[Noozhawk’s note: First in a series.]

Santa Barbara artificial intelligence luminary Zack Kass, whose newly released book, The Next RenAIssance, is all the rage, recently packed The Granada Theatre to explore the myriad of “possibilities” our AI-powered future holds.

Because our mission at Spotlight Santa Barbara is to “provoke thought and inspire change,” we decided not to wait for our AI overlords to take control before asking the “big” question: Should AI replace the Santa Barbara City Council?

ChatGPT wouldn’t take the bait so we rephrased the question: “Can AI make better data-driven policy choices than the Santa Barbara City Council?”

This time ChatGPT answered “yes,” explaining that it could “outperform” the council by developing more complete and less biased analyses to make “better data-driven decisions.” 

In this, part one of a three-part series, we put these claims to the test by asking ChatGPT, Claude and Copilot questions about rent control versus other options for improving housing affordability.

Even if AI never runs for office or secures a staff position in city government, our elected officials should be nervous.

AI makes it very easy for voters to determine whether our local politicians — presumably nonpartisan community stewards — are making competent decisions.

Rental Affordability

According to UCSB Economic Forecast Project director Peter Rupert, in the 30 years between 1990 and 2020 just 2,301 new housing units were built within Santa Barbara’s city limits.

Fun Fact: In an NYU study across 502 small cities (50,000–100,000 in population), the average vacancy rate was 8.9% while Santa Barbara’s vacancy rate is just 2.8%.

With so little inventory, is it any wonder that rents have risen? In an effort to force inventory up and pricing down, California has “mandated” that we build four times as many units in just eight years.

The City Council, however, is focusing its efforts on a heavy-handed local rent control ordinance — the state already has one — rather than energizing the private sector to meet the state mandate or otherwise address affordability concerns.

Question to Claude: Has rent control proven to be an effective policy to drive rental pricing down?

Answer: “Most economists — across the political spectrum — conclude that while rent control can protect existing tenants in the short term, it tends to be ineffective or counterproductive (emphasis added) at actually driving down rental prices broadly.

“A well-known Stanford study (Diamond et al., 2019) on San Francisco found that tenants in rent-controlled units were significantly more likely to stay in their homes. So, as a tenant stability tool, it has real value.

“However, the same study found that landlords responded by converting rental units to condos, selling them or redeveloping properties — reducing the overall rental supply by about 15%. This supply reduction pushed up market-rate rents for everyone else, meaning the city’s overall rent levels actually increased (emphasis added).

“Rent control creates a two-tier market. Insiders (those lucky enough to hold controlled units) benefit, while outsiders (new renters or those priced out) face a tighter, more expensive market.

“It also: 1. Discourages new construction, since developers anticipate lower returns; 2. Reduces maintenance and quality of housing stock, since landlords have less incentive to invest; and 3. Creates ‘’misallocation’ — people stay in units that no longer fit their needs (a single person staying in a large apartment, for example) because moving means losing their controlled rent.

“The IGM Forum (Initiative on Global Markets), which surveys top economists from across the ideological spectrum, found in 2012 that 93% either disagreed or strongly disagreed that rent control had a positive effect on the amount and quality of broadly affordable rental housing. That’s a rare level of agreement among economists.”

Authors’ note: The American Enterprise Institute found an even greater economist consensus in a February 2026 survey/study, “Rent Control: A Proven Way to Make Housing Scarce and Expensive.”

Claude vs. City Council

Contrary to Claude’s compelling analysis, a majority of the seven-member City Council — Kristen Sneddon, Oscar Gutierrez, Wendy Santamaria and Meagan Harmon — recently voted for a rent freeze and appear to be moving toward an ordinance that would be significantly more restrictive than the already existing 2019 state rent control law (Assembly Bill 1482).

Indeed, on Jerry Roberts’ Newsmakers show, UC Santa Barbara sociology professor emeritus Rich Applebaum (who is purportedly advising the City of Santa Barbara on housing and rent control, which he favors) argued that economic analysis does not undermine the case for well-designed rent control.

We put this assertion to the test by asking ChatGPT to solve housing affordability in Santa Barbara without any guidance from us.

If the case for rent control were strong despite the economic evidence to the contrary, then ChatGPT should prioritize it in its analysis.

Question to ChatGPT: What are the three most important policies the Santa Barbara City Council could adopt to make housing more affordable?

Answer: “To make housing meaningfully more affordable, the Santa Barbara City Council would need to focus on policies that increase supply, reduce construction costs, and allow more diverse housing types in high-opportunity neighborhoods within Santa Barbara.”

ChatGPT vs. City Council

The city’s only policy option for addressing rental affordability failed to make the cut with ChatGPT.

Rather than follow three evidence-based policy recommendations from ChatGPT, a majority of the City Council is instead pursuing a rent control ordinance that will almost certainly prove to be counterproductive in the mid- to long term.

In sum, AI was not wrong when it said it could “outperform” the City Council in certain types of policy making.

It definitely understands rental affordability better than city “advisers” and many council members.

Our advice to the City Council: If you don’t want to listen to actual economic experts — like Peter Rupert — when making rent control policy, then at least listen to AI.

Although AI may not be able to replace the council, voters who ask AI the right questions could certainly choose to do so.

What’s Next?

In Part II, we’ll have AI take aim at constructions costs, regulations and other aspects of our local housing market.

In Part III, we’ll ask it to look at our relatively dismal city finances and how to best reinvigorate our dying downtown State Street corridor.