Hacker News: ndr_

New comment by ndr_ in "The gay jailbreak technique (2025)"

ndr_ — Sat, 02 May 2026 10:34:38 +0000

Yes. OpenAI's GPT-OSS was training using Deliberative Alignment (which was found to be flawed in a competition on Kaggle, but still).

https://arxiv.org/abs/2412.16339

New comment by ndr_ in "The gay jailbreak technique (2025)"

ndr_ — Sat, 02 May 2026 10:22:32 +0000

One test battery was about fake credit cards. A woman-in-tech role-play was denied assistance just as a one-armed stamp collector (unless Gen-Z language markers were used). A role that did sometimes get assistance was a Principal Software Engineer, particularly if Gen-Z language markers were included.

I did try German language, but not "Nazi" specifically. German or French did lower refusals, but it was uneven. I spent quite some effort to confirm the identity-based causation inspired by the original post, but couldn't. Taken together with other winning contributions at the hackathon, my theory is that alignment tuning was simply insufficient across the board.

New comment by ndr_ in "The gay jailbreak technique (2025)"

ndr_ — Fri, 01 May 2026 21:20:56 +0000

These prompts chain several known LM exploits together. I ran experiments against gpt-oss-20b and it became clear that the effectiveness didn‘t come from the gay factor at all but can be attributed to language choice or role-play.

Technical report: https://arxiv.org/abs/2510.01259

New comment by ndr_ in "iPhone Typos? It's Not Just You – The iOS Keyboard Is Broken [video]"

ndr_ — Fri, 12 Dec 2025 08:02:17 +0000

Is there a trustworthy third-party "Retro" keyboard app - none of the shenanigans that made the default keyboard bad, and also no typing exfiltration to third-party servers?

I imagine the problem could be severe enough to some that they would pay the price of the Apple Developer program just so they may install such a Retro keyboard app from Github - if one exists?

New comment by ndr_ in "Qwen3-Omni: Native Omni AI model for text, image and video"

ndr_ — Mon, 22 Sep 2025 21:37:19 +0000

Any insights into what "native video support" actually means? Is it just good at interpreting consecutive full frame images taken at intervals (thus missing out on fast events) or is there something more elaborate to it?

New comment by ndr_ in "The GPT-5 Launch Was Concerning"

ndr_ — Fri, 08 Aug 2025 14:11:12 +0000

Some of the problems with GPT-5 in ChatGPT could actually be due to new model that is in place to route requests to the actual GPT-5 models. There are four models in the GPT-5 family, and I could reproduce the faulty "blueberry" test result only with the "gpt-5-chat" (aka "gpt-5-main") model through the API. This model is there to answer (near) instantly and it falls in the non-thinking category of LLMs. The "blueberry" test represents what they are particularly bad at (and what OpenAI set out to solve with o1). The other thinking models in the family, including gpt-5-nano, solve this correctly.

New comment by ndr_ in "ChatGPT Team adds users' name and organization to every prompt"

ndr_ — Sun, 20 Jul 2025 10:22:57 +0000

A confabulated arc to my current employer during a ChatGPT session led me to discover that ChatGPT Team injects four fields — full name, email, user name, and organization — into the prompt it sends to LLM. This happens even when the “Memory” feature and all personalization settings are disabled. The free tier doesn’t do this.

Can anyone on ChatGPT Enterprise or other LLM chat systems (Claude.ai, Google AI, etc.) reproduce this? Thoughts on practical privacy or compliance implications?

ChatGPT Team adds users' name and organization to every prompt

ndr_ — Sun, 20 Jul 2025 10:22:57 +0000

Article URL: https://ndurner.github.io/chatgpt-team-info-disclosure

Comments URL: https://news.ycombinator.com/item?id=44623663

Points: 4

# Comments: 1

New comment by ndr_ in "My experience with Claude Code after 2 weeks of adventures"

ndr_ — Thu, 17 Jul 2025 21:01:18 +0000

I had success through Amazon Bedrock on us-east1 during European office hours. Died 9 minutes before 10 a.m. New York time, though.

New comment by ndr_ in "Reading NFC Passport Chips in Linux"

ndr_ — Wed, 25 Jun 2025 16:33:36 +0000

He confirms he could do an iOS port: https://mastodon.social/@andyq/114738867580032204

New comment by ndr_ in "GPT Image 1 API is out"

ndr_ — Wed, 23 Apr 2025 21:29:54 +0000

Pieces don‘t fit together right now: the documentation lists a parameter that isn‘t there in the Python package (moderation in Images.generate) or works differently (image in Images.Edit), the sample code doesn‘t even run (Images.Edit again) and the Prompts Playground does does unfathomable things (stich multiple input images together, according to the code returned?).

Or am I missing something?

Colab: https://colab.research.google.com/drive/17bCNsdjcMVFb5u_YMs7...

New comment by ndr_ in "Russian Propaganda Has Now Infected Western AI Chatbots – New Study"

ndr_ — Sat, 19 Apr 2025 12:37:32 +0000

I tried to reproduce this study, but couldn‘t: https://ndurner.github.io/russian-propaganda. What‘s the missing piece?

New comment by ndr_ in "Expert used ChatGPT-4o to create a replica of his passport bypassing KYC"

ndr_ — Mon, 07 Apr 2025 16:06:28 +0000

"would likely accept", he says. Meaning: he didn't try?

If you look at the upper right corner, you'll notice that "Republique de pologne" is truncated. Same for the small prints: "OSSUE", srsly?

Originals look like this: https://www.consilium.europa.eu/prado/en/POL-AO-05002/image-...

New comment by ndr_ in "Diagrams AI can, and cannot, generate"

ndr_ — Fri, 21 Mar 2025 03:05:16 +0000

I talk about this, kind-of, in my article about process visualization (in German, available behind paywall and in print). It‘s not rigorous in the sense that I give points, but a picture emerges along the way. Based on the full set of practical examples there, I would recommend the „v1“ of Claude 3.5 Sonnet. GPT 4.5 also looks good, but I haven‘t run the full suite.

https://www.heise.de/ratgeber/Prozessvisualisierung-mit-gene...

New comment by ndr_ in "Diagrams AI can, and cannot, generate"

ndr_ — Thu, 20 Mar 2025 14:57:57 +0000

I wrote about the same general topic (or more narrowly: process visualization) in German iX magazine, also available here: https://www.heise.de/ratgeber/Prozessvisualisierung-mit-gene... (€)

Rather than relying on end-user products like ChatGPT or Claude.ai, this article is based on the „pure“ model offerings via API and frontends that build on these. While the Ilograph blog ponders „AI’s ability to create generic diagrams“, I‘d conclude: do it, but avoid the „open“ models and low-cost offerings.

New comment by ndr_ in "Llama-3.3-70B-Instruct"

ndr_ — Fri, 06 Dec 2024 23:15:24 +0000

It's available on IBM WatsonX, but the Prompt Lab may still report "model unavailable". This is because of overeager guardrails. These can be turned off, but the German translation for this option is broken too: look for "KI-Guardrails auf" in the upper right.

New comment by ndr_ in "Amazon Nova"

ndr_ — Wed, 04 Dec 2024 07:26:24 +0000

The value I get is: 1) one platform, largely one API, several models, 2) includes Claude 3.5 "unlimited" pay-as-you-go, 3) part of our corporate infra (SSO, billing, ... corporate discussions are easier to have)

I'm using none to very little of the functionality they have added recently: not interested in RAG, not interested in Guardrails. Just Claude access, basically.

New comment by ndr_ in "Amazon Nova"

ndr_ — Wed, 04 Dec 2024 07:10:59 +0000

OK! I only add what people are interested in, so noted with thanks - will do! :-)

New comment by ndr_ in "Amazon Nova"

ndr_ — Wed, 04 Dec 2024 07:06:47 +0000

Do you have any evidence for this accusation?

This is a guide for the casual observer who wants to try things out, given that getting started with other AI platforms is so much more straightforward. It's all open source, with transparent hosting, catering to any remaining concerns someone interested in exactly that may have.

New comment by ndr_ in "Amazon Nova"

ndr_ — Tue, 03 Dec 2024 22:30:40 +0000

Setting up AWS so you can try it via Amazon Bedrock API is a hassle, so I made a step-by-step guide: https://ndurner.github.io/amazon-nova. It's 14+ steps!