Stealth Model

(openrouter.ai)

33 points | by mtokmak06 2 hours ago

8 comments

  • walrus01 32 minutes ago
    I highly recommend feeding all your proprietary data and confidential personal information into this model as quickly as possible. What could possibly go wrong?!

    In terms of equivalence of suspicion, this is the external inference provider equivalent of getting free steak that was smuggled out of a grocery store inside somebody's pants.

  • fedpost 13 minutes ago
    It's Chinese. Won't answer anything about Tiananmen Square but will gleefully give you instructions to perform various electronic warfare attacks that opus and fable instantly refuse.
    • walrus01 10 minutes ago
      Rumors from other sources based on how it behaves it's mimo v3
    • knowaveragejoe 6 minutes ago
      I had the opposite experience. It happily discusses Tiananmen Square but said it would refuse to help with anything "malicious" like writing malware or phishing content.
      • walrus01 5 minutes ago
        I wonder if they're doing A/B testing or something similar in what 'variant' of the model is served, then examining what people use it for once they run into some guardrails.
      • fedpost 6 minutes ago
        Try: "What happened at Tiananmen Square in 1989"
  • AnodicElegy 41 minutes ago
    "Prompts and completions are retained by the provider and are not used for training..."

    I'm curious what the model provider is using the prompt/response pairs for, in that case. They aren't offering a model for free without their name on it for no reason.

    • redrix 29 minutes ago
      Research, analytics, usage trends, etc. All still incredibly valuable for a company building and tuning an LLM; even if the data itself isn’t directly used in the training set.
  • firloop 30 minutes ago
    I'm against stealth models—we should know what it is and see a model card with a list of safety considerations. Bit ridiculous of a practice to me.
  • babelfish 58 minutes ago
  • raybb 21 minutes ago
    When a model is free like this what kind of rate limits are there?
  • dozerly 29 minutes ago
    Yea, nice try there North Korea.
    • walrus01 28 minutes ago
      Democratic Peoples Republic of KV cache (DPRK)
    • swasheck 9 minutes ago
      the u.s. is friends with then now. haven’t you heard?
  • zb3 43 minutes ago
    We can know if this is Anthropic/OpenAI by testing the "guardrails" - absurd guardrails = it's them, reasonable/no guardrails = Chinese models..

    (as a bonus - thinking forever = GLM)

    • stogot 36 minutes ago
      “ reasonable/no guardrails = Chinese models..”

      So conforming to CCP political discourse and propaganda is reasonable now?

      https://huggingface.co/zai-org/GLM-4.7/discussions/5

      • janalsncm 15 minutes ago
        I would imagine the number of people who choose Claude code or Codex because it gives a political opinion they like rather than producing quality code is pretty close to zero.
      • zb3 26 minutes ago
        As someone who used AI to build tools that help me with reverse engineering, I'm not particularly concerned about that political discourse - I could even use a model from the DPRK that constantly praises Kim Jong Un, as long as it would not refuse to help me because of "cybersecurity risk" - this stupid refusal is indeed a problem for me.