Safe and Secure Innovation for Frontier Artificial Intelligence Models Act
Safe and Secure Innovation for Frontier Artificial Intelligence Models Act | |
---|---|
![]() | |
California State Legislature | |
Full name | Safe and Secure Innovation for Frontier Artificial Intelligence Models Act |
Introduced | February 7, 2024 |
Assembly voted | August 28, 2024 (48–16) |
Senate voted | August 29, 2024 (29–9) |
Sponsor(s) | Scott Wiener |
Governor | Gavin Newsom |
Bill | SB 1047 |
Website | Bill Text |
The Safe and Secure Innovation for Frontier Artificial Intelligence Models Act, or SB 1047, is a 2024 California bill intended to "mitigate the risk of catastrophic harms from AI models so advanced that they are not yet known to exist".[1] Specifically, the bill would apply to models which cost more than $100 million to train and were trained using a quantity of computing power greater than 1026 integer or floating-point operations.[2] SB 1047 would apply to all AI companies doing business in California—the location of the company does not matter.[3] The bill creates protections for whistleblowers[4] and requires developers to perform risk assessments of their models prior to release, under the supervision of the Government Operations Agency. It would also establish CalCompute, a University of California public cloud computing cluster for startups, researchers and community groups.
History
[edit]Background
[edit]The rapid increase in capabilities of AI systems in the 2020s, including the release of ChatGPT in November 2022, caused some researchers and members of the public to express concern about existential risks associated with increasingly powerful AI systems.[5][6] In May 2023, a statement on AI risk of extinction was released, which read "Mitigating the risk of extinction from AI should be a global priority alongside other societal-scale risks such as pandemics and nuclear war." It received signatures from the two most-cited AI researchers,[7][8][9] Geoffrey Hinton and Yoshua Bengio, along with industry figures such as OpenAI CEO Sam Altman, Google DeepMind CEO Demis Hassabis, and Anthropic CEO Dario Amodei.[10][11] Some thought existential concerns were overblown, and a distraction from the near-term harms of AI.[12] The concerns led to calls for stronger regulation of artificial intelligence, leading Sam Altman to ask politicians to take action during a congressional hearing.[13] Several technology companies have made voluntary commitments to conduct safety testing, for example at the AI Safety Summit and AI Seoul Summit.[14][15]
Governor Newsom of California and President Biden issued executive orders on artificial intelligence in 2023.[16][17][18] State Senator Wiener said SB 1047 draws heavily on the Biden executive order, and is motivated by the absence of unified federal legislation on AI safety.[19]
Legislative history
[edit]State Senator Wiener first proposed AI legislation for California through an intent bill called SB 294, the Safety in Artificial Intelligence Act, in September 2023.[20][21][22] SB 1047 was introduced by Wiener on February 7, 2024.[23][24]
On May 21, SB 1047 passed the Senate 32-1.[25][26] The bill was significantly amended by Wiener on August 15, 2024 in response to industry advice.[27] Amendments included adding clarifications, and removing the creation of a "Frontier Model Division" and the penalty of perjury.[28][29]
On August 28, the bill passed the State Assembly 48-16. Then, due to the amendments, the bill was once again voted on by the Senate, passing 29-9.[30][31]
Provisions
[edit]Prior to model training, developers of covered models and derivatives are required to submit a certification, subject to auditing, of mitigation of "reasonable" risk of "critical harms" of the covered model and its derivatives, including post-training modifications. Critical harms are defined with respect to four categories:[1][32]
- Creation or use of a chemical, biological, radiological, or nuclear weapon[33]
- Cyberattacks on critical infrastructure causing mass casualties or at least $500 million of damage
- Autonomous crimes causing mass casualties or at least $500 million of damage
- Other harms of comparable severity
Developers of covered models are also required to implement "reasonable" safeguards to reduce risk,[4] including the ability to shut down the model,[4] which has been variously described as a "kill switch"[34] and "circuit breaker".[35] Whistleblowing provisions protect employees who report safety problems and incidents.[4]
SB 1047 would also create a public cloud computing cluster called CalCompute, associated with the University of California, to support startups, researchers, and community groups that lack large-scale computing resources.[28]
Coverage
[edit]SB 1047 covers AI models with training compute over 1026 integer or floating-point operations and a cost of over $100 million.[2][36] If a covered model is fine-tuned using more than $10 million, the resulting model is also covered.[29]
Compliance and supervision
[edit]SB 1047 would require developers, beginning January 1, 2028, to annually retain a third-party auditor to perform an independent audit of compliance with the requirements of the bill, as provided.[28] The Government Operations Agency would review the results of safety tests and incidents, and issue guidance, standards, and best practices.[28] The bill creates a Board of Frontier Models to supervise the application of the bill by the Government Operations Agency. It is composed of 9 members.[36][needs update]
Reception
[edit]Support and opposition
[edit]Supporters of the bill include Turing Award recipients Yoshua Bengio[37] and Geoffrey Hinton,[38] Elon Musk[39], Kevin Esvelt,[40] Dan Hendrycks[41], Vitalik Buterin[42], OpenAI whistleblowers Daniel Kokotajlo[33] and William Saunders,[43] Lawrence Lessig,[44] Sneha Revanur,[45] Stuart Russell[44] and Max Tegmark.[46] The Center for AI Safety, Economic Security California[47] and Encode Justice[48] are sponsors. Yoshua Bengio writes that the bill is a major step towards testing and safety measures for "AI systems beyond a certain level of capability [that] can pose meaningful risks to democracies and public safety."[49] Max Tegmark likened the bill's focus on holding companies responsible for the harms caused by their models to the FDA requiring clinical trials before a company can release a drug to the market. He also argued that the opposition to the bill from some companies is "straight out of Big Tech's playbook."[46] The Los Angeles Times editorial board has also written in support of the bill.[50]
Andrew Ng, Fei-Fei Li,[51] Ion Stoica, Jeremy Howard, Turing Award recipient Yann LeCun, along with U.S. Congressmembers Nancy Pelosi, Zoe Lofgren, Anna Eshoo, Ro Khanna, Scott Peters, Tony Cárdenas, Ami Bera, Nanette Barragán and Lou Correa have come out against the legislation.[6][52][53] Andrew Ng argues specifically that there are better more targeted regulatory approaches, such as targeting deepfake pornography, watermarking generated materials, and investing in red teaming and other security measures.[49] University of California and Caltech researchers have also written open letters in opposition.[52]
Industry
[edit]The bill is opposed by industry trade associations including the California Chamber of Commerce, the Chamber of Progress,[a] the Computer & Communications Industry Association[b] and TechNet.[c][2] Companies including Meta[57] and OpenAI[58] are opposed to or have raised concerns about the bill, while Google,[57] Microsoft and Anthropic[46] have proposed substantial amendments.[3] Several startup founder and venture capital organizations are opposed to the bill, for example, Y Combinator,[59][60] Andreessen Horowitz,[61][62][63] Context Fund[64][65] and Alliance for the Future.[66]
After the bill was amended, Anthropic CEO Dario Amodei wrote that "the new SB 1047 is substantially improved, to the point where we believe its benefits likely outweigh its costs. However, we are not certain of this, and there are still some aspects of the bill which seem concerning or ambiguous to us."[67] Amodei also commented, "There were some companies talking about moving operations out of California. In fact, the bill applies to doing business in California or deploying models in California... Anything about 'Oh, we're moving our headquarters out of California...' That's just theater. That's just negotiating leverage. It bears no relationship to the actual content of the bill."[68] xAI CEO Elon Musk wrote, "I think California should probably pass the SB 1047 AI safety bill. For over 20 years, I have been an advocate for AI regulation, just as we regulate any product/technology that is a potential risk to the public."[69]
Open source developers
[edit]Critics expressed concerns about liability on open source software imposed by the bill if they use or improve existing freely available models. Yann LeCun, Chief AI Officer of Meta, has suggested the bill would kill open source AI models.[49] Currently (as of July 2024), there are concerns in the open-source community that due to the threat of legal liability companies like Meta may choose not to make models (for example, Llama) freely available.[70][71] The AI Alliance has written in opposition to the bill, among other open-source organizations.[52] In contrast, Creative Commons co-founder Lawrence Lessig has written that SB 1047 would make open source AI models safer and more popular with developers, since both harm and liability for that harm are less likely.[35]
Public opinion polls
[edit]The Artificial Intelligence Policy Institute, a pro-regulation AI think tank,[72] ran three polls of California respondents on whether they supported or opposed SB 1047.
Support | Oppose | Not sure | Margin of error | |
---|---|---|---|---|
July 9, 2024[73][74] | 59% | 20% | 22% | ±5.2% |
August 4–5, 2024[75][76] | 65% | 25% | 10% | ±4.9% |
August 25–26, 2024[77][78] | 70% | 13% | 17% | ±4.2% |
A David Binder Research poll commissioned by the Center for AI Safety, a group focused on mitigating societal-scale risk, found that 77% of Californians support a proposal to require companies to test AI models for safety risks, and 86% consider it an important priority for California to develop AI safety regulations.[79][80][81]
See also
[edit]- Artificial general intelligence
- Regulation of AI in the United States
- Regulation of artificial intelligence
Notes
[edit]References
[edit]- ^ a b Bauer-Kahan, Rebecca. "ASSEMBLY COMMITTEE ON PRIVACY AND CONSUMER PROTECTION" (PDF). California Assembly. State of California. Retrieved 1 August 2024.
- ^ a b c Daniels, Owen J. (2024-06-17). "California AI bill becomes a lightning rod—for safety advocates and developers alike". Bulletin of the Atomic Scientists.
- ^ a b Rana, Preetika (2024-08-07). "AI Companies Fight to Stop California Safety Rules". The Wall Street Journal. Retrieved 2024-08-08.
- ^ a b c d Thibodeau, Patrick (2024-06-06). "Catastrophic AI risks highlight need for whistleblower laws". TechTarget. Retrieved 2024-08-06.
- ^ Henshall, Will (2023-09-07). "Yoshua Bengio". TIME.
- ^ a b Goldman, Sharon. "It's AI's "Sharks vs. Jets"—welcome to the fight over California's AI safety bill". Fortune. Retrieved 2024-07-29.
- ^ Hinton, Geoffrey (2024-04-17). "Yoshua Bengio". TIME. Retrieved 2024-09-03.
- ^ "World Scientists Citation Rankings 2024". AD Scientific Index. Retrieved 2024-09-03.
- ^ "Profiles". Google Scholar. Retrieved 2024-09-02.
- ^ Abdul, Geneva (2023-05-30). "Risk of extinction by AI should be global priority, say experts". The Guardian. ISSN 0261-3077. Retrieved 2024-08-30.
- ^ "Statement on AI Risk | CAIS". www.safe.ai. Retrieved 2024-08-30.
- ^ "Artificial intelligence could lead to extinction, experts warn". BBC News. 2023-05-30. Retrieved 2024-08-30.
- ^ Zorthian, Julia (2023-05-16). "OpenAI CEO Sam Altman Agrees AI Must Be Regulated". TIME. Retrieved 2024-08-30.
- ^ Milmo, Dan (2023-11-03). "Tech firms to allow vetting of AI tools, as Musk warns all human jobs threatened". The Guardian. Retrieved 2024-08-12.
- ^ Browne, Ryan (2024-05-21). "Tech giants pledge AI safety commitments — including a 'kill switch' if they can't mitigate risks". CNBC. Retrieved 2024-08-12.
- ^ "Governor Newsom Signs Executive Order to Prepare California for the Progress of Artificial Intelligence". Governor Gavin Newsom. 2023-09-06.
- ^ "President Biden Issues Executive Order on Safe, Secure, and Trustworthy Artificial Intelligence". White House. 2023-10-30.
- ^ Riquelmy, Alan (2024-02-08). "California lawmaker aims to put up guardrails for AI development". Courthouse News Service. Retrieved 2024-08-04.
- ^ Myrow, Rachael (2024-02-16). "California Lawmakers Take On AI Regulation With a Host of Bills". KQED.
- ^ Perrigo, Billy (2023-09-13). "California Bill Proposes Regulating AI at State Level". TIME. Retrieved 2024-08-12.
- ^ David, Emilia (2023-09-14). "California lawmaker proposes regulation of AI models". The Verge. Retrieved 2024-08-12.
- ^ "Senator Wiener Introduces Safety Framework in Artificial Intelligence Legislation". Senator Scott Wiener. 2023-09-13. Retrieved 2024-08-12.
- ^ "California's SB-1047: Understanding the Safe and Secure Innovation for Frontier Artificial Intelligence Act". DLA Piper. Retrieved 2024-08-30.
- ^ Edwards, Benj (2024-07-29). "From sci-fi to state law: California's plan to prevent AI catastrophe". Ars Technica. Retrieved 2024-08-30.
- ^ Hendrycks, Dan (2024-08-27). "California's Draft AI Law Would Protect More than Just People". TIME. Retrieved 2024-08-30.
- ^ "California SB1047 | 2023-2024 | Regular Session". LegiScan. Retrieved 2024-08-30.
- ^ Zeff, Maxwell (2024-08-15). "California weakens bill to prevent AI disasters before final vote, taking advice from Anthropic". TechCrunch. Retrieved 2024-08-23.
- ^ a b c d Calvin, Nathan (August 15, 2024). "SB 1047 August 15 Author Amendments Overview". safesecureai.org. Retrieved 2024-08-16.
- ^ a b "Senator Wiener's Groundbreaking Artificial Intelligence Bill Advances To The Assembly Floor With Amendments Responding To Industry Engagement". Senator Scott Wiener. 2024-08-16. Retrieved 2024-08-17.
- ^ writer, Troy Wolverton | Examiner staff (2024-08-29). "Legislature passes Wiener's controversial AI safety bill". San Francisco Examiner. Retrieved 2024-08-30.
- ^ "Bill History - SB-1047 Safe and Secure Innovation for Frontier Artificial Intelligence Models Act". leginfo.legislature.ca.gov. Retrieved 2024-08-30.
- ^ "Analysis of the 7/3 Revision of SB 1047". Context Fund.
- ^ a b Johnson, Khari (2024-08-12). "Why Silicon Valley is trying so hard to kill this AI bill in California". CalMatters. Retrieved 2024-08-12.
- ^ Times, Financial (2024-06-07). "Outcry from big AI firms over California AI "kill switch" bill". Ars Technica. Retrieved 2024-08-30.
- ^ a b Lessig, Lawrence (2024-08-30). "Big Tech Is Very Afraid of a Very Modest AI Safety Bill". The Nation. Retrieved 2024-09-03.
- ^ a b "07/01/24 - Assembly Judiciary Bill Analysis". California Legislative Information.
- ^ Bengio, Yoshua. "Yoshua Bengio: California's AI safety bill will protect consumers and innovation". Fortune. Retrieved 2024-08-17.
- ^ Kokalitcheva, Kia (2024-06-26). "California's AI safety squeeze". Axios.
- ^ Coldewey, Devin (2024-08-26). "Elon Musk unexpectedly offers support for California's AI bill". TechCrunch. Retrieved 2024-08-27.
- ^ Riquelmy, Alan (2024-08-14). "California AI regulation bill heads to must-pass hearing". Courthouse News Service. Retrieved 2024-08-15.
- ^ Metz, Cade (2024-08-14). "California A.I. Bill Causes Alarm in Silicon Valley". New York Times. Retrieved 2024-08-22.
- ^ Jamal, Nynu V. (2024-08-27). "California's Bold AI Safety Bill: Buterin, Musk Endorse, OpenAI Wary". Coin Edition. Retrieved 2024-08-27.
- ^ Zeff, Maxwell (2024-08-23). "'Disappointed but not surprised': Former employees speak on OpenAI's opposition to SB 1047". TechCrunch. Retrieved 2024-08-23.
- ^ a b Pillay, Tharin (2024-08-07). "Renowned Experts Pen Support for California's Landmark AI Safety Bill". TIME. Retrieved 2024-08-08.
- ^ "Assembly Standing Committee on Privacy and Consumer Protection". CalMatters. Retrieved 2024-08-08.
- ^ a b c Samuel, Sigal (2024-08-05). "It's practically impossible to run a big AI company ethically". Vox. Retrieved 2024-08-06.
- ^ DiFeliciantonio, Chase (2024-06-28). "AI companies asked for regulation. Now that it's coming, some are furious". San Francisco Chronicle.
- ^ Korte, Lara (2024-02-12). "A brewing battle over AI". Politico.
- ^ a b c Edwards, Benj (2024-07-29). "From sci-fi to state law: California's plan to prevent AI catastrophe". Ars Technica. Retrieved 2024-07-30.
- ^ The Times Editorial Board (2024-08-22). "Editorial: Why California should lead on AI regulation". Los Angeles Times. Retrieved 2024-08-23.
- ^ Li, Fei-Fei. "'The Godmother of AI' says California's well-intended AI bill will harm the U.S. ecosystem". Fortune. Retrieved 2024-08-08.
- ^ a b c "SB 1047 Impacts Analysis". Context Fund.
- ^ "Assembly Judiciary Committee 2024-07-02". California State Assembly.
- ^ "Corporate Partners". Chamber of Progress.
- ^ "Members". Computer & Communications Industry Association.
- ^ "Members". TechNet.
- ^ a b Korte, Lara (2024-06-26). "Big Tech and the little guy". Politico.
- ^ Zeff, Maxwell (2024-08-21). "OpenAI's opposition to California's AI bill 'makes no sense,' says state senator". TechCrunch. Retrieved 2024-08-23.
- ^ "Little Tech Brings a Big Flex to Sacramento". Politico.
- ^ "Proposed California law seeks to protect public from AI catastrophes". The Mercury News.
- ^ "California's Senate Bill 1047 - What You Need to Know". Andreessen Horowitz.
- ^ "California's AI Bill Undermines the Sector's Achievements". Financial Times.
- ^ "Senate Bill 1047 will crush AI innovation in California". Orange County Register. 10 July 2024.
- ^ "AI Startups Push to Limit or Kill California Public Safety Bill". Bloomberg Law.
- ^ "The Batch: Issue 257". Deeplearning.ai. 10 July 2024.
- ^ "The AI Safety Fog of War". Politico. 2024-05-02.
- ^ "Anthropic says California AI bill's benefits likely outweigh costs". Reuters. 2024-08-23. Retrieved 2024-08-23.
- ^ Smith, Noah (2024-08-29). "Anthropic CEO Dario Amodei on AI's Moat, Risk, and SB 1047". "ECON 102" with Noah Smith and Erik Torenberg (Podcast). Event occurs at 55:13. Retrieved 2024-09-03.
- ^ Coldewey, Devin (2024-08-26). "Elon Musk unexpectedly offers support for California's AI bill". TechCrunch. Retrieved 2024-08-27.
- ^ Piper, Kelsey (2024-07-19). "Inside the fight over California's new AI bill". Vox. Retrieved 2024-07-29.
- ^ Piper, Kelsey (2024-06-14). "The AI bill that has Big Tech panicked". Vox. Retrieved 2024-07-29.
- ^ Robertson, Derek (2024-05-06). "Exclusive poll: Americans favor AI data regulation". Politico. Retrieved 2024-08-18.
- ^ Bordelon, Brendan. "What Kamala Harris means for tech". POLITICO Pro. (subscription required)
- ^ "New Poll: California Voters, Including Tech Workers, Strongly Support AI Regulation Bill SB1047". Artificial Intelligence Policy Institute. 22 July 2024.
- ^ Sullivan, Mark (2024-08-08). "Elon Musk's Grok chatbot spewed election disinformation". Fast Company. Retrieved 2024-08-13.
- ^ "Poll: Californians Support Strong Version of SB1047, Disagree With Anthropic's Proposed Changes". Artificial Intelligence Policy Institute. Retrieved 2024-08-13.
- ^ Gardiner, Dustin (2024-08-28). "Newsom's scaled-back surrogate role". Politico.
- ^ "Poll: 7 in 10 Californians Support SB1047, Will Blame Governor Newsom for AI-Enabled Catastrophe if He Vetoes". Artificial Intelligence Policy Institute. Retrieved 2024-08-28.
- ^ "California Likely Voter Survey: Public Opinion Research Summary". David Binder Research.
- ^ Lee, Wendy (2024-06-19). "California lawmakers are trying to regulate AI before it's too late. Here's how". Los Angeles Times.
- ^ Piper, Kelsey (2024-07-19). "Inside the fight over California's new AI bill". Vox. Retrieved 2024-07-22.