In DC Sep-Dec 2026 Associate Professor of Finance, @NYUStern. Finance/Real Estate/Urban newsletter: https://xitter.cf/t.co/lWVJvq6vKg.
Manhattan, NY
Joined September 2015
- Tweets67.2K
- Following5.2K
- Followers50.4K
- Likes82.6K
Pinned Tweet
🚨New predoc position!🚨
@AlexBartik and I are hiring a predoc at NYU Stern, starting Fall 2027
You would help us build AI tools to measure zoning rules and study how regulation impacts housing supply
Ideal for AI enthusiasts who want to get a PhD
apply.interfolio.com/194498
Arpit Gupta retweeted
Very happy to share a new paper, “Technology Bottlenecks: Measuring Where AI Could Raise the Value of Work It Cannot Yet Do.”
AI exposure measures ask about the work AI is capable of doing. We ask instead where the marginal value of work AI cannot yet do may rise as AI speeds up the rest of the job.
To do so, we employ 288 million U.S. job postings mapped line by line to their closest O*NET tasks and GPT-4 feasibility ratings from Eloundou et al. (2024) to construct a measure, “bottleneck pressure,” for each task in each occupation that posts it.
Using a simple task-based framework, we derive this pressure at the task level, showing that for an AI-infeasible task it is the exposure of the occupation that posts it, so the same task can face very different bottleneck pressure in different jobs.
Reporting bottleneck pressure alongside AI exposure shows both the scope for AI to perform work and where improving the work it cannot do may pay off more as capabilities expand.
Arpit Gupta retweeted
Finally got to put together a complete deck on the state of open models.
What they are, licenses, how competitive they are, signs people are using them, distillation's impact, and how we balance the risks and benefits of openness!
Will record a video lecture of these soon, and hopefully re-do them a few times a year.
The solution to stale unemployment insurance databases is in sight
A great research topic for someone is the role of recruiting filters at companies in contributing to the weakness in junior hiring
I wonder if they all have experience filters which hit the young
I'm a fan of UChicago's approach — reading and writing labs, but also Claude access for the masses
Arpit Gupta retweeted
The IBX is coming, and it should be automated. ETA’s newest report, published today, covers the benefits of automated light metro (ALM) technology in detail, and why now is the time for the MTA to incorporate automation into the project’s design.
I'm not sure where this new push for neighborhood schools will go. More geographical sorting within cities? More urban flight by parents who don't like their local school?
sfstandard.com/2026/10/15/la…
Arpit Gupta retweeted
Super interesting paper on U.S. elites, 1850–1950.
The top tail of U.S. asset owners experiences rapid turnover.
70% of individuals in the top 1% are no longer there 10 or 20 years later.
Arpit Gupta retweeted
🚨 JMP: How do firm promotion policies shape gender promotion gaps?
In large law firms, we find that:
1. Part-time work is stickier for women and explains much of the status quo promotion gap.
2. Policy changes can narrow the gap, but with varying effects on firm productivity.
You could imagine a world in which US labs focused on monetizing open source models, by selling inference services and add ons, rather than only (or primarily) around selling access to foundational models
🚀 Introducing DeepInfer, our in-house inference framework, now fully open source!
Powering DeepSeek-V4.1-Flash at scale with high throughput and low latency. We’re sharing our production stack to help the community lower compute costs, reduce energy use, and make AI more affordable.
🔗 github.com/deepseek-ai/DeepI…
1/4
Arpit Gupta retweeted
We're introducing TasteBench, a new standard to measure Research Taste on real world ML tasks.
Existing benchmarks have limited datasets and eval data ultimately becomes part of training data. In TasteBench, we continuously generate new ML tasks on OOS data no model has trained on.
Arpit Gupta retweeted
🚨 New paper🚨 Can LLMs predict how households spend, work and respond to policy? We build a benchmark to find out: HouseholdBench. With great coauthors @jinhuang9306000, @DiegoFerreras, Yutong Xie, Walter Yuan, Qiaozhu Mei and @ChenLian92. (1/7)
Arpit Gupta retweeted
Theoretical models of using parallel AI agents to achieve RSI assume scaling shapes that are mostly power laws. Real world experiments show much steeper diminishing returns - logarithmic or even sub-logarithmic - limiting RSI via mass parallelization. 16/25
Arpit Gupta retweeted
Over the weekend, I ran an experiment: Can LLM agents, individually or in teams, manage a weekly cross-asset portfolio better than simple systematic rules?
1/8
Arpit Gupta retweeted
TIL that France’s property tax system is even more generationally cursed than Californias Prop 13.
(Ontario fwiw has had them frozen for a decade now, something we have to change.)
- Property tax valuations/assessments frozen at 1970 values
- Revaluations based on renovations have disincentivized improving homes
- Capital gains taxes only charged on young people (basically)
- Property taxes higher on poorer and younger places
Obviously this isn’t as generationally horrifying as the French pension system.
But these dynamics are real across the western world, and if we don’t grapple with their unfairness, I’m not sure how we get through the next few decades.
We’ve saddled the young with too many bills they cannot afford to pay.
theborys.substack.com/p/how-…
These new *conditional* prediction markets are interesting, but odd that they (for now) imply Dems are better in the Senate with lower gas prices
Maybe: the events that lower gas prices would be on net good for Dems (recession?)
If you are voting in AFA elections, please vote for Stijn Van Nieuwerburgh as Second Vice President!
There are few people in the profession who have invested more in training and mentoring, or in communication and impact in research. He will upgrade what the AFA does here.
Arpit Gupta retweeted
Big new IFP report from @GeorgiaCAdamson, @KonstantinPilz, and me on the US-China AI competition.
We spent the last 6 months digging into how Chinese AI companies’ access to compute and distillation of American models translate into the current US-China capabilities gap.
The result is a 23k-word research report, including an interactive analytical model that visualizes changes in the US AI lead under different policy interventions.
Our findings:
1. The US AI lead over China is currently only around 6 months.
2. The US could roughly double this lead, to over a year, using authorities already available to the federal government.
3. On chips: China’s access to AI chips will roughly double over the next year, with chips housed in data centers overseas accounting for roughly half of China’s 2027 compute.
4. On distillation: the current ~16x US training compute advantage would normally yield a 13-month lead. But we estimate adversarial distillation halves this to around 6.5 months.
5. Increasing the lead would have some obvious benefits:
- Give the US broadly a greater cyberdefensive advantage against Chinese threat actors.
- Give US AI companies breathing room to spend more R&D resources on making their systems safer and more secure, without fear of being overtaken by less safety-conscious Chinese competitors.
6. Provide the US government more leverage in negotiations with China about the future of AI.
7. The two most important interventions are:
- For BIS: Create a new end-use control to restrict China’s remote access to training clusters abroad — China’s largest and fastest-growing source of frontier-scale compute.
- For DOJ/FTC: Create a narrow safe harbor to allow US companies to coordinate on countering Chinese adversarial distillation.
8. Both interventions (and all the others we discuss in the report) are things the federal government could do today.
9. To estimate the effects of each policy intervention, we model the US lead as a function of its compute advantage over China and China’s ability to adversarially distill the capabilities of the US frontier.
10. The model makes many simplifying assumptions, but provides a useful way to assess and rank the impact of different policy interventions at a high level.
Read the full report here: ifp.org/how-to-increase-amer…
