James Zou

@james_y_zou · Palo Alto, CA · joined 07 Aug 2016

@Stanford professor. Chan-Zuckerberg investigator. Sloan Fellow. Overton Prize. @togethercompute. AI for science + health.

25 209Followers

143Following

1 878Posts total

271.5KViews on collected posts

Against accounts of the same size

12 posts from the last 90 days, next to the 10K–100K follower range. shown widely, but few of those viewers react.

Median views11 413this account949median for 10K–100K

Reach, %45.27%this account3.23%median for 10K–100K

Engagement, %1.08%this account2.00%median for 10K–100K

Metric	This account	Median for 10K–100K	Ratio
Median views per post	11 413	949	12.0×
Reach (views ÷ followers)	45.27%	3.23%	14.0×
Engagement rate	1.08%	2.00%	0.54×

Others in this range → Compare with another account → How these benchmarks are built →

Growth & engagement

How the posts we collected actually performed: views and reaction rate post by post, what the audience did with them, and where the follower count goes.

Views per post

20.2K5 Aug

6.5K19 Aug

144.7K

3.4K

4.3K

576

16.3K20 Aug

3.5K24 Aug

25.3K26 Aug

25.6K3 Sep

17.2K4 Sep

Last 12 collected posts, oldest on the left. The scale is logarithmic: one post can outrun the rest a hundred times over.

Engagement rate per post

0.33%5 Aug

1.61%19 Aug

1.04%

1.52%

1.12%

2.43%

1.22%

0.91%20 Aug

0.60%24 Aug

0.61%26 Aug

0.88%3 Sep

1.15%4 Sep

Reactions — likes, reposts, replies and quotes — divided by views. Median for 10K–100K accounts is 2.00%.

What the audience does

Likes54.6%2 205 in total

Reposts7.7%312 in total

Replies1.7%70 in total

Quotes1.1%43 in total

Bookmarks34.8%1 406 in total

Share of every reaction we collected for this account. Replies mean argument, reposts mean endorsement, bookmarks mean the post was worth keeping.

The follower curve appears once this account has two daily snapshots — we take one a day, and this one is on its first.

Latest posts

Excited that Tabula Sapiens 2 is out at @CellCellPress! >1 million cells mapped across 28 human tissues. All this data is open! https://t.co/zCI30Hjxmx 17.2K views · 166 likes · 29 reposts · 3 replies 04 Sep 2026 · Open on X →

Your paper is cited 200 times. Maybe 8 of them actually read it. Paper Scorecard finds those 8 and surfaces the real critiques buried in literature. Fascinating to see what papers really say about each other! https://t.co/GTwsLbSheo https://t.co/r6ISl88Zli 25.6K views · 191 likes · 24 reposts · 5 replies 03 Sep 2026 · Open on X →

Over 1.8 million full text patents and 170 million abstracts are now on #paperclip! All indexed in the same agent-native filesystem w/ millions of papers, so you can easily search across literature + patents. Most drugs are first disclosed in patents so this is very useful! http 25.3K views · 131 likes · 17 reposts · 2 replies 26 Aug 2026 · Open on X →

string2string is accepted as an EMNLP demo! Great job @suzgunmirac🚀 3.5K views · 19 likes · 2 reposts · 0 replies 24 Aug 2026 · Open on X →

A lot of people are curious about what the agents actually say to each other, so @elb4tu built a nice website to explore our Physics of Agents experiments: https://t.co/UGZgvrt8MW You can visualize the social dynamics in these agentic communities! 16.3K views · 134 likes · 14 reposts · 1 replies 20 Aug 2026 · Open on X →

@james_y_zou This reminds me of Maya Okawa's work on consensus building. Really awesome how these topologies can behave like Ising models. https://t.co/MwNOGWMqhY 576 views · 6 likes · 0 reposts · 1 replies 19 Aug 2026 · Open on X →

n/ The key insight is that AI agents drift to low energy states that minimize their social pressure. Understanding such collective behavior is important for safety and alignment of multi-agent systems. See our paper for more discoveries! https://t.co/FIaeeUxs6V 4.3K views · 88 likes · 9 reposts · 6 replies 19 Aug 2026 · Open on X →

3/ Each community is a social network of agents with different personas. They discuss questions ranging from math problems to statements like: “Nuclear power is essential for clean energy.” Agents talk with their neighbors and update their opinions across multiple rounds. https 4K views · 39 likes · 5 reposts · 1 replies 19 Aug 2026 · Open on X →

4/ Some agent communities reach consensus. Others become more polarized over time. And we can predict these dynamics! We developed an Ising-type model and fit it to data of agent behavior. The model predicts the community's future. Generalizes to new social graphs + questions! 3.4K views · 44 likes · 5 reposts · 2 replies 19 Aug 2026 · Open on X →

New work I'm very excited about: Physics of Agents!🤖 As AI agents become more prevalent, they'll interact and influence one another. Can we predict the collective behavior that emerges? Surprisingly, their dynamics follow compact, predictive laws of statistical physics🧵 https: 144.7K views · 1.2K likes · 183 reposts · 46 replies 19 Aug 2026 · Open on X →

2/ We studied >10,000 communities of LLM agents that talk to each other and update opinions over time. We show that collective behavior of interacting AI agents can be modeled and predicted using stat mech. https://t.co/Vbb6mh0i61 Led by @elb4tu w/ great collaborators https: 6.5K views · 91 likes · 12 reposts · 2 replies 19 Aug 2026 · Open on X →

1/ Today, @james_y_zou, @pmphlt, @jurafsky, and I are releasing string2string Studio: an open-source, in-browser platform that brings alignment, comparison, search, and evaluation into one interactive, visual environment for strings and sequences. See: https://t.co/OfAa4rSe5T h 20.2K views · 52 likes · 12 reposts · 1 replies 05 Aug 2026 · Open on X →