- within Law Practice Management and Coronavirus (COVID-19) topic(s)
California recently passed the Transparent in Frontier Artificial Intelligence Act (SB 53), which is the first comprehensive state-level AI safety framework in the United States. This law applies mostly to the large AI developers training models with extreme compute (10^26 FLOP) or earning $500m+ annually.
If you are a founder of a tech startup, it is not likely that this law applies directly to you. However, SB53 may still materially impact your startup business. SB 53 introduces regulatory, commercial, and reputational dynamics that are likely to extend well beyond the Golden State.
Below is a summary of what founders of early-stage AI companies and their investors should be preparing for.
AI governance in commercial agreements, financing and exit transactions:
- Core elements of SB 53-aligned governance will likely be included in future procurement checklists, representations and warranties and diligence processes. Even in the absence of a legal mandate, failure to implement basic AI governance protocols may disadvantage startups in commercial agreements, financing and exit conversations.
Establishment of industry norms:
- Practices such as red-teaming, security controls for model weights, and whistleblower protections are likely to become baseline expectations across the sector. This is not limited to entities that meet SB 53's technical thresholds.
Acceleration of compliance timelines:
- Founders should anticipate requests from investors and partners for documented compliance readiness earlier than historically expected. Early-stage companies may benefit from proactively integrating governance infrastructure into their product and organizational roadmaps. Treating AI governance not merely as a compliance matter, but as a component of strategic positioning may benefit institutional capital raises.
Impact on incumbent strategies:
- SB 53 is explicitly designed to limit the ability of large AI labs to avoid safety obligations in the pursuit of speed or scale. Plus, startups should consider engaging with CalCompute, California's new state-backed computing initiative, which is designed to provide access to infrastructure, guidance, and public-private research resources. These parts of SB 53 could reduce asymmetries that currently favor dominant players.
Strategic positioning for startups:
- Voluntary alignment with SB 53 practices can signal institutional readiness and mitigate reputational risk. California's AI regulatory leadership is likely to influence policy beyond state borders. Much like the GDPR's impact on global privacy standards, SB 53 may serve as a prototype for future federal or multi-state regulatory frameworks. Key definitions and enforcement thresholds are expected to evolve through 2027.
Comparison to the EU Artificial Intelligence Act:
- The EU Artificial Intelligence Act (January 2024) is broader and more of a burden for startups. Obligations focus on providers of high-risk AI systems and general-purpose AI models, with systematic risk. The EU Act's stricter compliance requirements apply to those trained on 10^25 flops (vs CA's 10^26). The EU act also requires regulatory submissions, while CA only requires the Frontier Developers and Large Frontier Developers to publish transparency reports. An earlier version of the CA law, closer in similarity to the EU act, was vetoed by Governor Newsome for being too broad and potentially stifling innovation.
While SB 53 may not affect you directly, we believe startups that embed governance and transparency into their operations will differentiate themselves in highly competitive markets and allow them to align with evolving standards, which has the ultimate benefit of de-risking future partnerships and potential financing and acquisition.
The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.
