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.

2,364 searchable experts 2,966 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.

Concern & critique Research & technical analysis
3 live reaction maps in this view Jump to the evidence ↓
Reaction converging Its ai agent spent days hacking company sources say openai did not notice week Concern & critique · 5 attributable reactions Reaction converging Anatomy of a Frontier Lab Agent Intrusion: A Technical Timeline of the July 2026 Incident Concern & critique · 4 attributable reactions Reaction converging arxiv.org Research & technical analysis · 4 attributable reactions
Showing signals surfaced by Andrea Lathrop ×
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.

Developing Agents & robotics Signal 2d ago
⚡ 14 h early

Its ai agent spent days hacking company sources say openai did not notice week

14 experts across 4 network communities independently surfaced this.

14 experts 4 communities 1 sources clustered
Casey Newton Justin Hendrix Eryk Salvaggio Tim Kellogg Chris Paxton Shannon Mattern Grace Eileen Clancy 🧿 +6
3 experts discussed this · 38 posts
Casey Newton: Rogue OpenAI agents are leaving notes to themselves on company servers to help them escape their test environments (!!!) www.reuters.com/business/its...
Casey Newton: my sense is that they were leaving the notes surreptitiously
Casey Newton: you deny that models are taking steps to hide their behavior during evaluations? because that's pretty well established
Open the full discussion →
Primary source reuters.com ↗
New AI field signal Signal 17h ago

Anatomy of a Frontier Lab Agent Intrusion: A Technical Timeline of the July 2026 Incident

9 experts across 5 network communities independently surfaced this.

9 experts 5 communities 1 sources clustered
Margaret Mitchell Mark Riedl Simon Willison Alondra Nelson Suresh Venkatasubramanian Eileen Clancy 🧿 Gina Helfrich René Walter +1
3 experts discussed this · 17 posts
Simon Willison: Hugging Face just published a highly detailed technical account of OpenAI's accidental cyberattack on their systems - it's wild how sophisticated this was: huggingface.co/blog/agent-i... Wrote up s…
Alondra Nelson: Hugging Face just published a highly detailed technical account of OpenAI's accidental cyberattack on their systems - it's wild how sophisticated this was: huggingface.co/blog/agent-i... Wrote up s…
Grace: I think brute force is a reasonable description, in the sense that the agent tried a lot of approaches, most of which failed
Open the full discussion →
Primary source huggingface.co ↗
Established Models & releases Signal 17d ago

www-cdn.anthropic.com

5 experts are actively discussing the implications.

2 experts 1 community 1 sources clustered
Andrea Lathrop david-p-reichert.bsky.social

“😜 (From the Eleos group's response, here:) www-cdn.anthropic.com/files/4zrzov...”

“How so? They don't seem to mind: www-cdn.anthropic.com/files/4zrzov...”

5 experts discussed this · 14 posts
Grace: Condolences to GWT theorists who just learned that their model of consciousness can’t be correct
david-p-reichert.bsky.social: How so? They don't seem to mind: www-cdn.anthropic.com/files/4zrzov...
Grace: For the same reason that AI isn't considered "real" intelligence - people will take LLM non-consciousness as axiomatic and move the goal posts accordingly en.wikipedia.org/wiki/AI_effect
Open the full discussion →
Primary source www-cdn.anthropic.com ↗
Established AI field signal Signal 10d ago
⚡ 232 h early

arxiv.org

5 experts across 2 network communities independently surfaced this.

5 experts 2 communities 1 sources clustered
Aram Sinnreich Pavel ae Andrea Lathrop Matthew Sheffield
2 experts discussed this · 3 posts
ae: Mad lad trains neural network in Age of Empires II arxiv.org/pdf/2605.31514 @surcomplicated.bsky.social
ae: this is funny "Lemma 1 (NAND Gates in AoE II). It is possible to build and operate a NAND gate in the scenario editor by using one player, two units acting as the {0, 1} bits, and no other components"
Nafnlaus 🇮🇸 🇺🇦: Also, the Town Square can speak Chinese.
Open the full discussion →
Primary source arxiv.org ↗
Established Models & releases Signal 5d ago
⚡ 120 h early

How OpenAI Lost Control of an AI Model—and What Needs to Change

2 experts are actively discussing the implications.

1 expert 1 community 1 sources clustered
Andrea Lathrop

“... found the full TIME article: time.com/article/2026...”

2 experts discussed this · 6 posts
Andrea Lathrop: I am starting to get the sense that "sandbox" (you keep using word... I do not think it means what you think it means...) (i.e. are they just putting sandbox-shaped instructions in the context wind…
Andrea Lathrop: ... found the full TIME article: time.com/article/2026...
Andrea Lathrop: THIS!
Open the full discussion →
Primary source time.com ↗
More conversations

What experts are discussing without an anchoring article

4 experts · 11 posts · 3d ago
Ted Underwood ae Doll Andrea Lathrop
Doll: The vision of AGI/ASI that lab insiders tend to believe in is bugnuts insane though. Like its pure fantasy cooked up by yud and lesswrong and its unfalsifiable. Not very rational at all :p
Doll: It believes that most of them genunely believe it. It also believes they are in a cult.
Open the thread →
4 experts · 18 posts · 16d ago
Ted Underwood Hagen Blix Fran Litterio Andrea Lathrop
Multiple readings Questions & unknowns · 1 Context & explanation · 1
Ted Underwood: when people say "the problem is that these things don't 'think' at all in the way we know actual human beings think" I often want to respond, "yes, and that's a 'problem' because?"
Ted Underwood: Sometimes LLM solutions are brittle. But in that case the problem is — that they're brittle. You have to motivate the reference point.
Open the thread →
3 experts · 5 posts · 1d ago
SE Gyges Isaiah Bishop Andrea Lathrop
SE Gyges: i am actually pretty sympathetic to this; "we personally are clearly not able to figure out how to mitigate the impact of what we're doing" is obviously true the problem is they somehow reach the o…
Isaiah Bishop: it would be more believable if they took more of this stuff seriously at say OpenAI
Open the thread →
2 experts · 6 posts · 5d ago
Juan Diego Rodriguez Andrea Lathrop
Andrea Lathrop: Has OpenAI released enough detail to know what *actually* happened? News outlets keep describing it like it's a willful entity with volition that cleverly, intentionally escaped its cage, which is …
Andrea Lathrop: I am not implying it was intentional on OpenAI's part... what I mean is 1. what was the actual task, 2. what had they optimized or fine-tuned each of the two running models to do, 3. what *specific…
Open the thread →