California lawmaker revives AI whistleblower protections concerns in new bill
Sen. Scott Wiener (D-San Francisco) has introduced Senate Bill 53, which would protect AI lab whistleblowers from retaliation if they speak out about risks or irresponsible development. The bill would also create a compute cluster—CalCompute—to ensure a broad array of developers can access the compute they need to succeed.
Specifically, the bill would prohibit developers of certain artificial intelligence models from enforcing policies that prevent employees from disclosing information or retaliating against those who report concerns.
California bill would require insurers to pay full coverage without itemized lists
SB 53 defines 'critical risk' as a foreseeable and material risk that the development, storage or deployment of a foundation model could result in the death or serious injury of more than 100 people or cause more than $1 billion in damages.
Additionally, developers would be required to establish an internal process allowing employees to report concerns anonymously.
'We are still early in the legislative process, and this bill may evolve as the process continues. I'm closely monitoring the work of the Governor's AI Working Group, as well as developments in the AI field for changes that warrant a legislative response,' Wiener said in a statement.
'California's leadership on AI is more critical than ever as the new federal Administration proceeds with shredding the guardrails meant to keep Americans safe from the known and foreseeable risks that advanced AI systems present.'
Last year, when Wiener introduced a similar bill, other lawmakers and tech companies opposed the proposed legislation, saying it was 'ill-informed, will cause more harm than good, and will eventually drive developers away from the state.'
Copyright 2025 Nexstar Media, Inc. All rights reserved. This material may not be published, broadcast, rewritten, or redistributed.
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
Yahoo
18 minutes ago
- Yahoo
Ideal Systems Debuts "Tabatha" AI-Powered Newsroom Service at IBC2025
New service includes native integration of Comcast Technology Solutions' VideoAI™ and will be demonstrated with CTS at IBC2025 in Hall #5, in stand 5.C80 SINGAPORE, SG / / August 19, 2025 / Ideal Systems announced today it will be previewing "Tabatha," an integrated newsroom AI platform with native integration of Comcast Technology Solutions' (CTS) VideoAI™ at IBC2025 in Amsterdam. Tabatha is optimised and tailored for using AI in modern TV newsroom productions and is the latest module for Ideal Systems "Alice" cloud integration platform. Tabatha's native integration with VideoAI from CTS aims to dramatically increase newsroom productivity by reducing content search times and increasing search accuracy. It will enable users to search for any highlight without requiring content specific training and significantly simplify highlights workflows across all news and sports genres. Born out of the need for deep integration with both legacy broadcast systems and new media content management systems, Tabatha provides an enterprise integration layer for launching new AI assisted workflows to assist with the efficient creation of news content in legacy newsroom environments with minimal integration impact. Tabatha also allows for the management and control of multiple AIs within the same workflows, even if they are from different AI providers, with different billing models and full API level audits of all actions. "For news organizations, Tabatha serves as a seamless extension layer that enhances existing newsroom systems, archives, and editing workflows with advanced AI-driven media analysis, without altering the core properties of legacy infrastructure, minimizing the risk of introducing AI into operations," said Fintan Mc Kiernan, CEO at Ideal Systems SEA. "We're very proud to have VideoAI from Comcast Technology Solutions powering our efforts and to showcase our work together at IBC2025. Tabatha can also serve as a common platform enabling organizations to leverage AI insights across multiple detectors and billing models, all with full API-level auditing." While Tabatha can integrate with many AIs depending on customer requirements, it will launch with a native integration with VideoAI from CTS. Like Tabatha, VideoAI comes from a proven broadcast background having been developed and used by some of the largest broadcast, media and entertainment companies in the world. Based on years of real-word data training and refinements, VideoAI has won numerous accolades including a Technology & Engineering Emmy® Award, NAB Product of the Year Award, and numerous others. Using VideoAI, Tabatha allows for advanced functionality such as curation of content with accurate topic segmentation and a unique per segment synopsis. Deep search of video content for visual objects, celebrities, and semantic queries are all supported. Tabatha can generate advertising tags and standardised asset metadata sets. It can also automatically identify and select key highlights for distribution to social media platforms and apps or export them via Edit Decision Lists (EDLs) for use in editing platforms like Adobe Premiere, GV EDIUS, and others. Ideal Systems, powered by CTS VIdeoAI, is delivering a future-ready solution that modernizes newsroom operations through advanced automation, AI, and deep media intelligence. Ideal Systems will be demonstrating their integrated solution with CTS at IBC Hall #5, in stand 5.C80 and are now open to taking bookings. Contact AI@ for more information. Contact Information Melody Bilegt Marketing 6684-8770 SOURCE: Ideal Systems (Singapore) Pte. Ltd. View the original press release on ACCESS Newswire Error in retrieving data Sign in to access your portfolio Error in retrieving data Error in retrieving data Error in retrieving data Error in retrieving data
Yahoo
18 minutes ago
- Yahoo
Ramsey Theory Group Revolutionizes the Automotive Retail Sector with a Secure and Digital-First Sales Platform Proven to Streamline the Sales Process
Ramsey Theory Group is improving the automotive sales experience in a tougher market by giving car dealerships a secure, fully digital AI-driven platform to streamline transactions, ensure compliance, and deliver the convenience that car buyers demand. NEW YORK, Aug. 19, 2025 /PRNewswire/ -- Ramsey Theory Group, a leading provider of software development, quantitative analysis, information technology, cybersecurity, digital marketing, and product development, announced today that its AI-driven platform for the automotive retail sector is showing strong benefits for dealerships and customers alike. The seamless, compliant, and secure digital solution now is being used by car dealerships nationwide and delivers the following: Simplified, End-to-End Sales Workflow - streamlines the dealership sales process by enabling document uploads, credit submissions, and creation of compliant deal documents - all within a digital, AI-driven ecosystem designed for efficiency and clarity. This approach reduces paperwork, administrative burdens, and errors, significantly cutting time and costs for automotive retailers. Enhanced Compliance and Security - Built with regulatory and fraud-resistant measures in mind, the platform uses AI to ensure that all customer information and deal documentation adhere to compliance standards. This mitigates risk for dealerships and instills confidence in buyers throughout the transaction process. Improved Customer Engagement and Convenience - the app reimagines car buying as a fully virtual, user-friendly experience. Customers can remotely upload identification documents, submit credit application info, and receive compliant deal documentation from the comfort of their own homes. This consumer-centric approach caters to modern buyers' expectations for convenience and transparency. Thoughtfully Designed for Automotive Retailers – the solution was developed following extensive research and conversations with dealership staff and customers. Addressing real-world pain points, the platform delivers a user experience that saves time and enhances satisfaction for both parties in the automotive sales process. "Unlike other solutions for the automotive retail sector, we spent a great deal of time researching and evaluating the pain points of both the car buyers and also the dealerships," said Dan Herbatschek, CEO & Founder of Ramsey Theory Group. "This leapfrog technology brings speed, security, and simplicity to the automotive sales process and we're excited to continue innovating with new feature rollouts for the platform in the near future based on our customers' feedback." Visit to learn more. About Ramsey Theory GroupFounded by entrepreneur and technology innovator, Dan Herbatschek, Ramsey Theory Group leverages its expertise in software development, quantitative analysis, information technology, cybersecurity, digital marketing, and product development to better help organizations optimize their workflow. Working alongside entrepreneurs, the firm bridges the gap between business and software engineering matters - translating the vision of organizations into technologically executable problems. Based in New York, the Ramsey Theory Group specializes in Data-Intensive Application Design, Data Engineering, Business Intelligence, Custom Optimization, Mathematical & Statistical Modelling, Software Development, Data Visualization, Blockchain Development, Blockchain Consultancy, and Web and Mobile Application Development. View original content to download multimedia: SOURCE Ramsey Theory Group Sign in to access your portfolio
E&E News
28 minutes ago
- E&E News
Trump's AI plan meets a stressed American electric grid
President Donald Trump's plan to make the U.S. a leader in artificial intelligence is also set to remake the country's electric grid in his vision: Bigger, faster and bolstered by fossil fuels. But it's unclear how parts of the plan would be implemented and if it would be enough to ensure the U.S. has ample power to meet demand from energy-hungry data centers over the next decade. In a speech Wednesday, Trump declared that his administration would pursue 'all-out American energy dominance' to promote the AI industry. He promised the country would build as much new electricity capacity as China this year. Advertisement To do so, Trump signed an executive order Wednesday to streamline permitting for new data centers and associated power generation, along with two others related to artificial intelligence model development. The administration also released an action plan Wednesday morning that included ambitions for a 'comprehensive strategy to enhance and expand the power grid.' 'There's a global competition to lead in artificial intelligence and we want the United States to win that race,' David Sacks, the White House AI and crypto czar, said on a call with reporters. The action plan's provisions were designed to be implemented in six months to a year, said White House Office of Science and Technology Policy Director Michael Kratsios. The White House called for streamlining permitting requirements for new data centers and construction of data centers and power generation on federal land. The plan dictates the prioritization of 'reliable dispatchable power sources' like coal and gas plants, keeping existing generation on the grid and even reforming power markets 'to align financial incentives with the goal of grid stability.' Many of the provisions — including support for AI workforce programs — were backed by the Data Center Coalition, which represents the industry. The plan's 'focus on the removal of barriers to faster data center construction and operation — ranging from power access to supply chains to workforce development — which will allow data center companies to continue investing hundreds of billions of dollars to expand U.S. data center infrastructure across the country,' Josh Levi, the group's president, said in a statement. The group called for Congress to take further action on issues such as permitting. The plan was praised by groups representing industries supported by Trump and companies like telecommunications giant Lumen Technologies, which said cutting regulatory barriers would help build an AI backbone. While the plan mentions a need for 'new energy generation sources at the technological frontier' like geothermal and advanced nuclear, it is silent on wind and solar power. Renewable energy is typically the quickest form of new generation to build and has been a backbone for much of the new data center construction. The U.S. Energy Information Administration has reported that wind, solar and batteries accounted for more than 80 percent of new generation in 2024 and forecast that renewables would be 81 percent of new capacity in 2025. 'Companies are investing in renewables not because they love renewable energy, but because it provides the best return,' said Safak Yucel, associate director of the Business of Sustainability Initiative at Georgetown University. 'And the reason is not subsidies, it's that they're cheaper to run.' Large technology companies are continuing to find any energy source — including renewables — available on the grid. Meta, for example, announced a contract this week to purchase 600 megawatts of solar power from a Texas project developed by Canadian energy giant Enbridge to power its data centers. On Tuesday, OpenAI and Oracle announced next steps for the 'Stargate' project, a company with multiple partners backed by Trump that pledged in January to spend $500 billion in four years to build AI infrastructure in the United States. The companies inked a deal to develop 4.5 gigawatts of additional data center capacity, an amount that is more than double the power used in San Francisco. The companies did not specify where the new facilities would be or which power sources they would use but said parts of an existing data project in Abilene, Texas, tied to Stargate is now 'up and running.' In a research note Wednesday, Bank of America said it expected data centers to help U.S. electricity demand increase 2.5 percent annually over the next decade, almost five times more than the rate from 2014 to 2024. The action plan also came a day after United Nations Secretary-General António Guterres called on major tech firms to power their data centers entirely with renewable energy by 2030, warning that the forecast electricity demand is 'not sustainable — unless we make it so.' 'The future is being built in the cloud,' Guterres said in a speech Tuesday in New York. 'It must be powered by the sun, the wind, and the promise of a better world.' Outside view of the newly completed Meta's Facebook data center in Eagle Mountain, Utah on July 18, 2024. The data center is a complex of five large buildings, each over four football fields long and totaling 2.4 million square feet. | George Frey/AFP via Getty Images Abundance or scarcity? Critics charge that the administration's bid to expand the grid for AI is at odds with the policies coming out of the White House. The budget and tax megalaw signed by Trump earlier this month, for example, slashed tax credits and incentives for wind and solar projects. A forecast from BloombergNEF predicts that new wind, solar and energy storage additions could drop 23 percent through 2030 compared to what would have happened without the law. In all, that would be about 117 fewer gigawatts of new power. The Interior Department has also added new requirements for construction of solar and wind on federal lands, which developers have called a form of sabotage. The AI plan's release also came on the same day that Trump's Department of Energy nixed a conditional federal loan for the Grain Belt Express transmission project. That line, set to be one of the nation's largest new grid projects, would carry renewable energy from the Great Plains to large cities in the East, but had opposition from key Republican lawmakers. 'Their rhetoric doesn't match reality,' said Harry Godfrey, leader of Advanced Energy United's Federal Priorities team. 'They talk a game about energy abundance, yet their actions are fundamentally counter-abundance. They're creating scarcity.' In his speech, Trump talked up the potential of nuclear power to back data centers, calling the technology 'incredible.' But the technology — like advanced geothermal — is years away from commercial viability. Even building a new gas plant is expected to take at least five years because of supply chain backlogs. 'There is no AI revolution without power,' Godfrey said. 'But under this administration, where are you going to get the power?' The AI Action Plan mentions grid management technologies and tools to make existing power lines more efficient. It also calls for 'new and novel ways' for data centers to manage demand during periods of peak demand, a strategy researchers have said could allow more construction without additional generation. Rob Gramlich, president of the research firm Grid Strategies, said those were 'encouraging proposals' and could continue work on grid enhancements that started under the Biden administration. But, he added, other parts of the plan that tilt the scales toward fossil fuels, especially in interconnection queues, could result in 'market distortion' that slows down new power growth. President Donald Trump talked up the potential of nuclear power to back data centers, calling the technology 'incredible.' Cooling towers four, left and three are seen at the Alvin W. Vogtle Electric Generating Plant in Waynesboro, Georgia. | Mike Stewart/AP Streamlining NEPA A chief way the Trump administration is aiming to expedite a build out of data centers is through regulatory changes, including environmental review under the National Environmental Policy Act. The action plan calls for the creation of a categorical exclusion under NEPA 'to cover data center-related actions that normally do not have a significant effect on the environment.' That could free them from conducting environmental impact statements and assessments, if fully implemented. 'Where possible, adopt Categorical Exclusions already established by other agencies so that each relevant agency can proceed with maximum efficiency,' the plan states. That echoes the calls of many in the tech industry, who have complained that it takes too long to build infrastructure. Environmentalists say the language could lead to faster approval of large facilities that worsen air and water pollution. 'Across the South we are seeing states and communities clamor for more information about the infrastructure demands for AI, not less. Transparency is crucial to ensure that the physical infrastructure needed to support AI doesn't threaten our health, energy grid, access to water, and dramatically increase costs,' said Southern Environmental Law Center Climate Initiative Leader Alys Campaigne. The plan, if implemented, could make it harder for communities to obtain important information about nearby data centers, the group said. James Coleman, a law professor at the University of Minnesota, said the language could expedite projects, but it's not a 'silver bullet that gets you out of NEPA review.' Environmentalists could still challenge projects in court if they believe a data center facility did not have adequate environmental review. A court 'is going to make its own interpretation,' no matter what agencies do, he said. At the same time, the Trump administration may be in a better position to win many lawsuits, after the Supreme Court ruled in July that NEPA does not require agencies to consider far-reaching environmental impacts of projects, according to Coleman. Another lingering question is how agencies would look to implement categorical exclusions, since the administration has moved to roll back regulations governing the NEPA process. The Department of Energy issued an interim final rule this year rescinding its NEPA regulations and releasing alternative guidance in its place. 'What does a categorical exclusion mean if you aren't allowed to adopt regulations?' Coleman asked. Tanya Das, the director of artificial intelligence and energy technology policy at the Bipartisan Policy Center, said her organization supported the categorical exclusions to expedite the NEPA process. Language in the plan aiming to 'enable AI adoption' also could be useful, considering the time it takes to process applications under NEPA and other regulations. Going through data 'at a human pace can be very slow,' she said. AI tools that sift through data could be an option for expediting reviews, she said. Das said she would be watching closely to see how the administration fleshes out language in the action plan on the grid. The White House included broad directives such as 'stabilize the grid of today as much possible' and 'enhance the efficiency and performance of the transmission system,' but did not specify matching agency actions as of yet to implement many of those goals. Some groups that often back deregulatory actions said they were concerned that the plan puts too much focus on tapping the federal government to finance and guide the technology. The plan, for instance, calls for agencies 'with significant land portfolios' to identify sites suited for data center development. 'We don't want the Energy Department to overshadow private rollout and development of the infrastructure,' said Wayne Crews, a fellow in regulatory studies at the Competitive Enterprise Institute, a libertarian think tank.
