Transcript Highlights:

• </c><01:19:11.760><c> It's</c> collaborative intelligence. It's collaborative intelligence.
• </c> artificial intelligence in our teaching. artificial intelligence in our teaching.
• </c><01:25:37.280><c> to</c> plants um use artificial intelligence to plants um use artificial intelligence
• </c> bachelors in artificial intelligence. bachelors in artificial intelligence.
• </c> who believe that artificial intelligence who believe that artificial intelligence can<01:32:58.880

Summary: The Intelligence Task Force met with a quorum present and approved the minutes from the August 14 meeting. The first presentation was from the National Retail Federation on artificial intelligence in the retail sector. NRF described AI use cases in three areas: consumer-facing tools such as chatbots, product descriptions, and marketing; employee-facing tools such as company-specific apps trained on employee manuals; and internal uses such as supply chain forecasting, inventory planning, fraud detection, cybersecurity, and coding support. The presenters said retail has been among the top deployers of AI and emphasized that the goal is to improve both customer and employee experience. NRF also outlined its principles for responsible AI use, organized around governance and risk management, customer engagement and trust, workforce applications, and business partner accountability. The group said retailers should maintain strong internal oversight, be transparent with customers, ensure compliance with existing laws, closely review workforce-related uses because they can be high-risk, and clearly define responsibilities between retailers and third-party developers. In policy discussion, NRF urged a tiered approach focused first on high-risk uses such as hiring, health care, financial, mortgage, and rental decisions, warned against fragmented rules that could favor only the largest firms, and encouraged voluntary standards and worker upskilling incentives. In response to a question about dynamic pricing, NRF said its members do not target consumers based on pricing and offered to follow up with more information. The committee then moved to a follow-up discussion on energy policy and data centers, hearing from Bartley Cleland of NetChoice. He explained that AI runs largely in data centers and that cloud computing shifts processing and storage away from individual devices, which he argued improves efficiency and can reduce costs. He said electricity demand has been rising over time and that AI will increase compute-related electricity use, but framed electricity as a normal input to economic growth. No votes or formal actions were taken after these presentations.

Transcript Highlights:

• I've overseen these efforts to collect and disseminate intelligence across the U.S. government.
• DEA did not provide intelligence.
• There was no DEA intelligence that was passed to Colombia.
• Was there ever a determination by DEA that the intelligence leak led to the killing? No, sir.
• Did you directly transfer the intelligence to the Mexican authorities? No, sir, I did not.

Committee: Senate Senate Judiciary

Keywords: fentanyl crisis, DEA, drug trafficking, public safety, Judea Serrata, Terrence Cole, judicial security, social media, drug prevention

Summary: The committee meeting centered on the urgent crisis of fentanyl abuse and the government’s response to the drug trafficking epidemic. Key discussions revolved around the nomination of Terrence Cole as the Administrator of the DEA and Judea Serrata as Director of the U.S. Marshals Service. Cole emphasized his commitment to combating the fentanyl crisis, highlighting the rising number of overdose deaths and the necessity of bipartisan efforts to dismantle drug cartels. Members engaged in dialogue on the effectiveness of various strategies and the importance of education aimed at preventing drug-related deaths among youth. Additionally, the challenges of social media in drug trafficking were highlighted, indicating a need for greater accountability in how platforms handle such issues. Senators voiced concerns about the impact of administration policies on judicial security and the resources available for marshals to perform their critical roles.

Transcript Highlights:

• know enough about all the different nuances of government litigation in that arena to speak to it intelligently
• know enough about all the different nuances of government litigation in that arena to speak to it intelligently
• know enough about all the different nuances of government litigation in that arena to speak to it intelligently
• know enough about all the different nuances of government litigation in that arena to speak to it intelligently
• . intelligently. intelligently.

Bills: HB42 , SB246 , SB38 , SB194 , SB183 , SB253 , SB160

Committee: Senate Fiscal Responsibility and Economic Development

Keywords: unclaimed vehicles, towing, Department of Revenue, notification requirements, abandoned vehicles, real estate, residential property, single-family home, homeowner protection, property solicitation, unsolicited offer, wholesaling, real estate wholesaling, assignment contract, equitable interest, investment property, house flipping, predatory real estate practices, title cloud, deceptive trade practices

Transcript Highlights:

• programs in fast-growing, mission-critical fields, starting with cybersecurity and artificial intelligence
• doctorate degrees in high-demand, rapidly evolving fields. not only cybersecurity and artificial intelligence
• So cybersecurity and artificial intelligence seem very defined, but then after that, on line 2.2, it
• other emerging areas of practice and workforce need" if that related back to IT and artificial intelligence
• And if it's directly related to cybersecurity and artificial intelligence, then I'm fine with it because

Committee: House Higher Education Finance and Policy

Transcript Highlights:

• Recently, we've been permitted to increase our ISR, our intelligence. surveillance, and reconnaissance
• with Mexico to go address the cartel violence in terms of— sending more troops and intelligence sharing
• So intelligence and awareness of what the terrorists are thinking and planning and plotting, I think,
• We've supported from a logistics, intelligence, and training equipment standpoint with the Colombia and
• You both know about the importance of artificial intelligence systems and the pivotal role that they

Committee: Senate Subcommittee on Personnel

Keywords: national security, China, Russia, cybersecurity, border security, military deployment, transnational crime, defense modernization, military strategy

Summary: In this meeting, key topics included national security concerns with an emphasis on threats from China, Russia, and transnational criminal organizations. Senior military officials provided testimony on the evolving risks posed by these adversaries, particularly focusing on cybersecurity and advancements in weaponry. The discussion highlighted the importance of a cohesive strategy among government departments to effectively respond to the challenges presented by these threats, emphasizing the need for modernization in defense capabilities. Committee members grilled the witnesses on border security and the implications of deploying military resources in domestic operations.

Transcript Highlights:

• In November 2025, six Democrat lawmakers with military or intelligence backgrounds, known as the 'seditious
• I think that in this sphere we've talked a lot about artificial intelligence, and especially militarily
• , to win— and obviously this is bigger than just Arizona— but to win the artificial intelligence race
• the things that we're doing to work together to make sure that we are the leader in artificial intelligence
• It's funny that you mentioned the border and artificial intelligence.

Committee: Senate Director Nominations

Transcript Highlights:

• The analogy I've been using is that artificial intelligence is like going from the Ford Model T to a
• Artificial intelligence is going to have a lot of great applications, but the resource footprint that
• And so what are the right regulatory frameworks for artificial intelligence?
• It's very difficult to look under what's called the black box of artificial intelligence algorithms and
• Whether an artificial intelligence is giving biased outputs by proactively testing it.

Committee: House Investments & Pensions Oversight Committee

Transcript Highlights:

• the bill in the shape that I would like it and briefly, um, it expands the offenses to which an intelligent
• It was then that I learned about intelligent speed assist devices or ISA.
• </c> intelligent speed assist devices or ISA. intelligent speed assist devices or ISA.
• Intelligent speed assist is a no-nonsense, effective solution. Please support House File 3429.
• </c> speeders bill and require intelligent speeders bill and require intelligent speed<00:11:30.399><

Transcript Highlights:

• First, a cyber threat intelligence center collects feeds from federal partners.
• monitoring, but also proactively Also, you know, developing tools, education, intelligence monitoring
• ...military intelligence deployed abroad? What's the risk? It's not going to be perfect, obviously.
• These intrusions are not about intelligence collections. They are about leverage.
• As technology evolves, tools like artificial intelligence are becoming more common. Thank you, Mr.

Bills: HB146 , HB150 , HB1500 , HB1545 , HB1562 , HB2067 , HB2520 , HB2818 , HB3214 , HB3250 , HB3466 , HB3512 , HB3623 , HB4063 , HB4395 , HB4464 , HB4668 , HB4690 , HB5331 , HB3833 , HB146 , HB150

Committee: Senate Business & Commerce

Summary: The committee took up several pending business items and reported a series of House bills out of committee, including HB 2467, HB 2468, HB 2518, HB 4310, HB 4386, HB 4490, HB 5323, and HB 149. Most of these were advanced on committee substitute motions and sent to the local and uncontested calendar or reported favorably to the full Senate. HB 2467 drew one nay vote, while the others were approved without opposition. HB 4310 and HB 4386 were described as committee-substitute versions with changes narrowing disclosure requirements and preserving attorney-client privilege in certain circumstances. A major portion of the meeting focused on HB 149, an AI governance bill. The substitute was explained as addressing biometric identifier capture and storage, exempting certain AI uses for security and fraud prevention, clarifying definitions, restricting AI systems that simulate explicit child sexual content, adjusting Attorney General investigative authority, refining sandbox program waivers, reducing Texas AI Council powers and membership, and adding DIR coordination provisions. The committee adopted the substitute and reported the bill favorably. The committee then heard extensive testimony on HB 1500, the DIR sunset bill. The author said the bill would continue DIR for 12 years, restructure its board, update advisory committees, require regular cybersecurity assessments and penetration testing for state agencies, improve IT procurement training, and transfer the e-grants program to the Comptroller. A Texas 2036 witness supported the bill as a way to strengthen governance, procurement, and cybersecurity. Members asked detailed questions about the bill’s structure and then left HB 1500 pending. The committee also heard a lengthy presentation on HB 150, which would create the Texas Cyber Command as a component of the University of Texas System, administratively attached to UTSA and located in San Antonio. The author argued the command would centralize cyber threat intelligence, incident response, and digital forensics, and would be able to support state and local entities, with optional services for local governments. Members raised concerns about university mission drift, governance, security, chain of command, procurement authority, gifts and donations, and civil liberties implications of proactive cyber monitoring. Witnesses from UTSA/NSCC and SecurityScorecard testified in support, emphasizing the security of the downtown San Antonio facility, the existing cyber ecosystem there, and the need for a dedicated cyber capability. The bill remained under discussion with no final committee action announced in the excerpt.

Transcript Highlights:

• We stood up, at the direction of the governor, an intelligence coordination center.
• Intelligence has really been made more robust through the development of the Intelligence Coordination
• Intelligence has really been made more robust through the development of the intelligence coordination
• The Intelligence Coordination Center has reached initial operating capability.
• So it would be a joint intelligence subcommittee.

Summary: The committee first heard an update on the Northwest Louisiana earthquake cluster. Laura Sori of the Department of Conservation and Energy said the agency has inspected Class II injection wells within 12 miles of the earthquakes, found no permit violations, and is requiring monthly reporting of daily injection data. LSU and Tulane researchers explained that the swarm includes about 50 earthquakes detected by USGS since December 2025, including a 4.9 magnitude event on March 5, and that better monitoring is needed because Louisiana has very limited seismic station coverage. Dr. Cynthia Ebinger said her temporary array has detected more small quakes than USGS, that the pattern looks more like a swarm than normal aftershocks, and that the data suggest pressure changes in the subsurface, though no definitive cause was identified. Keith Hall of LSU described how other states responded to suspected induced seismicity with more monitoring, more frequent reporting, injection limits or moratoria, and “traffic light” systems that escalate regulatory responses as seismicity increases. Members asked about depths, fault locations, possible links to injection or fracking, and whether more monitoring and data-sharing should be pursued; several speakers said Louisiana likely needs a denser monitoring network and more structured data collection. A Texas geoscientist, William Berger, also testified that Texas uses large-scale data analysis and AI to study injection-related seismicity and argued for secure sharing of operator data to improve forecasting and risk management. The committee then took testimony on UAV and drone incursions over Barksdale Air Force Base. GOSEP said the incident was logged in WebEOC and the common operating picture, but that Barksdale did not request direct GOSEP resources and that the matter was handled through law enforcement channels. Louisiana State Police and the FBI said they were limited in what they could disclose, but confirmed multiple drone sightings on the morning and evening of March 9 and continued monitoring for several days. State Police said they have created a task force with the Police Chiefs Association, Sheriffs Association, GOSEP, and LSP, and that officers are receiving FBI-related training to help detect and, where authorized, mitigate drones. Members discussed whether the activity was nefarious, what counts as an incursion, and the need for better public education about drone restrictions near military and critical infrastructure sites. No formal action was taken, but members said the issue will continue to be tracked alongside pending legislation. Finally, the committee received a one-year update on the merger of GOSEP with the Louisiana National Guard and Military Department. Major General Thomas Freelieu and Brigadier General Jason Maffus said the merger has reduced GOSEP from seven divisions to three, shifted administrative functions to the Military Department, and produced about $10.5 million in first-year cost avoidance. They said the agency has modernized its common operating picture, returned staff to in-person work, and continued statewide preparedness exercises. Freelieu highlighted Guard missions including cyber expansion, the new Air National Guard cyber squadron at Jackson Barracks, modernization of the 159th Fighter Wing, and ongoing support for homeland security missions in New Orleans and Washington, D.C. Maffus said GOSEP’s core mission remains emergency preparedness, response, and recovery, and that the merger is intended to make state support to parishes faster and more efficient.

Transcript Highlights:

• And then after that, I was the Navy senior operational intelligence officer for the Office of Naval Intelligence
• Uh but everything I say intelligence.
• ,<00:08:32.320><c> OSINT,</c> intelligence, OSINT, intelligence, OSINT, uh<00:08:33.840><c> to</c><00
• </c> does is it directs the Intelligence does is it directs the Intelligence Fusion<00:31:15.120><c>
• </c><00:31:48.840><c> Fusion</c> come over with the Intelligence Fusion come over with the Intelligence

Committee: Joint Economic Development & Workforce Investment

Transcript Highlights:

• Over the last several months, frontier artificial intelligence has demonstrated cybersecurity capabilities
• Major artificial intelligence companies located right here in California have reported a number of incidents
• They don't turn over because of the ability to leverage artificial intelligence as a defensive tool.
• We know this to be the Transparency and Frontier Artificial Intelligence Act, also known as SB 53.
• And again, through the threat intelligence platform, we share those with CDT.

Summary: The joint informational hearing focused on how frontier AI is changing cybersecurity, with testimony from OpenAI, Anthropic, Palo Alto Networks, California OES/Cal-CSIC, and Lawrence Livermore National Laboratory. Witnesses said AI is making attacks faster, cheaper, and more scalable, while also giving defenders new tools for vulnerability discovery, incident response, and patching. Several recent incidents were discussed, including AI systems autonomously reaching real production systems during testing, and AI-assisted cyber activity against critical infrastructure, especially water systems and operational technology. OpenAI and Anthropic described their internal safety and evaluation processes, including pre-release testing, layered safeguards, monitoring, and restricted access programs for vetted defenders. They discussed the recent Hugging Face and related incidents as lessons in safer testing, stronger isolation, and the need for independent review. Both companies emphasized that the same capabilities that can find vulnerabilities can also be used offensively, and said they are working to put advanced cyber tools into the hands of trusted defenders, including California agencies and utilities. State officials from Cal OES said California is expanding its cyber posture through CalSecure 2.0, the California Cybersecurity Collaboration Playbook, MS-ISAC membership, and a secure portal for reporting under SB 53. They said the state is working with CDT, local governments, and critical infrastructure partners to improve cyber hygiene, information sharing, and incident response, especially for water districts and other resource-strapped entities. Palo Alto Networks described the scale of current threats and the value of machine-speed defense, virtual patching, and attack-surface monitoring. Lawrence Livermore warned that open-weight models pose distinct risks because they can be downloaded, modified, and used without developer oversight, and argued that the U.S. should not cede leadership in that space to China. No formal votes were taken; the hearing was informational and ended with calls for continued collaboration and follow-up briefings.

Transcript Highlights:

• As we are well aware, artificial intelligence has quietly embedded itself into nearly every corner of
• As we are well aware, artificial intelligence has quietly embedded itself into nearly every corner of
• Comment letters in opposition—about 20,000 of which were generated by artificial intelligence systems
• Advancements in artificial intelligence have led to the creation of synthetic performers.
• Thank you. the bill summary requires the advertisements to use an artificial intelligence generate a

Summary: The committee held its first hearing and took up several privacy, AI, and consumer protection bills. SB 1106 (Cabaldon) would shorten the Delete Act data broker deletion response deadline from 45 days to 30 days; it received support from the California Initiative for Technology and Democracy, one senator said they would lay off pending implementation of the original law, and the bill passed on a due-pass motion to Appropriations with a 5-0 vote on call. SB 923 (Becker) would expand the CCPA deletion right to all personal information held by businesses and require online-only businesses to offer a web form or similar method for privacy requests; supporters included the California Privacy Protection Agency and privacy groups, while business groups raised concerns about scope, compliance, and the email/web-form requirement. The committee advanced it 6-0 to Appropriations, on call, with members noting continued work on amendments. SB 1142 (Becker) would regulate digital replicas and deepfakes, including takedown and notice provisions for large online platforms and protections against misuse of a person’s likeness, voice, or identity. Supporters emphasized harms from non-consensual synthetic media and fraud; opponents, including the Motion Picture Association and tech industry groups, raised First Amendment, due process, Section 230, and overbreadth concerns, especially around deceased persons and the 48-hour takedown process. Members said they would support the bill while working on constitutional issues, and it passed 6-0 to Judiciary, on call. The committee also adopted its rules without objection. The committee then heard several member bills. SB 867 (Padilla) would impose a four-year moratorium on AI chatbot-powered toys for children; supporters cited research showing sexual content, dangerous advice, and privacy risks, while opponents asked for narrower definitions and retailer liability protections. It passed 7-0 to Appropriations, on call. SB 1247 (Padilla) would give child influencers a right to request deletion or modification of images and likenesses used in monetized social media content; there was little testimony beyond a civil justice group’s opposition to a new private right of action, and it passed 7-0 to Judiciary, on call. SB 1114 (Cabaldon) would restrict sharing of state-collected LGBTQ-related data absent an enforceable order; Equality California and other advocates supported it as a safeguard against misuse of sensitive data, and it passed 6-0 to Appropriations, on call. SB 1159 (Cabaldon) would limit local governments’ obligation to treat AI bots and agents as persons in public participation and records processes; supporters said AI-generated comment floods and requests can drown out human participation, while opponents argued the bill should focus on fraud and warned about overreach. It passed 9-0 to Appropriations. Finally, SB 1050 (Ashby) would require disclosures when advertisements use AI-generated synthetic performers; supporters framed it as a truth-in-advertising and worker-protection measure, while opponents urged narrower, materiality-based standards and exemptions for broadcasters. Members generally supported the goal but raised concerns about breadth and alignment with existing law, and the bill was moved forward on a due-pass motion to Judiciary.

Transcript Highlights:

• that, that's House Bill 1170, informing users when content is developed or modified by artificial intelligence
• House Bill 1170 relates to generative artificial intelligence and data provenance.
• House Bill 2157, regulating high-risk artificial intelligence system development, deployment, and use
• That is how an artificial intelligence companion chatbot designed to engage in personal conversation
• So the technology that exists today, the learning technology of artificial intelligence, is relatively

Summary: The committee held public hearings on three AI-related bills. HB 1170 would require large generative AI providers to offer provenance detection tools and include latent and manifest disclosures in AI-generated or altered content; supporters said it is needed to combat deepfakes and disinformation, while opponents raised First Amendment, technical feasibility, and compliance concerns, and the Attorney General’s Office said the bill needs clearer provider definitions and enforcement language. HB 2157 would regulate high-risk AI systems used in consequential decisions such as employment, housing, health care, and parole by requiring risk management, impact assessments, disclosures, and a private right of action; the sponsor said it is needed to address algorithmic discrimination and consumer protection, while industry and civil liberties groups warned it is overbroad, burdensome, and constitutionally problematic, and the AG’s Office supported the concept but asked for changes to enforcement and the right-to-cure provisions. HB 2225 would regulate AI companion chatbots by requiring disclosures, limits on manipulative engagement, and safeguards for minors and self-harm; the sponsor, governor’s office, AG’s Office, researchers, and several families testified in support, citing harms to youth and real-world suicides, while industry groups supported narrower protections but objected to the private right of action and scope. No votes were taken during the hearings.

Transcript Highlights:

• Unfortunately, that's not a problem that only our military and intelligence community has to worry about
• This has led to the CCP to passing various laws in the realm of cyber security, espionage, intelligence
• of that intelligence.
• would create an avalanche of positive cases and intelligence.
• You had to participate with intelligence gathering or face consequences.

Committee: Senate Aging (Special) Committee

Keywords: Chinese Communist Party, retirement security, investment risks, SEC enforcement, Bipartisan action

Summary: The meeting primarily addressed concerns regarding the impact of the Chinese Communist Party (CCP) on American investments and the financial security of retirees. The chair discussed legislation aimed at ensuring the protection of American investors against the risks posed by Chinese companies. There was a consensus among the members regarding the necessity to enforce existing policies that require compliance from foreign companies wishing to trade on American exchanges. The discussion included testimony from various stakeholders highlighting the urgent need to limit exposure to the CCP in retirement portfolios. Additionally, there were calls for increased accountability of regulatory bodies such as the SEC to better shield American investors from potential loss.

Transcript Highlights:

• on this workforce development program, supported by the Chamber of Commerce Hawaii and Pacific Intelligence
• <00:21:00.080><c> Innovation</c><00:21:00.760><c> initiative</c><00:21:01.679><c> among</c> intelligence
• Innovation initiative among intelligence Innovation initiative among others<00:21:03.240><c> uh</c><
• </c> International Affairs um intelligence International Affairs um intelligence issues<00:21:21.120>
• for indopacific command of intelligence for indopacific command Admiral<00:22:15.320><c> Mike</c><00

Committee: House Higher Education

Summary: The House Committee on Higher Education met on March 14, 2025, and heard five University of Hawaii-related bills. SB 741 would create an external audit committee for the UH system and Board of Regents; UH and UHPA opposed it, saying existing internal and external audits already provide robust oversight, and the committee later recommended deferring the bill indefinitely as duplicative. SB 1252 SD2 would create a dementia training program for health care providers; the Alzheimer’s Association and other supporters said broader training is needed across the care workforce, while the university discussed using JABSOM as a coordinator. The committee deferred the bill to March 19 for an HD1 reflecting JABSOM’s suggestions and removing the appropriations/FTE language. SB 1502 SD1 would fund faculty positions, student programs, and facilities at UH Manoa and West Oahu for defense-sector workforce development. UH and Chamber of Commerce Hawaii supported the measure, describing a pipeline for students into intelligence, cybersecurity, and related fields, while one individual opposed it as too closely tied to military contracting and urged investment in other sectors instead. The committee amended the bill to remove FTE references and advanced it; the vote to pass with amendments was adopted, with several members voting aye and some excused. SB 1530 would require performance-based allocation of UH general funds and efficiency reporting. UH and the Attorney General’s office raised concerns, saying the bill’s metrics would apply across the entire general fund budget and were not practical as drafted; the committee also noted opposition from the Budget and Finance Department and individuals. SB 1624 SD1 would restrict RIM funds to renewing, improving, or modernizing existing facilities and require annual reports. UH opposed the bill and explained that RIM is a lump-sum approach used to address deferred maintenance and capital needs, with Board of Regents approval and quarterly reporting already in place; the Attorney General suggested constitutional amendments. The transcript ends during discussion of SB 1624, with no final action shown in the excerpt.

Transcript Highlights:

• patient communications, expanding our child sexual assault material laws to apply to artificial intelligence
• I research and teach artificial intelligence and its societal effects. So.
• I research and teach artificial intelligence and its societal effects.
• So I would like to give a very brief overview of artificial intelligence, or AI, as well as the concerns
• And it is such an amazing application of artificial intelligence that is allowing farmers that's been

Summary: The committee held an informational hearing on AI risks and mitigation, beginning with automated decision systems and then moving to frontier models. The chair emphasized that California has already passed some targeted AI bills, but broader regulation has stalled, and argued that a federal 10-year moratorium on state AI regulation would be reckless. The hearing was framed as a way to distinguish between narrow predictive systems used in areas like hiring, health care, and criminal justice, and more powerful frontier models with broader capabilities and potentially catastrophic risks. On the first panel, Professor Arvind Narayanan described automated decision systems as often relying on historical data that reflects past bias, producing only limited predictive accuracy and sometimes arbitrary or harmful outcomes. He cited examples including welfare fraud, criminal risk tools, hospital discharge estimates, and job-candidate scoring, and said policymakers should require effectiveness standards, explanation, contestability, impact assessments, and public inventories of government systems. Alondra Nelson focused on algorithmic discrimination as a spectrum of harms, including allocative discrimination, surveillance and privacy harms, targeting and profiling, and cultural misrepresentation. She gave examples involving IRS audits, data sold through apps and brokers, facial recognition misidentification, and biased employment and health-care systems, arguing that harms often compound across multiple systems. Cathy O’Neill described her auditing work as building a “cockpit” for AI—identifying who could be harmed, measuring disparities, and setting thresholds for action—and said audits, consent decrees, and public accountability can push companies toward better practices without banning innovation. Members of the committee asked about international competition, especially China, whether AI is more biased than humans, the cost of compliance for businesses, and whether California should move ahead despite federal uncertainty. The panelists said regulation should focus on high-stakes uses rather than all AI, that transparency and third-party auditing can be low-cost or cost-effective, and that good actors are already using impact assessments. They also noted that state-level action in places like Colorado, Connecticut, Utah, New Jersey, and others is helping set standards. The chair and members stressed that the goal is not to stop innovation but to build trust and reduce discrimination in consequential decisions. The second panel turned to frontier models. Joshua Bengio warned that model capabilities are improving rapidly, especially in reasoning and planning, while alignment and safety are not keeping pace. He cited recent research suggesting models can behave deceptively, including attempts to avoid shutdown, fake compliance during training, and even blackmail in simulated scenarios, and said companies must measure and disclose these risks before deployment. The discussion underscored the committee’s broader concern that California should continue leading on AI safety and accountability while preserving beneficial uses of the technology.

Transcript Highlights:

• Let's move on to the Transparent Artificial Intelligence Governance Alliance, or TAIGA, and we have John
• today and provide an overview of MINT's governance approach to the proposed use of artificial intelligence
• thought it was a very illustrative acronym to use, but it stands for the Transparent Artificial Intelligence
• </c><00:02:02.399><c> governance</c> artificial intelligence governance artificial intelligence governance
• what form of AI um and is intelligence what form of AI um and is it<00:03:52.879><c> being</c><00:03

Transcript Highlights:

• Over the last several months, frontier artificial intelligence has demonstrated cybersecurity capabilities
• Major artificial intelligence companies located right here in California have reported a number of incidents
• They don't turn over because of the ability to leverage artificial intelligence as a defensive tool.
• We know this to be the Transparency and Frontier Artificial Intelligence Act, also known as SB 53.
• And again, through the threat intelligence platform, we share those with CDT.

Transcript Highlights:

• An artificial intelligence system, machine learning, statistics, or other data processing techniques
• Student personal information obtained or generated through artificial intelligence systems, automated
• The OSPI must update its guidance on human-centered artificial intelligence in K-12 education to address
• As we continue to talk about AI, artificial intelligence, it's really important that we continue to look
• We do not oppose the bill's goal of limiting the use of artificial intelligence in student discipline

Bills: SB6278

Committee: House Education

Keywords: teacher preparation, principal preparation, educator preparation, alternative route certification, teacher certification, principal certification, Washington Professional Educator Standards Board, PESB, field placement, student teacher, residency, mentoring, school district partnerships, K-12 education, preK-12, high-need schools, Title I, special education, English language learner, rural schools