line 25 In developing regulations pursuant to this subdivision, the
line 26 Government Operations Agency shall take into account both of
line 27 the following:
line 28 (A) Relevant standards or requirements imposed under federal

line 29 or state law or through self-regulatory or standards-setting bodies.
line 30 (B) Input from stakeholders, including academics, industry, and

line 31 government entities, including from the open-source community.
line 32 (2) Any regulations issued pursuant to paragraph (1) shall, at a

line 33 minimum, be consistent with guidance issued by the U.S. Artificial
line 34 Intelligence Safety Institute and the National Institute of Standards
line 35 and Technology.
line 36 (f) (1) On or before January 1, 2027, and annually thereafter,

line 37 the Government Operations Agency shall issue guidance for
line 38 preventing unreasonable risks of covered models and covered
line 39 model derivatives causing or materially enabling critical harms,
line 40 including, but not limited to, more specific components of, or

89


-----

**23** **SB 1047**


line 1 requirements under, the duties required under Section 22603 of
line 2 the Business and Professions Code.
line 3 (2) Any guidance issued pursuant to paragraph (1) shall, at a

line 4 minimum, be consistent with guidance issued by the U.S. Artificial
line 5 Intelligence Safety Institute and the National Institute of Standards
line 6 and Technology.
line 7 (g) Regulations and guidance adopted pursuant to this section

line 8 shall be approved by the Board of Frontier Models before taking
line 9 effect.

line 10 SEC. 5. Section 11547.6.1 is added to the Government Code,

line 11 to read:
line 12 11547.6.1. (a) There is hereby established in the Government

line 13 Operations Agency a consortium that shall develop, pursuant to
line 14 this section, a framework for the creation of a public cloud
line 15 computing cluster to be known as “CalCompute.”
line 16 (b) The consortium shall develop a framework for creation of

line 17 CalCompute that advances the development and deployment of
line 18 artificial intelligence that is safe, ethical, equitable, and sustainable
line 19 by doing, at a minimum, both of the following:
line 20 (1) Fostering research and innovation that benefits the public.

line 21 (2) Enabling equitable innovation by expanding access to

line 22 computational resources.
line 23 (c) The consortium shall make reasonable efforts to ensure that

line 24 CalCompute is established within the University of California to
line 25 the extent possible.
line 26 (d) CalCompute shall include, but not be limited to, all of the

line 27 following:
line 28 (1) A fully owned and hosted cloud platform.

line 29 (2) Necessary human expertise to operate and maintain the

line 30 platform.
line 31 (3) Necessary human expertise to support, train, and facilitate

line 32 use of CalCompute.
line 33 (e) The consortium shall operate in accordance with all relevant

line 34 labor and workforce laws and standards.
line 35 (f) (1) On or before January 1, 2026, the Government

line 36 Operations Agency shall submit, pursuant to Section 9795, a report
line 37 from the consortium to the Legislature with the framework
line 38 developed pursuant to subdivision (b) for creation and operation
line 39 of CalCompute.

89


-----

**SB 1047** **24**


line 1 (2) The report required by this subdivision shall include all of

line 2 the following elements:
line 3 (A) A landscape analysis of California’s current public, private,

line 4 and nonprofit cloud computing platform infrastructure.
line 5 (B) An analysis of the cost to the state to build and maintain

line 6 CalCompute and recommendations on potential funding sources.
line 7 (C) Recommendations for the governance structure and ongoing

line 8 operation of CalCompute.
line 9 (D) Recommendations on the parameters for use of CalCompute,

line 10 including, but not limited to, a process for determining which users
line 11 and projects will be supported by CalCompute.
line 12 (E) An analysis of the state’s technology workforce and

line 13 recommendations for equitable pathways to strengthen the
line 14 workforce, including the role of CalCompute.
line 15 (F) A detailed description of any proposed partnerships,

line 16 contracts, or licensing agreements with nongovernmental entities,
line 17 including, but not limited to, technology-based companies, that
line 18 demonstrates compliance with the requirements of subdivisions
line 19 (c) and (d).
line 20 (G) Recommendations regarding how the creation and ongoing

line 21 management of CalCompute can prioritize the use of the current
line 22 public sector workforce.
line 23 (g) (1) The consortium shall, consistent with state constitutional

line 24 law, consist of 14 members selected from among all of the
line 25 following:
line 26 (A) Representatives of the University of California and other

line 27 public and private academic research institutions and national
line 28 laboratories.
line 29 (B) Representatives of impacted workforce labor organizations.

line 30 (C) Representatives of stakeholder groups with relevant

line 31 expertise and experience, including, but not limited to, ethicists,
line 32 consumer rights advocates, and other public interest advocates.
line 33 (D) Experts in technology and artificial intelligence to provide

line 34 technical assistance.
line 35 (E) Personnel from other relevant departments and agencies as

line 36 necessary.
line 37 (2) Eight members of the consortium shall be selected by the

line 38 Secretary of Government Operations, and the President Pro
line 39 Tempore of the Senate and the Speaker of the Assembly shall each
line 40 select three members.

89


-----

**25** **SB 1047**


line 1 (h) If CalCompute is established within the University of

line 2 California pursuant to subdivision (c), the University of California
line 3 may receive private donations for the purposes of implementing
line 4 CalCompute.
line 5 (i) This section shall become operative only upon an

line 6 appropriation in a budget act for the purposes of this section.
line 7 SEC. 6. The provisions of this act are severable. If any

line 8 provision of this act or its application is held invalid, that invalidity
line 9 shall not affect other provisions or applications that can be given

line 10 effect without the invalid provision or application.
line 11 SEC. 7. This act shall be liberally construed to effectuate its

line 12 purposes.
line 13 SEC. 8. The Legislature finds and declares that Section 3 of

line 14 this act, which adds Chapter 22.6 (commencing with Section
line 15 22602) to Division 8 of the Business and Professions Code,
line 16 imposes a limitation on the public’s right of access to the meetings
line 17 of public bodies or the writings of public officials and agencies
line 18 within the meaning of Section 3 of Article I of the California
line 19 Constitution. Pursuant to that constitutional provision, the
line 20 Legislature makes the following findings to demonstrate the interest
line 21 protected by this limitation and the need for protecting that interest:
line 22 Information in unredacted safety and security protocols and

line 23 auditor’s reports may contain corporate proprietary information
line 24 or information about covered models and covered model
line 25 derivatives that could threaten public safety if disclosed to the
line 26 public.

O

89


-----

