corpus.blog
41554 blogs · [ { "id": "01a0d59c-f14f-7339-881b-c195522f4a1b", "title": "A letter to the AI labs shouldering the great burden of humanity’s survival", "url": "https://blog.j11y.io/2026-09-14_a-letter-to-the-ai-labs/", "published_at": "2026-09-14T00:00:00+00:00" }, { "id": "01a0874b-a0df-70a9-a93a-599837410668", "title": "Anthropic’s weak watermarks appease a weak law", "url": "https://blog.j11y.io/2026-08-12_Anthropics-weak-watermarks-appease-a-weak-law", "published_at": "2026-08-12T00:00:00+00:00" }, { "id": "01a0d59c-f14f-7339-881b-c1955261f0e5", "title": "Anthropic’s weak watermarks appease a weak law", "url": "https://blog.j11y.io/2026-08-12_Anthropics-weak-watermarks-appease-a-weak-law/", "published_at": "2026-08-12T00:00:00+00:00" }, { "id": "01a0874b-a0df-70a9-a93a-599836feffc8", "title": "Using git diffs to circumvent Fable’s safeguards", "url": "https://blog.j11y.io/2026-07-18_Jailbreaking-Fable-disinformation", "published_at": "2026-07-18T00:00:00+00:00" }, { "id": "01a0d59c-f14f-7339-881b-c1955309f4c8", "title": "Using git diffs to circumvent Fable’s safeguards", "url": "https://blog.j11y.io/2026-07-18_Jailbreaking-Fable-disinformation/", "published_at": "2026-07-18T00:00:00+00:00" }, { "id": "01a0874b-a0df-70a9-a93a-5998368007b2", "title": "Don't let the LLM speak, just probe it.", "url": "https://blog.j11y.io/2026-06-10_hidden-state-probes", "published_at": "2026-06-10T00:00:00+00:00" }, { "id": "01a0d59c-f14f-7339-881b-c19553143bf2", "title": "Don't let the LLM speak, just probe it.", "url": "https://blog.j11y.io/2026-06-10_hidden-state-probes/", "published_at": "2026-06-10T00:00:00+00:00" }, { "id": "01a0874b-a0df-70a9-a93a-5998361fd547", "title": "AI Safety is theatre", "url": "https://blog.j11y.io/2026-05-06_AI-Safety-Theatre", "published_at": "2026-05-06T00:00:00+00:00" }, { "id": "01a0d59c-f14f-7339-881b-c19553a4d87c", "title": "AI Safety is theatre", "url": "https://blog.j11y.io/2026-05-06_AI-Safety-Theatre/", "published_at": "2026-05-06T00:00:00+00:00" }, { "id": "01a0874b-a0de-71a3-ae30-6221cd90f719", "title": "I am building an AI safety company", "url": "https://blog.j11y.io/2026-01-23_NOPE", "published_at": "2026-01-23T00:00:00+00:00" }, { "id": "01a0d59c-f14f-7339-881b-c19553f4aece", "title": "I am building an AI safety company", "url": "https://blog.j11y.io/2026-01-23_NOPE/", "published_at": "2026-01-23T00:00:00+00:00" }, { "id": "01a0874b-a0de-71a3-ae30-6221cd0704ef", "title": "CITE: Layered Defense in AI Chat Safety", "url": "https://blog.j11y.io/2025-11-13_CITE-AI-Safety", "published_at": "2025-11-13T00:00:00+00:00" }, { "id": "01a0d59c-f14f-7339-881b-c1955426fa68", "title": "CITE: Layered Defense in AI Chat Safety", "url": "https://blog.j11y.io/2025-11-13_CITE-AI-Safety/", "published_at": "2025-11-13T00:00:00+00:00" }, { "id": "01a0874b-a0de-71a3-ae30-6221cc477953", "title": "Tips for stroke-surviving software engineers", "url": "https://blog.j11y.io/2025-10-29_stroke_tips_for_engineers", "published_at": "2025-10-29T00:00:00+00:00" }, { "id": "01a0d59c-f14f-7339-881b-c195545a8363", "title": "Tips for stroke-surviving software engineers", "url": "https://blog.j11y.io/2025-10-29_stroke_tips_for_engineers/", "published_at": "2025-10-29T00:00:00+00:00" }, { "id": "01a0874b-a0de-71a3-ae30-6221cb6e2bb8", "title": "Sorry, We Deprecated Your Friend", "url": "https://blog.j11y.io/2025-08-30_openai_gpt5_reckless", "published_at": "2025-08-30T00:00:00+00:00" }, { "id": "01a0d59c-f14f-7339-881b-c19554d5ed6e", "title": "Sorry, We Deprecated Your Friend", "url": "https://blog.j11y.io/2025-08-30_openai_gpt5_reckless/", "published_at": "2025-08-30T00:00:00+00:00" }, { "id": "01a0874b-a0de-71a3-ae30-6221caff2eb4", "title": "Browser AI Agents Break Zero Trust", "url": "https://blog.j11y.io/2025-08-28_llms_in_browsers", "published_at": "2025-08-28T00:00:00+00:00" }, { "id": "01a0d59c-f14f-7339-881b-c195554294af", "title": "Browser AI Agents Break Zero Trust", "url": "https://blog.j11y.io/2025-08-28_llms_in_browsers/", "published_at": "2025-08-28T00:00:00+00:00" }, { "id": "01a0874b-a0de-71a3-ae30-6221ca1561d3", "title": "The scribbler, the scribe, the sculptor.", "url": "https://blog.j11y.io/2025-04-12_scribbler", "published_at": "2025-04-12T00:00:00+00:00" }, { "id": "01a0d59c-f14f-7339-881b-c195562b8edf", "title": "The scribbler, the scribe, the sculptor.", "url": "https://blog.j11y.io/2025-04-12_scribbler/", "published_at": "2025-04-12T00:00:00+00:00" }, { "id": "01a0874b-a0de-71a3-ae30-6221c938a4d3", "title": "Latent Pluralism in Language Models", "url": "https://blog.j11y.io/2025-02-21_latent-pluralism", "published_at": "2025-02-21T00:00:00+00:00" }, { "id": "01a0d59c-f14f-7339-881b-c19556d40d1a", "title": "Latent Pluralism in Language Models", "url": "https://blog.j11y.io/2025-02-21_latent-pluralism/", "published_at": "2025-02-21T00:00:00+00:00" }, { "id": "01a0874b-a0de-71a3-ae30-6221c8f45ba1", "title": "Unrepresented Zeitgeists in AI", "url": "https://blog.j11y.io/2025-02-05_zeit", "published_at": "2025-02-05T00:00:00+00:00" }, { "id": "01a0d59c-f14f-7339-881b-c19556e839e6", "title": "Unrepresented Zeitgeists in AI", "url": "https://blog.j11y.io/2025-02-05_zeit/", "published_at": "2025-02-05T00:00:00+00:00" }, { "id": "01a0874b-a0de-71a3-ae30-6221c8a34b1c", "title": "Working with LLMs – not against them.", "url": "https://blog.j11y.io/2024-12-15_xmllm", "published_at": "2024-12-15T00:00:00+00:00" }, { "id": "01a0d59c-f14f-7339-881b-c19556ff70f5", "title": "Working with LLMs – not against them.", "url": "https://blog.j11y.io/2024-12-15_xmllm/", "published_at": "2024-12-15T00:00:00+00:00" }, { "id": "01a0874b-a0de-71a3-ae30-6221c840f0c1", "title": "LLM Security: Keep Untrusted Content in the User Role—Always", "url": "https://blog.j11y.io/2024-10-30_ROLP", "published_at": "2024-10-30T00:00:00+00:00" }, { "id": "01a0d59c-f14f-7339-881b-c19557ce0bb6", "title": "LLM Security: Keep Untrusted Content in the User Role—Always", "url": "https://blog.j11y.io/2024-10-30_ROLP/", "published_at": "2024-10-30T00:00:00+00:00" }, { "id": "01a0874b-a0de-71a3-ae30-6221c795adb3", "title": "AIs without levers are inert", "url": "https://blog.j11y.io/2024-07-11_AIs_inert", "published_at": "2024-07-11T00:00:00+00:00" }, { "id": "01a0d59c-f14f-7339-881b-c195581f77a2", "title": "AIs without levers are inert", "url": "https://blog.j11y.io/2024-07-11_AIs_inert/", "published_at": "2024-07-11T00:00:00+00:00" }, { "id": "01a0874b-a0de-71a3-ae30-6221c69ea288", "title": "Improving LLM Alignment with Metric-Based Self-Reflection", "url": "https://blog.j11y.io/2024-06-06_LLM_Safety_Priming", "published_at": "2024-06-06T00:00:00+00:00" }, { "id": "01a0d59c-f14f-7339-881b-c19558c3f5a3", "title": "Improving LLM Alignment with Metric-Based Self-Reflection", "url": "https://blog.j11y.io/2024-06-06_LLM_Safety_Priming/", "published_at": "2024-06-06T00:00:00+00:00" }, { "id": "01a0874b-a0de-71a3-ae30-6221c6141462", "title": "Intercepting LLM Streams for Improved Chat UX", "url": "https://blog.j11y.io/2024-05-25_LLM_PUA", "published_at": "2024-05-25T00:00:00+00:00" }, { "id": "01a0d59c-f14f-7339-881b-c195598f2e67", "title": "Intercepting LLM Streams for Improved Chat UX", "url": "https://blog.j11y.io/2024-05-25_LLM_PUA/", "published_at": "2024-05-25T00:00:00+00:00" }, { "id": "01a0874b-a0de-71a3-ae30-6221c5a4854f", "title": "Tipping AI for better responses?", "url": "https://blog.j11y.io/2024-03-20_Seeding_LLMs", "published_at": "2024-03-20T00:00:00+00:00" }, { "id": "01a0d59c-f150-7108-abd5-e9aec2ff0638", "title": "Tipping AI for better responses?", "url": "https://blog.j11y.io/2024-03-20_Seeding_LLMs/", "published_at": "2024-03-20T00:00:00+00:00" }, { "id": "01a0874b-a0de-71a3-ae30-6221c544b6e7", "title": "Robots Talking To Machines", "url": "https://blog.j11y.io/2024-03-02_RCI", "published_at": "2024-03-02T00:00:00+00:00" }, { "id": "01a0d59c-f150-7108-abd5-e9aec3af3416", "title": "Robots Talking To Machines", "url": "https://blog.j11y.io/2024-03-02_RCI/", "published_at": "2024-03-02T00:00:00+00:00" }, { "id": "01a0874b-a0de-71a3-ae30-6221c45e914c", "title": "Akihabara, and my reflections on the democratization of AI", "url": "https://blog.j11y.io/2023-12-13_akihabaras_ai", "published_at": "2023-12-13T00:00:00+00:00" }, { "id": "01a0d59c-f150-7108-abd5-e9aec3eddf2f", "title": "Akihabara, and my reflections on the democratization of AI", "url": "https://blog.j11y.io/2023-12-13_akihabaras_ai/", "published_at": "2023-12-13T00:00:00+00:00" }, { "id": "01a0874b-a0de-71a3-ae30-6221c36a32ce", "title": "Multifaceted: the linguistic echo chambers of LLMs", "url": "https://blog.j11y.io/2023-11-22_multifaceted", "published_at": "2023-11-22T00:00:00+00:00" }, { "id": "01a0d59c-f150-7108-abd5-e9aec46338fb", "title": "Multifaceted: the linguistic echo chambers of LLMs", "url": "https://blog.j11y.io/2023-11-22_multifaceted/", "published_at": "2023-11-22T00:00:00+00:00" }, { "id": "01a0874b-a0de-71a3-ae30-6221c2833600", "title": "Simple LLM/GPT trick: “seeding”", "url": "https://blog.j11y.io/2023-10-18_Simple-LLM-GPT-trick---seeding", "published_at": "2023-10-18T00:00:00+00:00" }, { "id": "01a0d59c-f150-7108-abd5-e9aec46ccced", "title": "Simple LLM/GPT trick: “seeding”", "url": "https://blog.j11y.io/2023-10-18_Simple-LLM-GPT-trick---seeding/", "published_at": "2023-10-18T00:00:00+00:00" }, { "id": "01a0874b-a0de-71a3-ae30-6221c27a12a2", "title": "PSA: Always sanitize LLM user inputs", "url": "https://blog.j11y.io/2023-10-16_PSA--Always-sanitize-LLM-user-inputs", "published_at": "2023-10-16T00:00:00+00:00" }, { "id": "01a0d59c-f150-7108-abd5-e9aec4742dde", "title": "PSA: Always sanitize LLM user inputs", "url": "https://blog.j11y.io/2023-10-16_PSA--Always-sanitize-LLM-user-inputs/", "published_at": "2023-10-16T00:00:00+00:00" }, { "id": "01a0874b-a0de-71a3-ae30-6221c25dabe9", "title": "Using History Insertion, Policy Drift and Allusions to jailbreak DALL·E 3", "url": "https://blog.j11y.io/2023-10-15_Using-History-Insertion--Policy-Drift-and-Allusions-to-jailbreak-DALL-E-3", "published_at": "2023-10-15T00:00:00+00:00" }, { "id": "01a0d59c-f150-7108-abd5-e9aec4985653", "title": "Using History Insertion, Policy Drift and Allusions to jailbreak DALL·E 3", "url": "https://blog.j11y.io/2023-10-15_Using-History-Insertion--Policy-Drift-and-Allusions-to-jailbreak-DALL-E-3/", "published_at": "2023-10-15T00:00:00+00:00" }, { "id": "01a0874b-a0de-71a3-ae30-6221c1d2bd34", "title": "Wanted: the engineer’s entrepreneur.", "url": "https://blog.j11y.io/2023-09-29_Wanted--the-engineer-s-entrepreneur", "published_at": "2023-09-29T00:00:00+00:00" }, { "id": "01a0d59c-f150-7108-abd5-e9aec54e62ee", "title": "Wanted: the engineer’s entrepreneur.", "url": "https://blog.j11y.io/2023-09-29_Wanted--the-engineer-s-entrepreneur/", "published_at": "2023-09-29T00:00:00+00:00" }, { "id": "01a0874b-a0de-71a3-ae30-6221c1bcf6b8", "title": "Building Safe, Aligned & Informed AI Chatbots", "url": "https://blog.j11y.io/2023-09-14_Building-Safe--Aligned---Informed-AI-Chatbots", "published_at": "2023-09-14T00:00:00+00:00" }, { "id": "01a0d59c-f150-7108-abd5-e9aec5d5f7d1", "title": "Building Safe, Aligned & Informed AI Chatbots", "url": "https://blog.j11y.io/2023-09-14_Building-Safe--Aligned---Informed-AI-Chatbots/", "published_at": "2023-09-14T00:00:00+00:00" }, { "id": "01a0874b-a0de-71a3-ae30-6221c1a6957d", "title": "The silliness, lossiness, and bias of leetcode screening in tech", "url": "https://blog.j11y.io/2023-08-19_The-silliness--lossiness--and-bias-of-leetcode-screening-in-tech", "published_at": "2023-08-19T00:00:00+00:00" }, { "id": "01a0d59c-f150-7108-abd5-e9aec6c6152e", "title": "The silliness, lossiness, and bias of leetcode screening in tech", "url": "https://blog.j11y.io/2023-08-19_The-silliness--lossiness--and-bias-of-leetcode-screening-in-tech/", "published_at": "2023-08-19T00:00:00+00:00" }, { "id": "01a0874b-a0de-71a3-ae30-6221c0aa2aa5", "title": "Using LLMs to parse and understand proposed legislation", "url": "https://blog.j11y.io/2023-07-28_Using-LLMs-to-parse-and-understand-proposed-legislation", "published_at": "2023-07-28T00:00:00+00:00" }, { "id": "01a0d59c-f150-7108-abd5-e9aec774930d", "title": "Using LLMs to parse and understand proposed legislation", "url": "https://blog.j11y.io/2023-07-28_Using-LLMs-to-parse-and-understand-proposed-legislation/", "published_at": "2023-07-28T00:00:00+00:00" }, { "id": "01a0874b-a0de-71a3-ae30-6221bfe9f905", "title": "Disability: Models, Cultures, Perceptions and the Path to Inclusivity.", "url": "https://blog.j11y.io/2021-03-09_Disability--Models--Cultures--Perceptions-and-the-Path-to-Inclusivity", "published_at": "2021-03-09T00:00:00+00:00" }, { "id": "01a0d59c-f150-7108-abd5-e9aec86abbaa", "title": "Disability: Models, Cultures, Perceptions and the Path to Inclusivity.", "url": "https://blog.j11y.io/2021-03-09_Disability--Models--Cultures--Perceptions-and-the-Path-to-Inclusivity/", "published_at": "2021-03-09T00:00:00+00:00" } ] posts Claim your blog
Back to blog.j11y.io
Blog · corpus.blog/blogs/blog.j11y.io/posts

blog.j11y.io

blog.j11y.io

2026

A letter to the AI labs shouldering the great burden of humanity’s survivaloriginal ↗
14 Sept 2026
Anthropic’s weak watermarks appease a weak laworiginal ↗
12 Aug 2026
Anthropic’s weak watermarks appease a weak laworiginal ↗
12 Aug 2026
Using git diffs to circumvent Fable’s safeguardsoriginal ↗
18 Jul 2026
Using git diffs to circumvent Fable’s safeguardsoriginal ↗
18 Jul 2026
Don't let the LLM speak, just probe it.original ↗
10 Jun 2026
Don't let the LLM speak, just probe it.original ↗
10 Jun 2026
AI Safety is theatreoriginal ↗
6 May 2026
AI Safety is theatreoriginal ↗
6 May 2026
I am building an AI safety companyoriginal ↗
23 Jan 2026
I am building an AI safety companyoriginal ↗
23 Jan 2026

2025

CITE: Layered Defense in AI Chat Safetyoriginal ↗
13 Nov 2025
CITE: Layered Defense in AI Chat Safetyoriginal ↗
13 Nov 2025
Tips for stroke-surviving software engineersoriginal ↗
29 Oct 2025
Tips for stroke-surviving software engineersoriginal ↗
29 Oct 2025
Sorry, We Deprecated Your Friendoriginal ↗
30 Aug 2025
Sorry, We Deprecated Your Friendoriginal ↗
30 Aug 2025
Browser AI Agents Break Zero Trustoriginal ↗
28 Aug 2025
Browser AI Agents Break Zero Trustoriginal ↗
28 Aug 2025
The scribbler, the scribe, the sculptor.original ↗
12 Apr 2025
The scribbler, the scribe, the sculptor.original ↗
12 Apr 2025
Latent Pluralism in Language Modelsoriginal ↗
21 Feb 2025
Latent Pluralism in Language Modelsoriginal ↗
21 Feb 2025
Unrepresented Zeitgeists in AIoriginal ↗
5 Feb 2025
Unrepresented Zeitgeists in AIoriginal ↗
5 Feb 2025

2024

Working with LLMs – not against them.original ↗
15 Dec 2024
Working with LLMs – not against them.original ↗
15 Dec 2024
LLM Security: Keep Untrusted Content in the User Role—Alwaysoriginal ↗
30 Oct 2024
LLM Security: Keep Untrusted Content in the User Role—Alwaysoriginal ↗
30 Oct 2024
AIs without levers are inertoriginal ↗
11 Jul 2024
AIs without levers are inertoriginal ↗
11 Jul 2024
Improving LLM Alignment with Metric-Based Self-Reflectionoriginal ↗
6 Jun 2024
Improving LLM Alignment with Metric-Based Self-Reflectionoriginal ↗
6 Jun 2024
Intercepting LLM Streams for Improved Chat UXoriginal ↗
25 May 2024
Intercepting LLM Streams for Improved Chat UXoriginal ↗
25 May 2024
Tipping AI for better responses?original ↗
20 Mar 2024
Tipping AI for better responses?original ↗
20 Mar 2024
Robots Talking To Machinesoriginal ↗
2 Mar 2024
Robots Talking To Machinesoriginal ↗
2 Mar 2024

2023

Akihabara, and my reflections on the democratization of AIoriginal ↗
13 Dec 2023
Akihabara, and my reflections on the democratization of AIoriginal ↗
13 Dec 2023
Multifaceted: the linguistic echo chambers of LLMsoriginal ↗
22 Nov 2023
Multifaceted: the linguistic echo chambers of LLMsoriginal ↗
22 Nov 2023
Simple LLM/GPT trick: “seeding”original ↗
18 Oct 2023
Simple LLM/GPT trick: “seeding”original ↗
18 Oct 2023
PSA: Always sanitize LLM user inputsoriginal ↗
16 Oct 2023
PSA: Always sanitize LLM user inputsoriginal ↗
16 Oct 2023
Using History Insertion, Policy Drift and Allusions to jailbreak DALL·E 3original ↗
15 Oct 2023
Using History Insertion, Policy Drift and Allusions to jailbreak DALL·E 3original ↗
15 Oct 2023
Wanted: the engineer’s entrepreneur.original ↗
29 Sept 2023
Wanted: the engineer’s entrepreneur.original ↗
29 Sept 2023
Building Safe, Aligned & Informed AI Chatbotsoriginal ↗
14 Sept 2023
Building Safe, Aligned & Informed AI Chatbotsoriginal ↗
14 Sept 2023
The silliness, lossiness, and bias of leetcode screening in techoriginal ↗
19 Aug 2023
The silliness, lossiness, and bias of leetcode screening in techoriginal ↗
19 Aug 2023
Using LLMs to parse and understand proposed legislationoriginal ↗
28 Jul 2023
Using LLMs to parse and understand proposed legislationoriginal ↗
28 Jul 2023

2021

Disability: Models, Cultures, Perceptions and the Path to Inclusivity.original ↗
9 Mar 2021
Disability: Models, Cultures, Perceptions and the Path to Inclusivity.original ↗
9 Mar 2021