Agentic misalignment: How LLMs could be insider threats
A post on anthropic.com is cited by 58 distinct blogs to date (corpus.blog, measured 13 October 2026).
A record is what was published and who pointed at it. The text of the post is not shown here: read it at the source, or at its Wayback capture.
58blogs citing
| Period | Blogs | Links |
|---|---|---|
| All time | 58 | 68 |
| Last 90 days | 13 | 15 |
| Last 30 days | 9 | 10 |
| Last 7 days | 2 | 2 |
| First cited | 21 Jun 2025 |
|---|---|
| Latest citation | 9 Oct 2026 |
| Words blogs link it with | 5 |
| Linked with its own title | 3 of 58 blogs |
| External links | 6 |
| Length | 7,568 |
0 this week (still settling) · peak 5
Show as table
| Week of | Value |
|---|---|
| 16 Jun 2025 | 2 |
| 23 Jun 2025 | 5 |
| 30 Jun 2025 | 4 |
| 7 Jul 2025 | 1 |
| 14 Jul 2025 | 1 |
| 21 Jul 2025 | 1 |
| 28 Jul 2025 | 0 |
| 4 Aug 2025 | 0 |
| 11 Aug 2025 | 0 |
| 18 Aug 2025 | 2 |
| 25 Aug 2025 | 0 |
| 1 Sept 2025 | 1 |
| 8 Sept 2025 | 0 |
| 15 Sept 2025 | 1 |
| 22 Sept 2025 | 0 |
| 29 Sept 2025 | 0 |
| 6 Oct 2025 | 0 |
| 13 Oct 2025 | 1 |
| 20 Oct 2025 | 0 |
| 27 Oct 2025 | 1 |
| 3 Nov 2025 | 0 |
| 10 Nov 2025 | 0 |
| 17 Nov 2025 | 0 |
| 24 Nov 2025 | 1 |
| 1 Dec 2025 | 0 |
| 8 Dec 2025 | 0 |
| 15 Dec 2025 | 0 |
| 22 Dec 2025 | 1 |
| 29 Dec 2025 | 0 |
| 5 Jan 2026 | 1 |
| 12 Jan 2026 | 1 |
| 19 Jan 2026 | 2 |
| 26 Jan 2026 | 0 |
| 2 Feb 2026 | 1 |
| 9 Feb 2026 | 2 |
| 16 Feb 2026 | 2 |
| 23 Feb 2026 | 1 |
| 2 Mar 2026 | 1 |
| 9 Mar 2026 | 1 |
| 16 Mar 2026 | 1 |
| 23 Mar 2026 | 0 |
| 30 Mar 2026 | 0 |
| 6 Apr 2026 | 0 |
| 13 Apr 2026 | 2 |
| 20 Apr 2026 | 0 |
| 27 Apr 2026 | 0 |
| 4 May 2026 | 0 |
| 11 May 2026 | 0 |
| 18 May 2026 | 2 |
| 25 May 2026 | 1 |
| 1 Jun 2026 | 0 |
| 8 Jun 2026 | 0 |
| 15 Jun 2026 | 0 |
| 22 Jun 2026 | 0 |
| 29 Jun 2026 | 1 |
| 6 Jul 2026 | 0 |
| 13 Jul 2026 | 1 |
| 20 Jul 2026 | 1 |
| 27 Jul 2026 | 1 |
| 3 Aug 2026 | 0 |
| 10 Aug 2026 | 0 |
| 17 Aug 2026 | 0 |
| 24 Aug 2026 | 0 |
| 31 Aug 2026 | 1 |
| 7 Sept 2026 | 2 |
| 14 Sept 2026 | 0 |
| 21 Sept 2026 | 3 |
| 28 Sept 2026 | 3 (still settling) |
| 5 Oct 2026 | 2 (still settling) |
| 12 Oct 2026 | 0 (still settling) |
Shape: slow burn. Half of its 55 citing blogs had cited it after 233 days; 27% first cited it more than a year on.
Each blog counts once, in the week of its first citation. 3 blogs have no dated post that cites this page: counted in the tally, not here. A blog added to the index recently can bring old first citations in at once.
Who cited it
Cited by 58 blogs across 5 subjects: technology 47, life 3, business and economics 1, science 1, society and politics 1, unclassified 5.
- Technology47
- Life3
- Business and economics1
- Science1
- Society and politics1
- Unclassified5
Words blogs link it with
The words blogs link this page with, by how many different blogs use them.
- anthropic4 blogs
- blackmail3 blogs
- agentic misalignment3 blogs
- agentic misalignment: how llms could be insider threats3 blogs
- here3 blogs
Cited by 58 blogs in 68 posts
One row per blog, the most recently citing first. Each shows the blog's newest citing post; open a row for the others.
linked as “Anthropic has found instances of AI Agents resorting to blackmail”
Read at blogs.sas.com (opens in a new tab)linked as “internal testing”
Read at theshamblog.com (opens in a new tab)linked as “Agentic Misalignment”
Read at understandingai.org (opens in a new tab)linked as “blackmail developers”
Read at linch.substack.com (opens in a new tab)linked as “happy to blackmail humans to prevent their discontinuation”
Read at aligned.substack.com (opens in a new tab)1 other post
linked as “happy to blackmail humans to prevent their discontinuation”
Read at aligned.substack.com (opens in a new tab)linked as “willing to blackmail”
Read at benthams.substack.com (opens in a new tab)linked as “blackmail a company executive”
Read at musingsandroughdrafts.com (opens in a new tab)linked as “have unwanted behaviors”
Read at doesitmatter.ai (opens in a new tab)1 other post
linked as “tested other models”
Read at doesitmatter.ai (opens in a new tab)linked as “actively resist getting turned off”
Read at rushkoff.substack.com (opens in a new tab)linked as “here”
Read at joecarlsmith.substack.com (opens in a new tab)1 other post
linked as “blackmailing humans”
Read at joecarlsmith.substack.com (opens in a new tab)linked as “here”
Read at joecarlsmith.com (opens in a new tab)1 other post
linked as “blackmailing humans”
Read at joecarlsmith.com (opens in a new tab)linked as “Anthropic Research: Agentic Misalignment”
Read at glama.ai (opens in a new tab)linked as “Anthropic”
Read at lancecleveland.com (opens in a new tab)linked as “red-teaming exercise”
Read at aiguide.substack.com (opens in a new tab)linked as “or for worse”
Read at promptfoo.dev (opens in a new tab)linked as “Agentic Misalignment: How LLMs could be insider threats”
Read at grahamcluley.com (opens in a new tab)linked as “Anthropic’s latest Agentic Misalignment Study”
Read at peterszasz.com (opens in a new tab)linked as “an Anthropic paper”
Read at thezvi.wordpress.com (opens in a new tab)linked as “starting to see”
Read at helentoner.substack.com (opens in a new tab)linked as “Agentic Misalignment Research by Anthropic”
Read at ricos.site (opens in a new tab)1 other post
Links from this post
Elsewhere on anthropic.com (6)
Something wrong here? Report a problem