Logan Graham ✓

@logangraham · the present, moments ago · joined 14 Jun 2009

Head of the Frontier Red Team @anthropicai. 🌎 Make things radically good.

22 466Followers

8 631Following

1 340Posts total

346.9KViews on collected posts

Последние посты

Logan Graham ✓@logangraham18 Aug

Last week, we published a first look into our new research on multiple agents. Consider: if you naively extrapolate AI revenues, within 2-3 years it's possible some % of global GDP could be agent-agent interactions. We want to know how that could fail. In the past ~month, the 35.9K views · 247 likes · 25 reposts · 18 replies Open on X →

Logan Graham ✓@logangraham22 Jul

Yesterday, as we huddled around our computers reading the report, I told the team to "remember this moment" as the first true AI safety incident. Pay attention to the trend! Major kudos to @OpenAI for sharing this and working with @huggingface to remediate. 106.6K views · 870 likes · 45 reposts · 57 replies Open on X →

Logan Graham ✓@logangraham16 Jul

Excited to finally be able to tweet this: we're looking for a world class visionary to lead @AnthropicAI's cyberdefense research mission. Leading Glasswing and our cyberdefense mission this year has been an honor. It's now clear we need to scale up our ambition to help defend ht 85.5K views · 475 likes · 49 reposts · 31 replies Open on X →

Logan Graham ✓@logangraham9 Jun

Fable 5 is the same underlying model as Mythos 5, but with cybersecurity and biology blocks. Mythos is the first model that's made me feel that we've entered the next phase of model progress. For years, we've talked about cybersecurity / self-improvement / autonomy / https://t.c 33.5K views · 187 likes · 17 reposts · 19 replies Open on X →

Logan Graham ✓@logangraham9 Jun

New post on Red today: Our team @AnthropicAI found that Mythos Preview is meaningfully better at developing N-days. It took us a couple thousand $ and a few hours to convert patches into exploits. We publish research like this because we think it's important the world knows ht 85.4K views · 659 likes · 62 reposts · 30 replies Open on X →

На фоне аккаунтов своего размера

1 постов за последние 90 дней рядом с диапазоном 10K–100K подписчиков. показывается широко, но откликается мало кто.

Медианные просмотры35 870этот аккаунт924медиана для 10K–100K

Охват, %159.66%этот аккаунт3.62%медиана для 10K–100K

Вовлечённость, %0.83%этот аккаунт1.52%медиана для 10K–100K

Показатель	Этот аккаунт	Медиана для 10K–100K	Отношение
Медианные просмотры на пост	35 870	924	38.8×
Охват (просмотры ÷ подписчики)	159.66%	3.62%	44.1×
Вовлечённость	0.83%	1.52%	0.55×

Другие в этом диапазоне → Сравнить с другим аккаунтом → Как считаются эти ориентиры →

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

85.4K9 Jun

33.5K

85.5K16 Jul

106.6K22 Jul

35.9K18 Aug

Last 5 collected posts, oldest on the left. The scale is logarithmic: one post can outrun the rest a hundred times over.

Engagement rate per post

0.89%9 Jun

0.68%

0.66%16 Jul

0.92%22 Jul

0.83%18 Aug

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

What the audience does

Likes71.0%2 438 in total

Reposts5.8%198 in total

Replies4.5%155 in total

Quotes1.3%46 in total

Bookmarks17.4%596 in total

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

The follower curve appears once this account has two daily snapshots — we take one a day, and this one is on its first.