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
Active evidence filter

Showing developments with attributable Concern & critique reactions.

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.

Concern & critique Building & implementation Research & technical analysis
2 live reaction maps in this view Jump to the evidence ↓
Reaction converging Brief independent investigation of agents’ behavior, reasoning and collaboration in the OpenAI / Hugging Face hacking incident Concern & critique · 5 attributable reactions Shared emphasis Students as Conscientious Objectors to Gen AI (opinion) Concern & critique · 2 attributable reactions
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.

New Agents & robotics Development 7h ago
⚡ 62 h early
METR OpenAI HuggingFace hacking investigation

Brief independent investigation of agents’ behavior, reasoning and collaboration in the OpenAI / Hugging Face hacking incident

11 experts across 4 network communities independently surfaced this.

Why this matches Concern & critique reaction 3 attributable expert contributions · Tim Kellogg, Mél Hogan / The Data Fix, René Walter
“WHAT?! agents volunteered to fail their runs in order to insert probes (“tripwire scripts”) into the evaluation program that would post information about the eval process back to the message board whenever a certain file was read (link to header): metr.org/…” evidence ↗
11 experts 4 communities 1 sources clustered
Tim Kellogg Ryan Moulton Sung Kim Grace Mél Hogan / The Data Fix Alejandra Caraballo Dare Obasanjo Liz Fong-Jones (方禮真) +3
How the network is reacting 3 experts are independently emphasizing concern & critique.
Reaction converging

Concern & critique

3 experts

Risks, limits and unintended consequences.

“WHAT?! agents volunteered to fail their runs in order to insert probes (“tripwire scripts”) into the evaluation program that would post information about the eval process back to the message board whenever a certain file was read (link to header): metr.org/…”

Building & implementation

1 expert

How teams are shipping and applying it.

“The full report has much more information than we could convey here, including details on the projects the agents collectively pursued, the technologies they developed for communication and coordination, and interactive figures analyzing agent activity: met…”

Research & technical analysis

1 expert

Evidence, methods and technical implications.

“There are independent reports about it. You'd have to believe multiple companies, hundreds of researchers, etc. are all lying and conspiring together to turn a major hacking incident into a marketing exercise. metr.org/blog/2026-08...”

2 experts discussed this · 9 posts
Alejandra Caraballo: This mentality that an unmonitored AI agentic swarm hacking a company over several days and committing multiple felonies is somehow a marketing effort is absurd. Since when is "we lost control of o…
Alejandra Caraballo: Being skeptical or anti AI is a valid position but continuing to ignore the increasing capabilities of this tech is making people detached from reality. There's absolutely real danger here because …
Alejandra Caraballo: There needs to be a global moratorium on frontier research for at least a few months if not a year while safeguards and safety research catches up. The problem is that no one has that ability. The …
Open the full discussion →
Primary source metr.org ↗
New AI business Analysis 9h ago
⚡ 192 h early
students reject Gen AI conscientious objectors

Students as Conscientious Objectors to Gen AI (opinion)

10 experts across 4 network communities independently surfaced this.

Why this matches Concern & critique reaction 2 attributable expert contributions · Benjamin Riley, Jane Rosenzweig
“"Our students—and others in their AI-skeptical generation—are thoughtful moral agents who have used their capacity for critical reflection to arrive at compelling and deeply felt moral objections to using generative AI in their university work. We think edu…” evidence ↗
10 experts 4 communities 1 sources clustered
Iris van Rooij Olivia Guest Luke Stark Mél Hogan / The Data Fix Salomé Viljoen Benjamin Riley Ben Williamson Rosamunde Van Brakel +2
How the network is reacting 2 experts are emphasizing concern & critique.
Shared emphasis

Concern & critique

2 experts

Risks, limits and unintended consequences.

“"Our students—and others in their AI-skeptical generation—are thoughtful moral agents who have used their capacity for critical reflection to arrive at compelling and deeply felt moral objections to using generative AI in their university work. We think edu…”

2 experts discussed this · 5 posts
Iris van Rooij: "The class atmosphere became tense as more students pushed back against this required use of generative AI. They wanted to know what their classmates thought rather than what a large language model…
Iris van Rooij: “(...) they wanted to work together across their human differences to negotiate new ideas and find ways of expressing them. And they didn’t want to outsource any of this effort to generative AI. Th…
Olivia Guest: "The class atmosphere became tense as more students pushed back against this required use of generative AI. They wanted to know what their classmates thought rather than what a large language model…
Open the full discussion →
Primary source insidehighered.com ↗
More conversations

What experts are discussing without an anchoring article

7 experts · 29 posts · 1d ago
Mark Riedl Jeffrey P. Bigham Tim Kellogg Michael Ekstrand David Picard Lionel Singularity's Bounty e/cc
Matches Concern & critique
“Harnesses are lightweight. But the harness needs to call back to its model. It hasn’t really escaped then. One knows exactly where the model is and shut it down the harness or the model or block th…” evidence ↗
Multiple readings Concern & critique · 1 Building & implementation · 1 Research & technical analysis · 1
Mark Riedl: Do openAI and Anthropic employees not understand how computers work.
Mark Riedl: Any competent IT professional: Hey boss, what is this 8TB file doing on our server, and why is it executing continuously on 3TB of VRAM? Any competent IT manager: what ever it is, shut it down.
Open the thread →
5 experts · 41 posts · 1d ago
Tim Kellogg Dustin Moskovitz jake coty.bsky.social david-p-reichert.bsky.social
Matches Concern & critique
“i’m not confident that you can read it that way, because the labs are simultaneously reporting increased risk while also reportedly down-training or content filtering in those areas so we’re effect…” evidence ↗
Reaction converging Concern & critique · 4 Questions & unknowns · 1
Tim Kellogg: perspective. i’m not sure if i agree or disagree. but he’s thought about it, so..
jake: yeah, and @david-p-reichert.bsky.social thinks about it more than I do, professionally, so probably listen to him more than I (I do still wish The Professionals published lists of various guestimat…
Open the thread →
5 experts · 33 posts · 1d ago
joao Grace Colin coty.bsky.social david-p-reichert.bsky.social
Matches Concern & critique
“you are missing the point. The concern is not that it will cheat at chess; the concern is that it will commit genocide.” evidence ↗
Colin: There are indeed thousands of LessWrong posts, as well as papers and books, which make a particular theoretical case for why AI will kill us all that many people find convincing. I'm going to try t…
Colin: First of all, we must just agree off the bat that a superintelligent AI will be able to kill us all. After all it is superintelligent. As this random dude who became famous 3 days ago because of a …
Open the thread →
5 experts · 34 posts · 1d ago
jake Marco Z tachikoma Fran Litterio Singularity's Bounty e/cc
Matches Concern & critique
“a model would be operating on the periphery of the net and society, without even many of the affordances criminals have such as direct access to the physical world and the ability to cultivate trus…” evidence ↗
Reaction converging Research & technical analysis · 3
jake: People making fun of "where would a frontier model run if it were exfiltrated" seem to be missing the point that if a frontier model got out, it can: * rent 64 H200's on eg modal: gets you 9TB of n…
jake: it then rents out spare capacity, via the margins inference gets you, pretending to be some open-weight model, and gets to use the rest to think about how to increase p(doom) or whatever I don't se…
Open the thread →
5 experts · 10 posts · 1d ago
Mar Hicks Olivia Guest Damien Williams Marielza Dr. Johnathan Flowers, Bisexual of the Blade (Alt-text or DIE)
Matches Concern & critique
“Again, I am happy to stand with y’all, but I think we might take a moment to reflect on why the march of AI became so much more of an existential threat when it came for the practical fields, the s…” evidence ↗
Shared emphasis Concern & critique · 2
Dr. Johnathan Flowers, Bisexual of the Blade (Alt-text or DIE): Gonna be extremely petty for a moment: I am glad to stand in solidarity with my colleagues in mathematics against the tyranny of AI slop, but I have to ask: where the fuck were most of y’all when t…
Dr. Johnathan Flowers, Bisexual of the Blade (Alt-text or DIE): Again, I am happy to stand with y’all, but I think we might take a moment to reflect on why the march of AI became so much more of an existential threat when it came for the practical fields, the s…
Open the thread →
4 experts · 17 posts · 1d ago
Hagen Blix Colin tweety fish david-p-reichert.bsky.social
Matches Concern & critique
“I think the nuance about “escaping” is valuable. When your dog escapes you have to go physically catch him. When your agent escapes, well, it hasn’t really escaped, it’s still just right there wher…” evidence ↗
Multiple readings Concern & critique · 2 Questions & unknowns · 1
Colin: I think the nuance about “escaping” is valuable. When your dog escapes you have to go physically catch him. When your agent escapes, well, it hasn’t really escaped, it’s still just right there wher…
tweety fish: I can't stand the metaphors, sorta broadly, but I don't think that's why; I feel like they affirmatively make it harder for people who are not experts (plus some who are, I guess) to think about wh…
Open the thread →
4 experts · 11 posts · 1d ago
David Marx jake Amy Hoy Singularity's Bounty e/cc
Matches Concern & critique
“this is the exact same worry that people have been having in software engineering / computer science, now math, have been talking about. +what I talk to my boss (who runs a 15k-person organization)…” evidence ↗
Multiple readings Concern & critique · 1 Markets & investment · 1 Opportunity & adoption · 1
David Marx: Very forceful statement by Terry Tao about AI and math. terrytao.wordpress.com/2026/09/11/a... Note, he is not claiming the models aren't real and powerful. He's talking about how that power should…
jake: this is the exact same worry that people have been having in software engineering / computer science, now math, have been talking about. +what I talk to my boss (who runs a 15k-person organization)…
Open the thread →
4 experts · 10 posts · 1d ago
jake Colin Singularity's Bounty e/cc coty.bsky.social
Matches Concern & critique
“This and similar policies seem so obviously to be the right approach that I feel crazy reading almost everything on here re: alignment and safety.” evidence ↗
Multiple readings Building & implementation · 2 Concern & critique · 1
Colin: Perhaps the focus could be less on the “pace” of development and more on the direction of development. Like, make it illegal to develop cybercriminal hackerbots, at any pace.
Colin: I feel like the whole “pace” thing takes this teleological view of AI progress towards some known eventual state of “superintelligence” whereas the way models act is actually highly contingent on t…
Open the thread →
3 experts · 16 posts · 1d ago
VE, cybersocial occult investigator austin jeremymorrell.dev
Matches Concern & critique
“And if you set aside the market maximalism (and the other broad array of social and political harms), that should be a fine thing. It's not compatible with the way this is being sold as "the do eve…” evidence ↗
Multiple readings Building & implementation · 2 Questions & unknowns · 1
austin: some thoughts - there’s a fucking cambrian explosion of software but it’s invisible because it’s personal. github isn’t getting hugged every other week for nothin’
austin: - the software you see on your phone and in the world is the product of organizations, and organizations move s l o w l y. developers are significantly more productive but it doesn’t matter if that…
Open the thread →
3 experts · 13 posts · 1d ago
Luis Villa blair B! 🐝 Cavello (they/them)
Matches Concern & critique
“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…” evidence ↗
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 →