Annie Tsai

I’ve been told more than once that running companies in the private sector is very different from driving change in the public sector. I’ve learned this to be somewhat true in practice, but I don’t think it is always true in all cases. 

A notable difference is that there are legal constraints protecting our democratic process like  the Brown Act, which requires 72-hour agenda notification periods, that force pick up conversations to mostly get tabled until there’s a spot on the agenda in a scheduled meeting. 

Also included in Brown Act is the perceived requirement to not discuss anything not agendized, so non-agendized public comment can’t receive a reply but can be requested to be discussed on a future agenda. This felt strange to me in the beginning as a private citizen, but you learn fast. 

For leading in both environments, I do think it’s a balancing act of “best compromises” to “best outcomes” in order to achieve vision and drive tactics simultaneously given the known constraints and opportunities – what do we want for the long term, and what can we do in the short term while we work towards the long term? Any leader will tell you that it’s rare for the initially envisioned “ideal state” to be waiting for you at the finish line, and sometimes even, what was once ideal is now an impediment. In my day job, it’s all about navigating this balancing act and ensuring the two work streams support each other and eventually converge. 

Doing so successfully is both art and science and I don’t think the public sector works differently here. 

Speaking of art and science, Hollywood went big on backing SB 1047 last week but fortunately for everyone, the bill sat unsigned on Governor Newsom’s desk up until hours before his deadline to veto, which he did so on Sunday early afternoon. 

As I previously stated, “The size of the LLM is not the determining factor for harm capacity. Rather, the determining factor for harm is how the model is used,” and continue to advocate for regulation that focuses on harm reduction and threats as opposed to regulation for regulation’s sake.

Governor Newsom echoes these concerns in his Veto Message when he says, “By focusing only on the most expensive and large-scale models, SB 1047 establishes a regulatory framework that could give the public a false sense of security about controlling this fast-moving technology. Smaller, specialized models may emerge as equally or even more dangerous than the models targeted by SB 1047 – at the potential expense of curtailing the very innovation that fuels advancement in favor of public good.” 

I for one am glad that it was sent back to the drawing board as I am a strong proponent for application layer driven definitions. Regulate at the application layer and you will actually protect consumers from real and not theoretical threats that are happening to us today and into the near future.

A great example of application layer regulation is the set of five bills signed on September 27 regulating AI and deepfakes, which I also wrote about previously. It makes a lot of sense to regulate this very big known issue, and the definitions are specific yet broad enough where regulation can be both “exact” and “protective”. 

Bills that get passed don’t always check both of these checkboxes, of course. For example, SB 35 was first passed as a part of something like 14 housing bills from 2017-18 intended to beef up the RHNA (Regional Housing Needs Assessment) process, which the state uses to calculate and allocate every city’s housing target every 8 years and speed up development. However, further refinement was needed and SB 423 was passed last year, which significantly increases the ability for the state to hold jurisdictions accountable to their RHNA targets. The governor recently signed another 32 housing bills to make further streamlinings to existing law.

San Francisco was the first city in the state to trigger SB 423 over the summer by not having a certified Housing Element, taking their permitting process from one of the longest to now one of the shortest with no community, planning commission, or CEQA (California Environmental Quality Act) review…. through 2036. Any city that fails to show that they are on track to achieving their RHNA allocation at the cycle’s midpoint – or beginning of 2027 – will trigger SB 423 and ten years of revoked local control.  

It will certainly be interesting to see how loss of local control impacts the development pipeline and makeup of San Francisco. Come 2027, all cities who are not on track to achieve their RHNA targets will have an idea of what to expect. Even with high interest rates and cost of construction, a lot can happen in 10 years.