Transcript Highlights:

• Teachers are using AI systems to generate less plans tied to the state standards.
• We also have issues about are we over relying on AI content generation.
• And so it's important that if we're using AI systems, they're AI systems that actually support learning
• So is there actual AI coursework that teaches you how to learn AI and actual AI coursework that teaches
• you how to learn AI and use it?

Transcript Highlights:

• And then talking also about the AI that has already been in place that does not necessarily generate
• <00:04:12.480> image<00:04:12.920> that<00:04:13.040> is AI this is a a generated
• image that is AI this is a a generated image that is not<00:04:13.439> real<00:04:14.159>
• the promise of AI.
• the promise of AI.

Keywords: 1187, senate, all

Transcript Highlights:

• county from transferring money appropriated for the office of sheriff or constable to the county's general
• fund or any other county account. the county's general fund or any other county account, prohibiting
• In 1951, General MacArthur addressed Congress after being relieved by President Truman, and in addressing
• I'll fade away from the state senate, but I'm going to do my best General MacArthur impersonation.

Bills: SB 3, SB 16, SB 2, SB 5, SB 10, HB17, SB 54, SB 9, SB 7, SB 17, SB 4, HB8, HB25, HB26, HB192, HB8, HB25, HB26, HB48, HB149, HB192, HB254, HB17, SB54, SB9, SB7, SB17, SB4, SB3, SB16, SB2, SB5, SB10

Keywords: flash flood, flood warning, outdoor warning siren, emergency alert, disaster preparedness, flood mitigation, Hill Country floods, Texas Water Development Board, municipalities, counties, local government mandate, public safety, grant program, backup power, weather siren, emergency management, flood-prone area, warning system, real property theft, real property fraud

Transcript Highlights:

• The Secretary of State must request the assistance of the Office of Attorney General to determine if
• filed and a termination statement is filed with respect to the lien, the Office of the Inspector General
• State to refuse a document filed to create a lien on real property if the Office of the Attorney General

Bills: SB 3, SB 16, SB 2, SB 5, SB 10, HB17, SB 54, SB 9, SB 7, SB 17, SB 4, HB8, HB25, HB26, HB192, HB8, HB25, HB26, HB48, HB149, HB192, HB254, HB17, SB54, SB9, SB7, SB17, SB4, SB3, SB16, SB2, SB5, SB10

Keywords: flash flood, flood warning, outdoor warning siren, emergency alert, disaster preparedness, flood mitigation, Hill Country floods, Texas Water Development Board, municipalities, counties, local government mandate, public safety, grant program, backup power, weather siren, emergency management, flood-prone area, warning system, real property theft, real property fraud

Bills: SB 3, SB 16, SB 2, SB 5, SB 10, HB17, SB 54, SB 9, SB 7, SB 17, SB 4, HB8, HB25, HB26, HB192, HB8, HB25, HB26, HB48, HB149, HB192, HB254, HB17, SB54, SB9, SB7, SB17, SB4, SB3, SB16, SB2, SB5, SB10

Keywords: flash flood, flood warning, outdoor warning siren, emergency alert, disaster preparedness, flood mitigation, Hill Country floods, Texas Water Development Board, municipalities, counties, local government mandate, public safety, grant program, backup power, weather siren, emergency management, flood-prone area, warning system, real property theft, real property fraud

Transcript Highlights:

• They're advising us on how to appropriately use generative AI in the practice of law.
• The bill also adds digital and AI-generated images to the definition of sexual exploitation under the
• Those photos can be taken off of real children and then used to generate AI images and video and live
• California must be clear that child sexual exploitation in any form, streamed, digital, AI-generated,
• , and we've seen agentic AI, generative AI, all the AIs, as they reflect, as they impact human society

Summary: The committee heard several bills focused on artificial intelligence, child safety, mental health, and privacy. SB 574 by Senator Umberg would require transparency and human oversight when attorneys, judges, and court neutrals use AI; it drew support from privacy advocates and committee members, with no opposition. SB 1276, the End Child Exploitation Act, would update child sexual exploitation laws to cover live-streamed and AI-generated abuse material and clarify that viewing such content can be criminally punishable; prosecutors, child advocacy groups, and others supported it, while no one appeared in opposition despite opposition on file. SB 813 would create a California AI standards and safety commission and a voluntary two-tier certification framework for AI safety standards; supporters said it would create scalable, independent oversight, while TechNet and CalChamber opposed it as duplicative, under-defined, and likely to create a de facto mandate. The committee discussed market pressure, federal preemption concerns, and the role of voluntary standards, but no final vote was taken in the excerpt. Senator Padilla also presented SB 300, which would strengthen protections for minors from sexually explicit chatbot content by moving from a reasonableness standard to an affirmative duty to prevent such exposure and to prohibit facilitation. Supporters said new evidence showed greater risks and that companies can and should build stronger safeguards; opponents, including TechNet and CCIA, argued the bill was premature because SB 243 had only recently taken effect and warned it could create strict-liability-like exposure. Padilla then presented SB 903, which would bar AI chatbots from being advertised as therapists, require licensed clinician oversight and informed consent for AI use in psychotherapy, and protect patient confidentiality; it received broad support from mental health professionals and labor groups, while industry and health associations were opposed unless amended over triage and crisis-detection language. The committee members emphasized the need for human judgment in mental health care and noted ongoing negotiations on amendments. The committee also heard SB 1119, a companion to AB 2020, which would require annual risk assessments, crisis-response protocols, default child settings, parental controls, limits on data use, public incident reporting, and third-party audits for chatbots used by children. Supporters argued the bill would address documented harms and improve transparency, while industry groups objected to ambiguous standards, liability exposure, and the private right of action. A roll call vote was taken on SB 1119 after quorum was established; the motion to pass to Appropriations succeeded on a 5-1 vote, with one no vote and the measure left on call for absent members. Finally, SB 354, a privacy bill for insurance consumers, would modernize outdated insurance privacy rules, bar sale of personal information, and expand consumer rights to know, correct, and delete data. Supporters said it would implement Proposition 24’s privacy mandate, while a large coalition of insurers, agents, brokers, and related businesses opposed unless amended, mainly seeking a small-business exemption and narrower treatment of publicly available information; members and the author said negotiations were ongoing and the bill had already been substantially revised.

Transcript Highlights:

• And then talking about also the AI that has already been in place, that doesn't necessarily generate
• c><00:02:09.879> actually what generative AI is where it actually what generative AI is where
• the promise of AI.
• the promise of AI.
• AI.

Keywords: 1187, senate, all

Transcript Highlights:

• AI are supercomputers.
• AI are supercomputers. They are enough. AI are supercomputers.
• I think that AI and the forever.
• <00:10:51.360> tools encourage revisions to allow AI tools encourage revisions to allow AI
• about is is that AI AI chat bots should about is is that AI AI chat bots should have<00:21:11.520

Keywords: 1183, house

Transcript Highlights:

• AI are supercomputers.
• AI are supercomputers. not good enough. AI are supercomputers.
• The enforcement with the attorney general is for the AI itself if it's found to be in violation.
• The enforcement with the attorney general is for the AI itself if it's found to be in violation.
• The enforcement with the attorney general is for the AI itself if it's found to be in violation.

Keywords: 919, house, all

Summary: House File 3893, which would regulate the use of artificial intelligence in psychotherapy and counseling, was heard with an A3 amendment adopted by voice vote at the start of the meeting. The amendment was described as the product of stakeholder discussions and was intended to clarify the roles of licensing boards, the attorney general, and other stakeholders, including language about large corporations versus individual practitioners. The bill author argued that AI chatbots should not be allowed to provide psychotherapy or counseling, citing concerns about suicide risk, the need for licensed human professionals, and examples of harmful chatbot interactions. Supportive testimony came from Eric Meshy of SAVE, who said AI cannot assess risk or intervene in crises, and from Dr. Steven Gerardo of the Minnesota Psychological Association, who said AI is designed to keep users engaged and monetize conversations rather than provide treatment. A virtual testifier from TechNet supported the bill’s overall goal but urged clearer definitions so the measure would not unintentionally restrict supervised, administrative, or wellness uses of technology, and asked for clearer limits around emotional distress and AI’s role in treatment decisions. Members raised several concerns and questions, including why religious counseling was exempted, whether the bill could conflict with interstate licensure compacts, privacy issues around AI used for scheduling or administrative support, and whether chatbot screening language would itself require analyzing sensitive mental health disclosures. The author said the bill was still being refined and that the attorney general language was intended to apply to AI companies, while boards would handle individual licensed professionals. After discussion, the committee renewed the motion to re-refer House File 3893 as amended to the Committee on Commerce, Finance, and Policy, and the motion prevailed by voice vote.

Transcript Highlights:

• GENERATIVE AI TOOLS.
• Porras: THANK YOU FOR THE PRESENTATION I AM FASCINATED BY AI IN GENERAL.
• MOVING ON TO THE ADMINISTRATION SIDE, RIGHTFULLY GENERATIVE AI, CHAT GDP IS THE HIGHLIGHT OF THE DAY.
• EU WE PROVIDED GUIDELINES WITH FACULTY RESEARCHERS AND GENERAL COUNSEL ON HOW TO USE AI BOTH INSIDE AND
• IF WE ARE NOT CAREFUL THEN THE GENERATIVE AI TOOLS WILL BASICALLY INSPIRE THE STUDENTS TO UPDATE THEIR

Transcript Highlights:

• AI in the practice of... ...they're advising us on how to appropriately use generative AI in the practice
• The bill also adds digital and AI-generated images to the definition of sexual exploitation under the
• Those photos can be taken off of real children and then used to generate AI images and video and live
• And noted that it is a part of generative AI, correct? It can be.
• And we've seen agentic AI, generative AI, all the AIs, as they affect human society.

Keywords: 988, house, all

Transcript Highlights:

• everything um from you know generative everything um from you know generative AI<01:09:34.000>
• that is uh the gener the AI revolution. that is uh the gener the AI revolution.
• > which of AI generated content is which of AI generated content is which includes<01:24:51.120
• AI model or input to train a generative AI model or an<01:24:56.960> employer<01:24:57.440>
• AI tools that are output of generative AI tools that are state<01:25:54.239> sanctioned<01:25

Keywords: 1183, house

Transcript Highlights:

• Um, obviously, data centers are needed to, you know, generate AI, particularly generative AI.
• Um, obviously, data centers are needed to, you know, generate AI, particularly generative AI.
• Um, obviously, data centers are needed to, you know, generate AI, particularly generative AI.
• You know, generate AI, particularly generative AI.
• /c> at AI in total, what it generates, what at AI in total, what it generates, what it<01:31:19.360><

Keywords: 1187, senate, all

Transcript Highlights:

• It is AI.
• These are what are called AI agents. These are what are called AI agents.
• the AI.
• And so the thing is this is also not just an AI that is just arresting AI.
• This is a proactive AI to arresting AI.

Keywords: 958, all

Summary: A presenter from Fast Health Corporation described a proposed Kentucky Health Command System tied to Senate Bill 175, which would create a state-sanctioned AI platform for rural hospitals and telehealth. The company said the system would help rural residents get health information remotely, triage minor issues, and escalate more serious cases to Kentucky providers, with use cases including blood pressure, diabetes, maternity care, smoking cessation, and other preventive-care topics. The presenter argued the system would help rural hospitals compete with out-of-state telehealth companies and keep patients connected to local care. The presentation also emphasized a commercial model the sponsor said would generate new revenue through ads and branded interactions, with the bill reportedly directing 80% of that revenue to rural hospitals and 20% to the state to maintain the system. The presenter said the technology would augment, not replace, doctors and nurses, and claimed it could improve access and convenience in underserved areas. Committee members raised concerns about liability, whether the AI could provide medical advice, and whether there was evidence it had reduced emergency room visits; the presenter said the system could not give medical advice and acknowledged the technology is still very new. The sponsor of the bill said the goal was to help transform rural health care, reduce unnecessary ER use, and capture revenue that would otherwise go to commercial search engines and out-of-state companies. No vote or final action was taken during the portion of the meeting provided, and the discussion ended with questions about branding, loyalty, and the legal limits of the AI system.

Transcript Highlights:

• AI is not a therapist.
• district on safe use of AI with kids. district on safe use of AI with kids.
• generate their own recommendations. generate their own recommendations.
• generated the image?
• You know, I saw earlier this week, I think her name was Tilly Norwood, the first AI-generated actress

Keywords: 958, all

Summary: The committee met with a quorum, approved the prior meeting minutes, and then heard testimony on the use of artificial intelligence in therapy and mental health settings. Representative Lisa Willner and Brenda Rosen of NASW Kentucky argued for “guard rails” on AI chatbots so they cannot present themselves as licensed therapists or replace school counselors, psychologists, or social workers. They said AI can support licensed professionals, but warned that chatbots cannot reliably recognize nonverbal cues, escalate crises, or provide accountable care, and they cited examples of harmful chatbot interactions, including a suicide case and a chatbot telling a user to “Please die.” The witnesses also raised concerns about data privacy, commercialization of sensitive mental health conversations, and the use of personal clinical content to train AI models. They said minors should require parental consent and suggested transparency about how a chatbot is trained and who created it. They distinguished between unvetted consumer chatbots and AI tools that have been scientifically validated or approved as digital therapeutics, noting that some evidence-based tools may be useful for specific conditions such as depression, anxiety, or eating disorders. Committee members asked whether regulation should be handled by the legislature or by professional boards, and whether a multi-state model would be preferable to 50 different state approaches. The witnesses generally favored expert-led standards and said a board or panel of experts could review and approve mental health chatbots, but members cautioned that boards can become too restrictive and that legislation should preserve flexibility and avoid discouraging children from seeking help. The discussion ended with a request for the witnesses to restate their proposed policy ideas, including privacy protections, bans on commercialization, limits on training AI with clinical content, transparency requirements, and informed consent.

Transcript Highlights:

• For example, if you go into Facebook today, you can ask it to make an AI-generated picture of you with
• This is true for many generative AI programs that we're familiar with, but there are apps, platforms,
• generative AI that maybe some of us use doesn't have this turned on.
• <00:16:20.480> generative<00:16:20.880> AI would say commercial uh AI generative AI
• would say commercial uh AI generative AI that<00:16:21.680> maybe<00:16:22.000> some<00

Keywords: 1187, senate, all

Transcript Highlights:

• The bill addresses how social media displays these AI images and gives credit to our AI companies that
• But it gives credit to those AI companies who are embedding in all their AI images that their technology
• provenance data or the digital tag that's attached to these images, you can, you know, if you're a user of AI-generated
• So the Attorney General, can we talk about the Attorney General involvement in this?
• One, the Attorney General gives notice to the AI company or the social media company, saying, “I believe

Summary: The Information Technology Budget and Policy Subcommittee met to consider House Bill 369, relating to digital content provenance and AI-generated images, sponsored by Representative McFarland. McFarland explained that the bill is intended to address concerns about people being misled by AI-generated content, especially on social media, by preserving digital provenance data or watermark-like signatures attached to AI images. He said the measure is meant to support transparency without stifling innovation. A strike-all amendment was adopted that narrowed the bill in response to industry feedback. It removed an initial pilot program for the Division of Emergency Management, deleted provisions dealing with AI in elections, tightened definitions, and delayed the effective date to July 2026. Members asked about user control over provenance data, whether the data could include personal information, and how enforcement would work. McFarland said users may add extra information but are not required to, and that enforcement would begin with a complaint to the Attorney General, followed by notice and a 30-day cure period before possible action under the Fair, Unfair, and Deceptive Trade Practices Act. Public testimony included support from an artist advocate, who said AI transparency is needed to protect creators, and support statements from Adobe and Comcast. TechNet appeared for information only and said it appreciated the changes, including the removal of the capture device, the 30-day cure period, and the delayed effective date. The amendment was adopted, and HB 369 was then reported favorably by a 17-0 vote. After the bill, the chair presented the subcommittee’s recommended 2025-2026 IT budget, totaling $528 million, including $146.5 million in general revenue and $382.2 million in trust funds. The proposal reflected a roughly 13% decrease from the prior year and emphasized funding for major ongoing projects such as the PALM/FLAIR replacement, the AHCA FX program, the Department of Revenue’s CAMS modernization, and the APD I-Connect system. No questions were raised, and the meeting adjourned without further business.

Transcript Highlights:

• And so these shouldn't be AI projects for AI projects' sake.
• And so these shouldn't be AI projects for AI projects sake.
• AI for protecting privacy.
• That's kind of where I'm at in policy in general, but especially with this undiscovered country of AI
• . ...and policy in general, but especially with this undiscovered country of AI.

Bills: HB2452, HB2592

Keywords: comprehensive plans, county governance, data centers, zoning, state regulations, land use, energy production, infrastructure, environmental impact, artificial intelligence, government regulation, innovation, procurement, administrative burdens, 1182, all

Summary: The House Artificial Intelligence and Innovation Committee heard a presentation from Steven Garrison of the Cicero Institute on how Arizona state government could use AI to improve service delivery, reduce costs, and streamline procurement. He argued that AI should be treated as software deployed across many use cases, not just chatbots, and urged the state to identify opportunities proactively, avoid unnecessary agency-level AI rules, use existing procurement structures, and measure outcomes. Members asked about workforce impacts, agency involvement, privacy, and the balance between innovation and regulation; Garrison said AI would likely augment workers first, create new jobs over time, and should be guided by the legislature rather than broad agency rulemaking. The committee then considered HB 2592, which directs the Department of Administration to have budget units identify AI opportunities and streamline implementation, and an amendment adding reporting requirements to state leaders and the Secretary of State. After limited public testimony, the amendment was adopted and the bill received a due pass recommendation on a 4-2 vote, with one member absent. The committee next took up HB 2452, which would add data centers and small modular reactors to county comprehensive planning and adjust county land-use planning requirements, including changes related to renewable energy planning in larger counties. Supporters said the bill would help counties plan ahead for data center growth and future energy needs, including SMRs and other emerging technologies. County representatives opposed the measure, arguing it improperly singled out specific uses, blurred the line between comprehensive planning and zoning, and reduced local control by altering established public planning processes; they also raised concerns about vague standards and the bill’s treatment of renewable energy. After debate, the committee approved HB 2452 on a 4-3 vote for a due pass recommendation and then adjourned.

Transcript Highlights:

• a minor account holder, and reasonable measures to prevent the conversational AI service from generating
• It's not that it's AI. It's just that AI fools us better than anything else.
• Don't try to define AI. Because, yeah, we don't have any AI experts here in this legislature.
• things that should apply for every conversation with AI in general.
• Those AI models.

Bills: HB2056, HB2057, HB2224, HB2257, HB2265, HB2311, HB2352, HB2367, HB2369, HB2446, HB2532, HB2592, HB2752, HB2809, HB2957

Keywords: brackish water, groundwater, desalination, water resources, feasibility study, environmental impact, special plates, centennial, Arizona Department of Transportation, vehicle registration, commercial discounts, Arizona Centennial, produce incentive, agriculture, economic support, funding, Arizona, boating, watercraft licensing, law enforcement

Summary: The committee approved the March 10, 2026 minutes and then considered several bills. HB 2056, appropriating $100,000 for an Arizona Department of Water Resources feasibility study on brackish groundwater desalination, passed 6-3 despite concerns that it could encourage more groundwater pumping and worsen subsidence or impacts to tribal lands. HB 2057, which lowers the Arizona Centennial Special Plate renewal fee from $25 to $20, also received a do-pass recommendation, 7-2, with some members objecting to the special plate’s focus and the reduced fee. HB 2224, appropriating $1 million annually to DES for the produce incentive program, drew strong support from nonprofit and food-access advocates who said the Double Up Food Bucks program helps families, farmers, and local markets; it passed 9-0. HB 2257, changing the allocation of watercraft registration revenues among the Watercraft Licensing Fund, Lake Improvement Fund, and boating safety/law enforcement funds, passed 6-3 after criticism that the bill was heard without a sponsor present and concern about shifting money away from Game and Fish. HB 2265, barring courts from charging defendants for public defender or court-appointed counsel costs and related criminal-case fees, passed 7-1 amid debate over constitutionality, cost shifting to counties, and whether the bill should be amended to offset lost revenue. The committee then took up HB 2311, which imposes disclosure, content, parental-control, and self-harm response requirements on publicly available conversational AI services for minors. The sponsor and supporters framed it as child protection, while opponents warned about privacy, surveillance, and overbroad definitions; an Epstein amendment expanding the bill to broader conversational technology was rejected, and the bill passed 7-2. HB 2752, moving authority over international trade offices from the Arizona Commerce Authority to legislative appropriations and reporting, passed 6-3 after a proposed data-and-research amendment was also rejected. Finally, HB 2957, requiring ADOT to continue offering non-REAL ID licenses and limiting biometric collection and data sharing, prompted extensive testimony for and against; supporters emphasized privacy and choice, while opponents warned it could conflict with federal CDL and REAL ID requirements. The transcript cuts off during the discussion of HB 2957, before a final vote is shown.

Transcript Highlights:

• But generative AI is the most exciting technology I've seen.
• But just in terms of generative AI, it is moving very, very quickly.
• And the use of generative AI in this environment is... ...came as a surprise to me.
• of AI technology; and four, preventing AI-generated evidence tampering in our courtrooms.
• of AI technology; and four, preventing AI-generated evidence tampering in our courtrooms.

Summary: The committee heard several AI- and consumer-protection-related bills, with extensive testimony from authors, supporters, and industry opponents. SB 53 by Senator Wiener would create transparency requirements for large AI developers, including disclosure of safety and security protocols, reporting of critical safety incidents, whistleblower protections, and the CalCompute public cloud. Supporters said it is a narrower, transparency-based follow-up to last year’s vetoed AI safety bill, while opponents argued it still relies too much on company size, could expose trade secrets, and should be narrowed further. The committee approved SB 53 on a do-pass-as-amended vote to Appropriations, with the roll held open for absent members. SB 766 by Senator Allen would codify the FTC’s Cars Rule and create a three-day cooling-off period for certain used-car purchases, along with stronger disclosure rules on pricing, add-ons, and government affiliation claims. Supporters said it would save consumers money and time and help buyers avoid bad deals, while dealer and industry groups said amendments addressed many of their concerns. Several former opponents moved to neutral, and the committee passed SB 766 unanimously as amended to Appropriations. SB 7 by Senator McNerney would regulate automated decision-making systems in employment by requiring notice, human review for discipline and termination, and limits on predictive behavior analysis. Labor and consumer advocates supported the bill as a safeguard against biased or overly automated workplace decisions, while employer and industry groups raised concerns about scope, notice burdens, and the predictive-analysis ban. The committee passed SB 7 to Appropriations on a 4-2 vote, with the roll held open. SB 833, also by Senator McNerney, would require human oversight of AI used in critical infrastructure, along with training and system assessments; it drew limited opposition focused on scope, and the committee passed it as amended to Appropriations on a 5-0 vote, also holding the roll open. Later, the committee took up SB 11, which would address AI-generated voice, image, and video cloning and deepfakes by clarifying likeness protections, requiring consumer warnings, and addressing misuse and evidence tampering. Supporters framed it as a targeted response to nonconsensual deepfakes, while industry groups said recent amendments improved the bill but still had concerns about penalties and warning language. The committee also heard SB 720, the Safer Streets Act, which would let cities opt into a revised red-light camera system that shifts from driver to owner liability, removes facial identification, makes violations civil rather than criminal, and directs revenue toward transportation safety projects; the author presented the bill, but the transcript ends before any final action on SB 720.