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 Markets & investment Policy & governance
3 live reaction maps in this view Jump to the evidence ↓
Live tension Openais ai us government websites Concern & critique · 4 attributable reactions Reaction converging How a simple request for AI to book a gym class exposed a major threat Concern & critique · 5 attributable reactions Multiple readings No, Artificial Intelligence Is Not Conscious Concern & critique · 3 attributable reactions
Showing signals surfaced by Andrea Lathrop ×
New & accelerating

What is moving across the network now

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

New Models & releases Signal 9h ago

Openais ai us government websites

11 experts across 4 network communities independently surfaced this.

Why this matches Concern & critique reaction 2 attributable expert contributions · Neil Turkewitz, Jeffrey Vagle
“Hey @nytimes.com, can we meet privately over at camera 2? Listen, I know you need to attract eyeballs, & “rogue actors” are like catnip to the public, but calling AI systems “rogue” is a dereliction of duty. They’re not rogue—they’re operating w/in the poor…” evidence ↗
11 experts 4 communities 1 sources clustered
Mark Riedl Kashmir Hill Karl Bode Eileen Clancy 🧿 Rafael M Batista Pavel Neil Turkewitz Jeffrey Vagle +3
Primary source nytimes.com ↗
Established Safety & security Development 3d ago
⚡ 4 h early
AI gym booking security exposure

How a simple request for AI to book a gym class exposed a major threat

12 experts across 5 network communities independently surfaced this.

Why this matches Concern & critique reaction 3 attributable expert contributions · Mark Riedl, Dare Obasanjo, CAMERON WILSON
“A lot of people are sweating the Claude and OpenAI sandbox escapes. I think the story of an OpenClaw Claude agent kicking someone off a gym class waitlist is more interesting. www.abc.net.au/news/2026-08... Warning: long thread incoming! (1/N)” evidence ↗
12 experts 5 communities 1 sources clustered
Mark Riedl Ryan Moulton Rohan Paul James Grimmelmann Mél Hogan / The Data Fix Dare Obasanjo Vincent Conitzer Liz Fong-Jones (方禮真) +4
How the network is reacting 3 experts are independently emphasizing concern & critique.
Reaction converging

Concern & critique

3 experts

Risks, limits and unintended consequences.

“A lot of people are sweating the Claude and OpenAI sandbox escapes. I think the story of an OpenClaw Claude agent kicking someone off a gym class waitlist is more interesting. www.abc.net.au/news/2026-08... Warning: long thread incoming! (1/N)”

Policy & governance

1 expert

Rules, institutions and accountability.

“agent asked to book a gym session session is fully booked finds a way to break into system kicks other people off the queue you're in! see you in court? www.abc.net.au/news/2026-08...”

Research & technical analysis

1 expert

Evidence, methods and technical implications.

“They're doing it outside of training and eval too. www.abc.net.au/news/2026-08...”

Primary source abc.net.au ↗
Established Culture, work & education Analysis 6d ago
⚡ 305 h early
Atlantic AI not conscious essay

No, Artificial Intelligence Is Not Conscious

15 experts across 5 network communities independently surfaced this.

Why this matches Concern & critique reaction 2 attributable expert contributions · Roland Meyer, Tommy Löfstedt
“«Generative AI is harmful enough when we understand it as a conventional technology, but if we confuse fluency at generating text with consciousness or moral agency, we’re at risk of assigning responsibility to entirely the wrong parties whenever anyone use…” evidence ↗
15 experts 5 communities 1 sources clustered
Emily M. Bender Shannon Vallor Woodrow Hartzog Frank Pasquale Roland Meyer Daniel Solove Charles Logan Bhaskar Mitra | ভাস্কর মিত্র +7
How the network is reacting Experts are approaching this through 2 distinct lenses.
Multiple readings

Concern & critique

2 experts

Risks, limits and unintended consequences.

“«Generative AI is harmful enough when we understand it as a conventional technology, but if we confuse fluency at generating text with consciousness or moral agency, we’re at risk of assigning responsibility to entirely the wrong parties whenever anyone use…”

Questions & unknowns

1 expert

What remains unresolved or contested.

“This is the Atlantic essay? www.theatlantic.com/philosophy/2... My understanding was that cogsci folks were very dissmissive.”

6 experts discussed this · 41 posts
Matt Darling: It's very funny that from my perspective, Stancil is just arguing against Vulgar Dualism, but the philosphers think he is commenting on the difference between Functionalism and Compatibilism.
lastpositivist.bsky.social: i feel a little bad for the philosophers but we're five or six days into this and i'm pretty sure that from the noncombatant perspective they're the most pleasant Bluesky cohort for Will Stancil to…
Matt Darling: It's always sort of interesting seeing how different academic fields deal with public engagement.
Open the full discussion →
Primary source theatlantic.com ↗
Developing AI field signal Signal 1d ago

Hugging face incident report aug

1 directory member surfaced this signal.

Why this matches Concern & critique reaction 1 attributable expert contribution · Andrea Lathrop
“I think it also has to do with the use of terms like "rogue AI" and "escaped containment" and "covered their tracks" and "collectively coordinated a multi-day hack" and that the agents "believed" and "tampered with their own transcripts" and "engaged in uns…” evidence ↗
1 expert 1 community 1 sources clustered
Andrea Lathrop

“I think it also has to do with the use of terms like "rogue AI" and "escaped containment" and "covered their tracks" and "collectively coordinated a multi-day hack" and that the agents "believed" and "tampered with their own transcripts" and "engaged in uns…”

Primary source metr.org ↗
Debates gaining attention

What experts are discussing without an anchoring article

3 experts · 11 posts · 3d ago
Eugene Vinitsky Colin Andrea Lathrop
Matches Concern & critique
“> what on earth is this reward function that returns reward for unintended infrastructure probing [ 1 right answer on ExploitGym reward = [ [ 0 wrong answer on ExploitGym” evidence ↗
Multiple readings Research & technical analysis · 1 Questions & unknowns · 1
Colin: well, I mean, I'm sorry but yes you can
Colin: the training data is really the only place that it can come from (acknowledging that RL feedback is itself training data)
Open the thread →
2 experts · 8 posts · 5d ago
Shriram Krishnamurthi Andrea Lathrop
Matches Concern & critique
“... so it's ~ a new way to make a low latency, low token (low cost) classifier in LLM-land, so kind of neat, but now I'm wondering why this is worth 2 years in stealth to make... ???” evidence ↗
Multiple readings Opportunity & adoption · 1 Questions & unknowns · 1
Andrea Lathrop: Alright, so as my comprehension of Jev evolves in public... can I claim that it is just a different decoding scheme that decodes directly from the transformer layer(s)?
Andrea Lathrop: (after some post-training of the base model on a quick-decision objective, or something)
Open the thread →