One of California’s Most Influential Labor Unions Speaks Out on AI Safety Bill

As California Gov. Gavin Newsom considers signing or vetoing the hotly contested AI safety bill, SB 1047, SAG-AFTRA and two women’s groups are pushing him to pass it — adding even more voices to an already heated debate. The artists’ union, the National Organization for Women (NOW), and Fund Her sent letters to Newsome, all of which were obtained by Edge and are published here for the first time.

Letters from SAG-AFTRA, NOWAND Finance it underscore concerns about the potential for AI to cause catastrophic harm if the technology is left unregulated. SAG-AFTRA introduces SB 1047 mandate for developers to test and protect against AI-induced disasters, such as cyberattacks on critical infrastructure or the development of biological weapons. NOW and Fund Her cite sedate warnings With People at the forefront of artificial intelligence developments and discuss the potentially disproportionate impact of this technology on vulnerable groups.

SAG-AFTRA published yesterday from X a request for support 160,000 membersWhich switch on stars like Scarlett Johansson and Tom Hanks. NOW, the biggest a US feminist organisation with about 500,000 members said it was motivated by claims by experts “about how dangerous this amazing technology could be if not developed and implemented responsibly”. Finance it, PAC, which has helped choose 12 Progressive Women in Prominent Offices in California in 2022, writes about the “race to develop the first independently thinking AI,” at which point “it will be too late to put safeguards in place.”

SAG-AFTRA and NOW are the latest political forces to speak out on a California law that has generated unprecedented national attention and disrupted previous partisan divisions.

SB1047by state senator Scott Wiener, would be the most significant AI safety law in the U.S. It establishes civil liability for creators of next-generation AI models, such as ChatGPT, if they cause disasters without implementing adequate safeguards. The bill also includes whistleblower protections for employees of AI companies, ensuring support from OpenAI whistleblowers Daniel Kokotajlo and William Saunders.

NOW writes in its letter that “California’s AI safety standards will change the world,” which is reflected in the bill Co-sponsor And Hendrycksdirector AI Security Center. Hendrycks says Edge that SB 1047 could be Newsome’s “Pat Brown moment,” referring to then-California Gov. signing the landmark 1966 automotive emissions law cites what later became known as The California Effect: “Where California leads the way on important regulations, the rest of the country follows.”

After passing both houses of the state legislature strong majorities in tardy August, the bill now awaits Governor Newsom’s decision, which is expected to be made by September 30. The governor’s office said it “does not normally comment on pending bills. This measure will be evaluated on its merits.”

Despite that comment, the fate of SB 1047 may hinge on political calculations — a reality that both sides appear to recognize as they rally support in the bill’s final moments.

The strange political coalitions that emerged in the fight over SB 1047 portend an upside-down future for AI policy battles. Billionaire Elon Musk backs social justice groups and unions supporting the bill, while former House Speaker Nancy Pelosi, progressive House of Representatives Congressman Ro KhannaTrump-backing venture capitalist Marc Andreessenand the “godmother” of AI Fei Fei Li everyone is against it.

Artificial intelligence is a sporadic problem that There isn’t Already sorted into clear partisan camps. As technology becomes more crucial, the debate over how to govern it is likely to intensify and may continue to confound ordinary loyalties.

These latest letters confirm support for the bill from organizations such as almost 2 million people SEII and Latino Community Foundation.

SAG-AFTRA has been home to some of the most organic anti-AI sentiment. Many film actors See generative AI as an existential threat to their livelihoods. The operate of this technology has been a major a moot point in the actors’ strike in 2023, which led to requirement studios must obtain informed consent from artists before creating digital replicas of them (actors must also be compensated for their operate).

Relationship letter writes that “SAG-AFTRA is acutely aware of the potential dangers that AI poses,” citing issues its members have experienced with unwanted deepfake pornography and theft of performers’ likenesses. It concludes that “policymakers have an obligation to intervene and protect our members and the public. SB 1047 is a measured first step that will get us there.”

Christian Nunes, the organization’s president, said in a phone interview that NOW got involved because the group is concerned about how unregulated AI could affect women. She AND NOW have previously supported efforts to ban unwanted deepfakes.

IN THE NOW letterNunes writes that the dangers that AI experts warn about “will disproportionately affect vulnerable groups, including women.” She highlights Newsome’s “bold support for us in the face of intense lobbying pressure” on reproductive rights, equal pay, and paid family leave, and that this support “is one of the reasons women voted for [him] “time and time again.”

While SB 1047 is not explicitly designed to address the more central concerns of these groups, the organizations appear to see strategic value in joining the coalition behind it. Nunes said Edge believes the bill is part of a broader effort to hold big tech companies accountable.

This support for SB 1047 complements other pending AI legislation that more directly addresses the specific concerns of these groups. For example, federal NO FAKES Act aims to combat deepfakes, while another AI bill sits on Newsome’s desk, supported by SAG-AFTRA, would regulate use of digital replicas. By supporting SB 1047, along with these more targeted initiatives, these organizations appear to be taking a comprehensive approach to AI governance.

NOW and Fund Her’s letters draw parallels between unregulated AI and the history of social media. Fund Her founder and CEO Valerie McGinty writes to Edge, “We have seen the incredible harm that social media has done to our children and how difficult it is to reverse. We will not have to play catch-up again if Gov. Newsom signs SB 1047 into law.”

It’s unclear whether the letters will be enough to overcome the powerful forces against the bill. While Wiener and others advocates Describing the regulations as “soft” and “common sense,” the industry is largely panic.

The United States currently relies almost exclusively on self-regulation and non-binding voluntary commitments to manage artificial intelligence and industry I’d like to keep it that way. As the first U.S. AI security regulation with a claw, SB 1047 would set a mighty precedent, which is likely the motivation behind these letters and the mighty industry pushback.

Google, FinishAND OpenAI he took extremely step of writing their own letters opposing the bill. The resistance of AI investors was even stronger, and the prestigious startup incubator Combiner Y (YC) and venture capital Andreessen Horowitz (a16z) leading full-court pressure to block SB 1047. These and other prominent opponents warn the bill could hint exodus from California, cede The United States is ahead of China in artificial intelligence, destroy open source community.

Naturally, supporters dispute each of these arguments. In the July letter Referring to YC and a16z’s claims about the bill, Wiener points out that SB 1047 would apply to any AI company covered by it that does business in California, the world’s AI hub and fifth-largest economy. Dario Amodei, CEO of leading AI company and eventual de facto supporter of SB 1047 Anthropic, called the threat of leaving “only the theatre” (however, it was also invoked by OpenAI, FinishAND Google).

In it statement in opposing the bill, Pelosi called it “well-intentioned but misinformed.” In a phone interview, Wiener said, “I have tremendous respect for the retired speaker. She’s the best,” but called Pelosi’s statement “unfortunate” and noted that “some of the best machine learning pioneers in the world support the bill,” citing support from the “godfathers” of deep learning Geoffrey Hinton AND Joshua BengioWiener also emphasizes support open letter published on Monday from over 100 employees and graduates of leading artificial intelligence companies.

To assess the validity of SB 1047, the most persuasive letter may be one published by Anthropic, which distanced itself from its competitors and wrote that “the benefits of the revised legislation likely outweigh the costs.” The letter was sent after round of corrections made directly in response to the company previous complaintsAnthropic’s Claude family of chatbots is a world leader some indicatorsand the company will likely be one of the few AI companies that will be directly affected by the law in the near future.

With key leaders of the Congress promising to hinder crucial federal regulations regarding artificial intelligence and opposing SB 1047California can go it alone, as it has done in the past. Net neutrality AND data privacyAs NOW’s Nunes writes, “California’s AI safety standards will change the world,” giving Governor Newsom a chance to make history and model “sustainable AI leadership.”

Finance it McGinty summed up the proponents’ position in an email to Edge: “We should listen to those experts who are more concerned with our well-being than Big Tech executives skimping on AI security.”

As the Sept. 30 deadline approaches, all eyes are on Gov. Newsom to see how he shapes the future of AI governance in California and beyond. “My experience with Gavin Newsom is—whether you agree or disagree—he makes thoughtful decisions based on what he thinks is best for the state,” Wiener says. “I’ve always appreciated that about him.”

Adjustment: The article originally quoted the “godfather” of deep learning Yann LeCun as a supporter of SB 1047. LeCun opposes the bill. We regret the error.