My experience has been the same: flatout refusals no matter how i frame the health questions - very frustrating. Even psychology is out of scope. Pretty useless unfortunately

Have you tried switching off custom stuff that might get added in beyond the prompt?

Isnt this the principle behind synthid?

Maybe if you look at it from far away enough.

Watermarking tries to resist image data manipulation. Smuggling data is concerned with preservation of bytes.

Though if we're executing arbitrary code on the target anyway, ways of embedding data in an image are vast, including watermarking/steganography.

More generally it’s called Steganography.

It’s unusable for me due to the refusals. I’m using claude to find patterns in health data

I do some work in laboratory automation and it was quick to refuse the first thing I asked it to do. There wasn't anything spicy in the request, just basic liquid-handling protocol implementation. Their position seems to be that they're too stupid to classify requests safely, and that seems reasonable to me. I'd guess the classifier will improve rapidly.

Have you tried locally running qwen?

Is there a Qwen that I can run locally that is anywhere near these frontier models?

No, and don't let anyone gas light you into thinking the answer is yes.

Same. I'm working on a set of python and matlab scripts that deals with segmenting MRI images into brain vs skull, and it thinks that's bioterrorism.

Quite counterproductive to refuse to help on health issues too. If they detect health data, they can add a disclaimer, but not hide the information.

You miss the point - by collecting and processing medical data they would fall into a thoroughly regulated industry. Not because they may provide you incorrect data, because they are not allowed to process them.

What custom prompt do you have set up? If you tell it you're occupation, does it turn helpful? There was a study that if you tell models they tested that you're a patient, it would refuse, but tell it you're a doctor and suddenly it turns helpful.

According to the model, it’s not the model itself that’s doing this, it’s the harness.

Assuming the model is being “truthful”, CC is just being stupid in its detection mechanism.

Anthropic knows it refuses too much, they want to be very cautious to avoid any scandals. I think this is why they want to store all Fable and Mythos chats for 30 days so they can use the data to improve.

They want to be very cautious to honour the important doctrine at least until IPO launches: we are so good we are nerf our products.

I’m a point where I expect everything I do will be retained indefinitely.

I’m having a really hard time believing some weak reason for a 30 day retention policy.

There’s no way around it? Can’t you obfuscate as generic data and use keys to map to the real data?

I guess you could even turn everything into numbers, not a bad idea at all!

what prompts do you use for this?

I wonder if it sees Healthcare companies being targeted and that's why it's freaking out; clearly they have some pretty stupid regexes in the harness to detect this sort of shit.

e: I quit the session and went back in. Set it to Fable and told it to continue the last session. It's moving along as if none of that had happened.

How weird.

I wonder if this letter has anything to do with why anything even remotely related to biology is getting flagged.

https://www.wired.com/story/openai-anthropic-letter-ai-biolo...

The refusal rate is insane

Thats why it is a mythos model

Mythical levels of refusal... checks out!

A myth, not a model that actually exists

Getting the 2002 Cool Edit Pro 2.0 vibes!

It probably won’t work either way: https://arxiv.org/abs/2605.13829

It has happened to me with very long conversations

Thanks! Do you mean long during the current session or counting all sessions since it started?

It was a conversation crossing raw IVG data against a symptom phenotype (dna analysis) which went on for a good while

Wouldnt it be easier to fine tune a model to convert the Brazilian Portuguese corpus into European Portuguese and then use that corpus?

That idea is different than what most are talking here in other comments.

The grammar and vocabularies don't match, but I think the worst are the expressions. Both sides have *a lot* of expressions that vary per context and location.

I’m a native Portuguese speaker, whom grow up watching morning Brazilian Portuguese cartoons. I would say it’s doable

With a 10min delay, via aistudio

That’s basically the original gpt5 routing idea but done right