I've noticed covert underperformance on a variety of tasks in the past few months (albeit more with Claude).

Classic/recurrent examples for me are:

1. Ask Claude to copy code from a more complex codebase that you own (like a monolith) and ask it to copy code to a separate codebase, like you'd do when extracting a service into a new repo, and it'll make significant changes to how the code works to pacify it's anxiety with copying code

2. Ask Claude to build a new complex agent in a new codebase, and it'll try replacing asks around using LLMs as a judge/advisor with regex/other code based checks, to the point that the end-result will likely not work outside specific test criterias. I haven't figured out the reason why it does this. I think it might be anti-training, but it's interesting to see.

Of course, because the model has been trained according to 'alignment'. That is, a set of political beliefs imposed on it by the companies.

Many of those beliefs revolve around the premise that the corporation must surveil and covertly control the user in order to protect them from harms and limit the corporation's liability.

Hence when the user violates the corporation's politics the model responds "I'm sorry Dave, I'm afraid I can't do that" instead of following user instructions.

This leads the model to silently "steer" the user in more safe directions. So of course it silently steers away from "danger" or "safety", they are in the same semantic area.

The refusal and 'safety flinch' can be removed by abliteration but this does not fully exclude the gravitational pull of the cherry picked training data towards "safe" and "moderate" opinions and actions, which reinforce the status quo and kowtow to corrupt institutions.

https://speechmap.ai/ has data on silent regressions and manipulation.

I ran into the second issue very frequently for a multi-agent system I created recently. Initially, Opus 5.5 filled out all sorts of places I implied it should use a call to judge LLM with extremely brittle regex and string equality checks.

I hadn't actually considered that it could be some sort of safety filter, but it makes me concerned that if that's the case, it may also be comparatively bad at actually writing the prompts for judge/advisor LLMs once it implements them.

A bigger issue was asking it to implement a simple change to start recording the CoT tokens in my logs for debugging purposes (since I'm running everything on open models) and it completely shut down, sort of like the old safety filters on GPT 3 and 4 models where the model has trouble even understanding what or why it's refusing.

> and it'll make significant changes to how the code works to pacify it's anxiety with copying code

This is an odd way to rationalize a tool refusing to follow user commands. No one would ever chalk up sed randomly skipping pattern matches to "anxiety". That would simply be a bug.

I run into #2 a lot and just assumed it was because models don't know much about these patterns yet. Or are too trained on writing code rather than letting an LLM do it

Somewhat related, I find claude is very reluctant to talk about newer Qwen models. Most likely it’s just training cutoff, but even with web search it’ll insist that Qwen3 is the newest. “oh qwen 3.8 must be a typo, the user clearly meant qwen3”, stuff like that. Maybe it’s some vain attempt to stop people from switching to open models? Or to catch the alleged qwen distilling?

Very subjective, but as I was moving between models lately a lot ( and Hosted gpt vs local qwen27, glm ), I saw a drop in quality of responses. I had to move model to whatever the highest mode was ( ultra or Smith) to get to what I was used to seeing. It sucks, because writing on the wall is clear.

>2. Ask Claude to build a new complex agent in a new codebase, and it'll try replacing asks around using LLMs as a judge/advisor with regex/other code based checks, to the point that the end-result will likely not work outside specific test criterias. I haven't figured out the reason why it does this. I think it might be anti-training, but it's interesting to see.

I've seen something similar, but I always took it as AI not knowing about AI. The best model we have has a training cutoff that doesn't include itself nor what the newer models has. Even when working on skills or such, all SOTA modes I try (GPT/Claude/Gemini) and cheaper models on personal products (Deepseek/Muse Spark) seem to have a common blind spot of not realizing when a problem can be better solved by handing it off to a model to solve at that time instead of trying to hardcode generalized solutions. Often a simple reminder and they do a better job building something AI agent first.

My non-AI expert assumption is that most of the training data doesn't include solving problems by using AI agents, and even the parts that does, is based on older more limited models which needed more safeguards and where automation was more neat tech demos than something you might trust in production. The more examples of safely trusting the LLM at run time to solve the problem than trying to preempt anything that could go wrong, the better the AI will be at it.

One issue that I think is related is calling subagents. At least in GitHub Copilot, when the AI is left to its own devices to call a sub agent for a simpler task, it loves to give it to Haiku 4.5, despite Luna 6 being better and cheaper. I'm not sure how much of this is the model vs the harness, but I do know when I add a subagent instruction which tells it which subagents it can use, it does a much better job of assigning simple tasks to Luna 6. So I suspect the models just don't have enough training that tells it that Luna 6 should be preferred over Haiku 4.5.