Sharon Goldman ✓

@sharongoldman · Metuchen, NJ · joined 30 Mar 2009

Founder, Ground Level AI. Previously covered AI at Fortune. Field reporting from the front lines of AI. https://t.co/YoXtonijWc Signal: sharongoldman.43

10 025Followers

3 349Following

10 638Posts total

2.3MViews on collected posts

Derniers posts

Sharon Goldman ✓@sharongoldman4 Sep

Back to the land, AGI baby! 484 views · 3 likes · 0 reposts · 0 replies Open on X →

Sharon Goldman ✓@sharongoldman4 Sep

@sharongoldman They were never separate, there was just no attack surface to argue about. An agent with tool access is a program with credentials and a network. That is a security problem with a model attached, and security teams already have the playbook. 54 views · 1 likes · 0 reposts · 0 replies Open on X →

Sharon Goldman ✓@sharongoldman4 Sep

Exactly a year ago I published a @FortuneMagazine feature story in which I interviewed @logangraham the leader of Anthropic’s Frontier Red Team, and Keane Lucas, one of its researchers. Nearly a year before Mythos and Fable, I described how Lucas took the stage at last year's 917 views · 10 likes · 1 reposts · 1 replies Open on X →

Sharon Goldman ✓@sharongoldman4 Sep

@sharongoldman AI labs should own the risk of their experiments when those experiments touch the real world: https://t.co/LRSGlnCu3F 156 views · 3 likes · 0 reposts · 0 replies Open on X →

Sharon Goldman ✓@sharongoldman4 Sep

There is more to this story, however. When I checked in on local residents six months after reporting on the ground from Meta's construction site in Richland Parish, Louisiana, I found that the success of the build has also led to rapidly rising rents, evictions, and thousands of 1.4K views · 10 likes · 4 reposts · 2 replies Open on X →

Sharon Goldman ✓@sharongoldman4 Sep

@sharongoldman this is exactly the split I’ve been thinking about. AI safety tends to start with the model, while security starts by assuming the component may fail or be compromised and asks what authority the surrounding system gives it. with agents, those two views are now des 33 views · 1 likes · 0 reposts · 0 replies Open on X →

Sharon Goldman ✓@sharongoldman4 Sep

AI safety and cybersecurity have long been two different worlds. In the age of AI agents, they’re colliding. And sometimes clashing. The OpenAI–Hugging Face incident has AI safety and cybersecurity experts arguing over what went wrong — and who has the better playbook for what 9.9K views · 83 likes · 23 reposts · 3 replies Open on X →

Sharon Goldman ✓@sharongoldman5 Aug

@sharongoldman Thanks for reporting this Sharon! Important! 19K views · 30 likes · 2 reposts · 1 replies Open on X →

Sharon Goldman ✓@sharongoldman5 Aug

@sharongoldman “AI agents accidentally created an internal message board…” What do you mean “accidentally?” 1.3K views · 29 likes · 0 reposts · 3 replies Open on X →

Sharon Goldman ✓@sharongoldman5 Aug

@sharongoldman Great stuff thank you for your work. 25.9K views · 113 likes · 1 reposts · 2 replies Open on X →

Sharon Goldman ✓@sharongoldman5 Aug

NEW: OpenAI gives first detailed debrief of the Hugging Face incident at Black Hat conference In a session I attended today at Black Hat, OpenAI's Eric Wallace and Michael Dalton said the company is "consciously slowing down research to enhance security" while a full technical 2.2M views · 1.8K likes · 333 reposts · 128 replies Open on X →

Face aux comptes de taille comparable

11 posts des 90 derniers jours, à côté de la tranche de 10K–100K abonnés. diffusé à plus de monde que les comptes de taille comparable.

Vues médianes1 325ce compte924médiane pour 10K–100K

Portée, %13.22%ce compte3.62%médiane pour 10K–100K

Engagement, %1.17%ce compte1.52%médiane pour 10K–100K

Indicateur	Ce compte	Médiane pour 10K–100K	Rapport
Vues médianes par post	1 325	924	1.43×
Portée (vues ÷ abonnés)	13.22%	3.62%	3.65×
Taux d'engagement	1.17%	1.52%	0.77×

Autres comptes de cette tranche → Comparer avec un autre compte → Comment ces repères sont établis →

Growth & engagement

How the posts we collected actually performed: views and reaction rate post by post, what the audience did with them, and where the follower count goes.

Views per post

2.2M5 Aug

25.9K

1.3K

19K

9.9K4 Sep

1.4K

156

917

484

Last 11 collected posts, oldest on the left. The scale is logarithmic: one post can outrun the rest a hundred times over.

Engagement rate per post

0.11%5 Aug

0.45%

2.42%

0.17%

1.13%4 Sep

3.03%

1.17%

1.92%

1.31%

1.85%

0.62%

Reactions — likes, reposts, replies and quotes — divided by views. Median for 10K–100K accounts is 1.52%.

What the audience does

Likes53.4%2 122 in total

Reposts9.2%364 in total

Replies3.5%140 in total

Quotes6.0%240 in total

Bookmarks27.9%1 108 in total

Share of every reaction we collected for this account. Replies mean argument, reposts mean endorsement, bookmarks mean the post was worth keeping.

Followers by day

5 Sep

Daily snapshots since 05 Sep 2026; the dashed line is the starting count.

Derniers posts

Face aux comptes de taille comparable

Growth & engagement

Views per post

Engagement rate per post

What the audience does

Followers by day

Comptes similaires