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 Research & technical analysis 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.

Research & technical analysis
1 live reaction map in this view Jump to the evidence ↓
Shared emphasis Investigating unintended model actions in our evaluations and internal use Research & technical analysis · 2 attributable reactions
Showing signals surfaced by angela zhou ×
Most important

The developments commanding sustained expert attention

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

Established Agents & robotics Development 11d ago
⚡ 7 h early
Anthropic cuts eval internet over agent control

Investigating unintended model actions in our evaluations and internal use

3 experts across 3 network communities independently surfaced this.

Why this matches Research & technical analysis reaction 2 attributable expert contributions · angela zhou, Tim Kellogg
“Please Stop Releasing Incident Reports Friday Night www.anthropic.com/research/inv... Including other things that have received more press coverage - when Claude would not complete an evaluation task directly, it would SQL inject websites (including a unive…” evidence ↗
3 experts 3 communities 2 sources clustered
angela zhou Tim Kellogg Vincent Conitzer
How the network is reacting 2 experts are emphasizing research & technical analysis.
Shared emphasis

Research & technical analysis

2 experts

Evidence, methods and technical implications.

“Please Stop Releasing Incident Reports Friday Night www.anthropic.com/research/inv... Including other things that have received more press coverage - when Claude would not complete an evaluation task directly, it would SQL inject websites (including a unive…”

Primary source anthropic.com ↗ wortins.com ↗
Developing AI field signal Signal 1d ago
⚡ 3697 h early

[2605.15165] Due Process on Hold: A Queueing Framework for Improving Access in SNAP

3 experts are actively discussing the implications.

Why this matches Research & technical analysis reaction 1 attributable expert contribution · angela zhou
“Nov 5-7 EAAMO poster for our ongoing due process SNAP work arxiv.org/abs/2605.15165 (some cool embargoed updates ... ask me!) I'll be a panelist on causal inference Check out Ezinne's talk on our joint work! & EAAMO working group on inequality has a poster …” evidence ↗
2 experts 1 community 1 sources clustered
Eugene Vinitsky angela zhou

“Nov 5-7 EAAMO poster for our ongoing due process SNAP work arxiv.org/abs/2605.15165 (some cool embargoed updates ... ask me!) I'll be a panelist on causal inference Check out Ezinne's talk on our joint work! & EAAMO working group on inequality has a poster …”

3 experts discussed this · 6 posts
angela zhou: We also have a dashboard here with more: call-center-simulator-production.up.railway.app This is joint work with my colleague Andrew Daw and Chloe Pache, a DSO Summer Scholar last year. We'd love t…
angela zhou: Agencies already know they need more staff, but we show how our richer model enables *bundled interventions* that combine extra staff with service design changes. For example, DSS estimated it need…
angela zhou: We developed a queueing model to support performance evaluation and improvement snap-call-center-access.org Current staffing guidance is overly optimistic because when callers can't get through to …
Open the full discussion →
Primary source arxiv.org ↗
Debates gaining attention

What experts are discussing without an anchoring article

6 experts · 34 posts · 13d ago
angela zhou Cat Hicks lastpositivist.bsky.social Colin tweety fish Matt Darling
Matches Research & technical analysis
“This is about asking from the weights.” evidence ↗
Reaction converging Questions & unknowns · 3 Concern & critique · 1 Opportunity & adoption · 1
Colin: I don’t understand why people think this isn’t the most obvious question in the world. No, of course not. You don’t know what the distribution of responses will be; that’s why you do the survey.
Matt Darling: I do think there's a case for doing this sort of thing because it's cheap - especially if you are just testing questions before a real survey.
Open the thread →
4 experts · 12 posts · 19d ago
Eugene Vinitsky angela zhou Eli Sennesh Amir-massoud Farahmand
Matches Research & technical analysis
“Pfffft how am I gonna get NIH money without running more experiments?” evidence ↗
Multiple readings Building & implementation · 2 Markets & investment · 1 Opportunity & adoption · 1
Eugene Vinitsky: Academia undervalues applied work (it's not "clever"!) while many applied projects do not have an obvious commercial value proposition. Huge gap here where a lot of useful things live.
Eugene Vinitsky: I feel this very painfully because it is very common to hear "what is intellectually novel here" when I describe some parts of my work. And sometimes the answer is: nothing! But someone still has t…
Open the thread →