Expert attention map

The Who's Who of AI

What credible people across AI noticed, why it matters, and where the field is converging or disagreeing.

2,397 searchable experts 4,280 tracked across all sources
Build my stream Create my edition
Filter the conversation Who is saying what?

Combine a professional role with a reaction lens. Both must match the same attributed contribution.

Clear all
New Network reaction maps

See how experts are reacting—not just what they shared.

Posts are grouped by conversation and tone. Select a lens to filter the stream; these are never permanent labels on people.

Building & implementation Research & technical analysis Policy & governance
2 live reaction maps in this view Jump to the evidence ↓
Multiple readings State media control influences large language models | Nature Policy & governance · 5 attributable reactions Multiple readings Inference Cooperative — AI, governed together Building & implementation · 2 attributable reactions
Showing signals surfaced by B! 🐝 Cavello (they/them) ×
Debated

The questions experts are actively pulling apart

One card per development. Sources are clustered; reaction bundles describe these posts, never the people behind them.

Established Culture, work & education Analysis 4d ago
Anti-AI art manifesto

anti-ai art manifesto

9 experts are actively discussing the implications.

1 expert 1 community 1 sources clustered
B! 🐝 Cavello (they/them)

“one of my fave pieces on this topic: multimodal.art/manifesto/”

9 experts discussed this · 64 posts
SE Gyges: the frustrating part of bad ai takes is that nobody who has them wants to stand by them or admit their misses later. it is just the rocket goalposts game.
SE Gyges: see i don't remember you ever having especially bad takes, the sort of person who has really bad takes is the sort of person who does not learn from or even admit their mistakes
Eli Sennesh: the frustrating part of bad ai takes is that nobody who has them wants to stand by them or admit their misses later. it is just the rocket goalposts game.
Open the full discussion →
Primary source multimodal.art ↗
Established AI field signal Analysis 5d ago
⚡ 6 h early
essay how to talk AI

How to talk about

11 experts across 2 network communities independently surfaced this.

11 experts 2 communities 1 sources clustered
Emily M. Bender Timnit Gebru Hypervisible Luke Stark Eileen Clancy 🧿 Frank Pasquale Charles Logan Jeffrey Vagle +3

“For reasons mysterious to me, Will Stancil is continuing his crash out about language used to describe LLMs, moving on from raning about "stochastic parrots" to picking up on this piece that @nannainie.bsky.social and I published earlier this summer: button…”

“Instead of railing against the latest “I asked Claude” article in a mainstream publication, I’ll just post this.”

2 experts discussed this · 16 posts
Emily M. Bender: So many conversations where otherwise serious people throw around the term "AI" without defining it, despite it being notoriously ill-defined. Thinking about why this might be >>
Emily M. Bender: In some cases, no doubt, the speaker fully believes that autonomous thinking machines are among us. (Though I said "otherwise serious people" in the prev post, so.) >>
Emily M. Bender: In other cases, I suspect that people feel like they lack the expertise to call bullshit and are repeating what they hear others say. >>
Open the full discussion →
Primary source buttondown.com ↗
Established Models & releases Research 3d ago
⚡ 953 h early
State media influence LLM training paper

State media control influences large language models | Nature

10 experts across 4 network communities independently surfaced this.

10 experts 4 communities 1 sources clustered
Justin Hendrix Luke Stark Henry Farrell Eli Sennesh Gina Helfrich Dr. Chinasa T. Okolo Debora Nozza MilaNLP Lab +2
How the network is reacting Experts are approaching this through 3 distinct lenses.
Multiple readings

Policy & governance

2 experts

Rules, institutions and accountability.

“"...LLMs exhibit a stronger pro-government valence in the languages of countries with lower media freedom than in those with higher media freedom." www.nature.com/articles/s41...”

Research & technical analysis

2 experts

Evidence, methods and technical implications.

“Researchers from Oregon, Purdue, UC San Diego, NYU and Princeton ran six experiments. The overarching finding: state control of media in many countries is already baked into the training data of widely used commercial models, and is shaping the answers thos…”

Building & implementation

1 expert

How teams are shipping and applying it.

“Builds on stodgier academic work here - www.science.org/stoken/autho... - this, by Roberts et al, is also very interesting on language and LLMs albeit much more political www.nature.com/articles/s41...”

2 experts discussed this · 2 posts
VE, cybersocial occult investigator: Starting a conspiracy theory that the papal encyclical was actually trying to get ahead of and cement Catholicism as the official machine religion
René Walter: OMG www.nature.com/articles/s41...
Open the full discussion →
Primary source nature.com ↗
New Policy & governance Resource 22h ago
inference cooperative AI governance

Inference Cooperative — AI, governed together

3 experts across 3 network communities independently surfaced this.

3 experts 3 communities 1 sources clustered
Ronen Tamari Nathan Schneider B! 🐝 Cavello (they/them)
How the network is reacting Experts are approaching this through 2 distinct lenses.
Multiple readings

Building & implementation

1 expert

How teams are shipping and applying it.

“Maybe relevant @ntnsndr.in in context of inference.coop (congrats on the launch!)”

Research & technical analysis

1 expert

Evidence, methods and technical implications.

“We are now about half way to the threshold at Inference.coop where I'll have to give up my benevolent dictatorship. Want to join the ethical-ish AI train and help unseat me from power?”

2 experts discussed this · 3 posts
Nathan Schneider: We are now about half way to the threshold at Inference.coop where I'll have to give up my benevolent dictatorship. Want to join the ethical-ish AI train and help unseat me from power?
B! 🐝 Cavello (they/them): We are now about half way to the threshold at Inference.coop where I'll have to give up my benevolent dictatorship. Want to join the ethical-ish AI train and help unseat me from power?
B! 🐝 Cavello (they/them): @roblach.bsky.social shall we?? #NoKings 😜
Open the full discussion →
Primary source inference.coop ↗
Established AI field signal Analysis 18d ago
⚡ 742 h early
Percepta AI-native hospital vision

Building the AI-native hospital | Percepta

3 experts across 2 network communities independently surfaced this.

3 experts 2 communities 1 sources clustered
Eugene Vinitsky angela zhou B! 🐝 Cavello (they/them)

“Lets make hospitals better: https://t.co/YkHWGNexJi”

2 experts discussed this · 8 posts
Eugene Vinitsky: At the startup, we are seeing large productivity gains time after time in deployments of AI that are done thoughtfully. There are so many important things that aren't done because there simply aren…
Eugene Vinitsky: What does thoughtful mean? Domains that have reasonable oversight, or a human-in-the-loop, or where people already have high failure rates, or where being able to do the activity at all outweighs a…
ae: At the startup, we are seeing large productivity gains time after time in deployments of AI that are done thoughtfully. There are so many important things that aren't done because there simply aren…
Open the full discussion →
Primary source percepta.ai ↗
Established AI research Resource 5d ago
deep learning AI educational videos

B’s AI video recommendations

2 experts are actively discussing the implications.

1 expert 1 community 2 sources clustered
B! 🐝 Cavello (they/them)

“CodeEmporium’s breakdown of transformers uses some more scary looking diagrams and words (“vector”), but I think the actual explanations are pretty accessible, if you can just encourage people to gloss over the unfamiliar terminology youtu.be/TQQlZhbC5ps?...”

2 experts discussed this · 7 posts
Grace Lindsay: Anyone know of any good short videos or visualizations that cover how modern AI is built and trained in a way that is suitable for a college gen ed class? It can lean a bit technical. The stuff I'v…
Grace Lindsay: Ooo that paper ChatGPT demo is good!
B! 🐝 Cavello (they/them): I have a playlist of some of my favorite videos for exploring concepts in machine learning/algorithms/AI! youtube.com/playlist?lis...
Open the full discussion →
Primary source youtube.com ↗ youtu.be ↗
Established Models & releases Signal 20d ago
⚡ 4 h early

A new class action lawsuit questions whether Anthropic broke the law by misleading power users

3 experts across 3 network communities independently surfaced this.

3 experts 3 communities 1 sources clustered
Dare Obasanjo Hayden Field B! 🐝 Cavello (they/them)

“In an expanded class-action lawsuit filed today, a group of Claude subscribers say the company deceptively advertised the limits of its Max subscription tier. www.theverge.com/ai-artificia...”

2 experts discussed this · 2 posts
Hayden Field: In an expanded class-action lawsuit filed today, a group of Claude subscribers say the company deceptively advertised the limits of its Max subscription tier. www.theverge.com/ai-artificia...
B! 🐝 Cavello (they/them): In an expanded class-action lawsuit filed today, a group of Claude subscribers say the company deceptively advertised the limits of its Max subscription tier. www.theverge.com/ai-artificia...
Mike Masnick: Hey Hayden. Am I missing it or does your article not include a link to the filing? Would like to read the underlying complaint and I don't see a link, or even the name of the lead plaintiff to look…
Open the full discussion →
Primary source theverge.com ↗
Established AI field signal Resource 8d ago
Public AI Network launched

Public AI Network

3 experts are actively discussing the implications.

1 expert 1 community 1 sources clustered
B! 🐝 Cavello (they/them)

“Yes yes yes yes! #PublicAI publicai.network/whitepaper/”

3 experts discussed this · 4 posts
Eryk Salvaggio: Should we be talking more about AI as infrastructure? Modern infrastructure is "collective intelligence," as @debcha.bsky.social argues. I'm thinking about this today because of @tressiemcphd.bsky.…
Matthew Kirschenbaum: I also think @miriamposner.com’s brilliant soon to be published book on SEEING LIKE A SUPPLY CHAIN will help a lot here, including moments when she makes the parallels between supply chains and the…
Miriam Posner: Thank you so much! It seems to me that a number of authors are sort of converting on this point independently. The more time you spend staring at it, the more it just seems clear.
Open the full discussion →
Primary source publicai.network ↗
Established AI field signal Signal 19d ago

static1.squarespace.com

2 experts are actively discussing the implications.

2 experts 2 communities 1 sources clustered
Serife (Sherry) Wong B! 🐝 Cavello (they/them)

“Many trust Anthropic leaders' good intentions and it's being confused with trustworthiness of mechanisms & institutions. These aren't the same thing as I learned from the inimitable bad ass political scientist Margaret Levi when we wrote this brief: static1…”

2 experts discussed this · 4 posts
Serife (Sherry) Wong: All bots are cops.
Serife (Sherry) Wong: @ali-alkhatib.com coined this phrase and has stickers :) Pop it on your signboards and march with it. Says it all really. It is a solid framing and a message to rally behind that can travel quickly…
Eileen Clancy 🧿: All bots are cops.
Open the full discussion →
Primary source static1.squarespace.com ↗
Established Responsible AI Analysis 8d ago
⚡ 56 h early
AI taboos argument essay

A case for AI taboos – Writings and rehearsals by Nathan Schneider

2 experts are actively discussing the implications.

2 experts 2 communities 1 sources clustered
Nathan Schneider B! 🐝 Cavello (they/them)

“I keep coming back to this list I started on AI taboos: nathanschneider.info/2026/06/a-ca... What else should be on it?”

2 experts discussed this · 3 posts
Nathan Schneider: A case for AI taboos: https://nathanschneider.info/2026/06/a-case-for-ai-taboos/
B! 🐝 Cavello (they/them): A case for AI taboos: https://nathanschneider.info/2026/06/a-case-for-ai-taboos/
B! 🐝 Cavello (they/them): Love this! You’ve been doing so much awesome writing and public thinking. Thank you!
Open the full discussion →
Primary source nathanschneider.info ↗
Established Models & releases Signal 10d ago

Most devs don't understand how LLM tokens work

3 experts are actively discussing the implications.

1 expert 1 community 1 sources clustered
B! 🐝 Cavello (they/them)

“I was JUST researching tokenization today! I found this pretty helpful: youtu.be/nKSk_TiR8YA”

3 experts discussed this · 3 posts
samwho.dev: ok dorks: without looking it up, what is a token? (in LLMs) hand-wavy BS is fine if you're not sure. im doing some research
Amy Hoy: a symbol for some tiny slice of semantic meaning aka a word or word part
alanau.bsky.social: You mean like tokenizing data, or do you mean the proprietary scrip used to buy cloud services? 🤷
Open the full discussion →
Primary source youtu.be ↗
Established Models & releases Signal 14d ago

Drey Dossier on Instagram: "The billionaires have bravely volunteered to regulate everyone else. #economics #anthropic"

2 experts are actively discussing the implications.

1 expert 1 community 1 sources clustered
B! 🐝 Cavello (they/them)

“Not sure about writing, but @jmpasner.bsky.social and I were talking about this: www.instagram.com/reel/DdP2vKY...”

2 experts discussed this · 3 posts
Tamara Kneese: Is anyone writing (in a critical/compellimg way) about frontier lab AI slowdown calls amid the broader data center pause/moratorium push in San Francisco/the larger Bay Area?
Tamara Kneese: In case it’s not clear this is why I’m asking bsky.app/profile/evan...
B! 🐝 Cavello (they/them): Not sure about writing, but @jmpasner.bsky.social and I were talking about this: www.instagram.com/reel/DdP2vKY...
Open the full discussion →
Primary source instagram.com ↗
More conversations

What experts are discussing without an anchoring article

3 experts · 48 posts · 2d ago
Grace B! 🐝 Cavello (they/them) Singularity's Bounty e/cc
Multiple readings Policy & governance · 1 Building & implementation · 1 Questions & unknowns · 1
Grace: AI writing, unfortunately
Grace: What do we do about this? Substack lets you check for AI with pangram, but it doesn’t show a label upfront. I assume if it started doing that authors would optimize against it. Maybe build it into …
Open the thread →
6 experts · 6 posts · 5d ago
Naomi Saphra Maria Antoniak David Mimno Gautam Kamath The Data Therapist in the Blue Sky B! 🐝 Cavello (they/them)
Naomi Saphra: A pattern that I've been surprised by is people writing followup emails for their slop emails. Like, people are getting really insistent that I respond in earnest to emails they did not write.
The Data Therapist in the Blue Sky : s/people writing/AI agents generating/ s/people are getting really insistent/AI agents emailmaxxing/
Open the thread →
3 experts · 13 posts · 17d ago
Luis Villa blair B! 🐝 Cavello (they/them)
Shared emphasis Concern & critique · 2
B! 🐝 Cavello (they/them): This is a good point, but I actually think it would be good to discuss what the difference is. I think that there are some technologies, the existence of which is actually dangerous, in part becaus…
B! 🐝 Cavello (they/them): I think that many of us are aligned in being concerned about the harms and have tried to intervene before things got to such a point. Others felt that such a point was unavoidable (I disagree, but …
Open the thread →
4 experts · 4 posts · 17d ago
Mark Riedl Ben Recht Eryk Salvaggio B! 🐝 Cavello (they/them)
Eryk Salvaggio: It's been lower on my priority list for things to write up, but METR being in the strict-intentionalist soup is an important part of this story. So is their use GPT-5.6 Sol as a classifier to parse…
Ben Recht: Also important, they are the same people: - their founder and a large part of their staff have worked at these companies. - their most outspoken researcher is married to someone on the OpenAI board…
Open the thread →
3 experts · 6 posts · 15d ago
Prof. Dr Beth Singler Jack Stilgoe B! 🐝 Cavello (they/them)
Shared emphasis Policy & governance · 2
Jack Stilgoe: For AI governance, 'pacing', 'pausing' and 'slowing down' are the wrong metaphors. Only used by people who want this to be seen as a race. This is about steering, not braking. But...
Jack Stilgoe: We can think about companies going slower would actually mean. It would not mean 'we need to stop ourselves being so clever'. It would mean 'we need to build infrastructures that make everyone safe…
Open the thread →