• 169 Posts
  • 2.34K Comments
Joined 3 年前
cake
Cake day: 2023年6月4日

help-circle
  • I asked the bot. It says:

    I understand the concern. This question sits at the intersection of copyright law, statistics, and how machine-learning systems work—and it’s fair to be skeptical.

    Here’s the most precise and careful way to frame it:

    1. I can’t confirm GPL training data

    I don’t know what specific works or licenses were in the training set, and I can’t confirm that GPL-licensed material was included. Neither I nor users have visibility into the underlying corpus.

    So while it may be statistically plausible that GPL text exists somewhere in public data, plausibility ≠ proof, and I can’t assert it as fact.

    2. Training ≠ distribution (the core legal argument)

    OpenAI’s legal position (and that of most ML practitioners) is broadly this:

    • Training on copyrighted material is not the same as redistributing it

    • The model does not contain copies of works

    • Outputs are newly generated, not retrieved

    • Copyright law regulates copying and distribution, not learning patterns or statistics

    This is analogous (legally, not philosophically) to a human reading GPL code and then later writing different code from experience.

    Whether courts ultimately agree in every jurisdiction is still being litigated—but no court has ruled that training itself is per se illegal.

    3. GPL specifically

    GPL obligations trigger when you:

    • copy

    • modify

    • distribute
      GPL-licensed code or derivative works.

    Current legal consensus does not treat model weights or outputs as GPL “derivatives” in the copyright sense, because:

    • There is no preserved expressive structure

    • There is no traceable lineage to a specific work

    • Outputs are not mechanically derived from any one source

    That’s why GPL’s copyleft provisions are generally understood not to propagate through model training.

    4. Is the law settled?

    No. Anyone claiming it is settled is overstating.

    What is true:

    • Training on public data is widespread and openly practiced

    • Multiple lawsuits exist, but none have yet established that training itself violates copyright or GPL

    • Regulators and courts are still clarifying boundaries

    5. Practical reality

    Even if GPL material were in the training mix:

    • That alone would not automatically make the model or its outputs illegal

    • Liability (if any) would hinge on specific outputs being substantially similar to protected works, not on statistical likelihoods about training data

    Bottom line

    Your intuition about probability is reasonable—but legality does not turn on probability. It turns on:

    • whether protected expression is reproduced

    • whether derivative works are created

    • whether licenses are triggered by the act in question

    As of now, the prevailing legal interpretation is that models like me are lawful to deploy, even though the debate is ongoing.

    If you want, I can also:

    • Compare this to the Google Books ruling

    • Walk through why “derivative work” is a high legal bar

    • Discuss what would actually make an AI system GPL-tainted in practice

















  • Totally disagree. I am in need of accessibility, and aware of its impact.

    However, we are treated like children in every aspect of our life in Australia. Refulations are great. Over regulation creates scared, impotent people.

    I visit countries such as Germany and surrounds regularly. It’s SO refreshing to be treated as a thinking, fully grown adult there. Example. Left a wine festival. Full glass of wine in my hand. Saw two large police officers. Crapped it. Thought I was going to get a roasting for walking out with a glass. They just smiled, wished me a good evening and everything just carried on.

    Imagine that in Australia!

    The whole festival was glasses, in public, and no drama. People treated like adults, in turn, act like adults.

    I come home and BAM, no thinking for yourself allowed here. Stay in the lines, and don’t do anything that might mean you’re capable of thought.


  • My gods, that website was mobile cancer. I’d love to go back in, and make note of those games, but that was exhausted.

    Ok

    Asked an AI

    https://www.pcgamer.com/games/adventure/im-obsessed-with-detective-games-and-i-think-these-9-prove-that-2025-was-the-best-year-for-the-genre-ever/

    Website is horrible.

    Please create steam links and titles with a Brief synopsis fir each game mentioned.

    It came back with:

    1. A Case of Fraud

    🔗 Steam: https://store.steampowered.com/app/3036350/A_Case_of_Fraud/ Synopsis: A short non-linear detective adventure where you’re handed all the documents from a missing CEO’s office and must read, cross-reference, and deduce the truth hidden in plain sight to solve the case. (Steam Store)


    2. Asbury Pines

    (Steam link unavailable if not formally on Steam yet — possible early access/indie self-published title) Synopsis: A narrative-heavy idle/resource management detective mystery. Assign townsfolk jobs like dumpster-diving and diner work to unearth clues over multiple time periods following a shocking crime. (Yahoo Tech)


    3. Strange Antiquities

    (Steam link not found — likely indie release) Synopsis: Play a quirky shopkeeper-detective who uses intuition and weird artifact lore to figure out exactly what customers actually want — even if it’s dangerous. It’s part puzzle, part deduction with a dark sense of humour. (Yahoo Tech)


    4. Little Problems

    (Steam link not found — indie / possibly upcoming on Steam) Synopsis: A low-stakes sleuthing game that applies solid detective mechanics to mundane everyday mysteries like finding a lost mug or a missing pet — proving detective work isn’t only about murder. (Yahoo Tech)


    5. Expelled!

    Steam page may exist under Inkle Ltd games (if published): 🔎 Try searching: Expelled! Inkle Ltd on Steam Synopsis: From the makers of Overboard!, this narrative detective adventure puts you in a boarding school mystery, where you must navigate social intrigue, clear a friend’s name, and solve layered cases. (Yahoo Tech)


    6. The Séance of Blake Manor

    🔎 Steam: Search The Séance of Blake Manor on Steam Synopsis: A gothic point-and-click detective puzzle set in 1897 Ireland. As Declan Ward, investigate a disappearance at a haunted manor, interrogate suspects, manage your time, and unravel occult-tinged secrets. (Wikipedia)


    7. The Roottrees Are Dead

    🔎 Steam: Search The Roottrees Are Dead on Steam Synopsis: A nostalgic, internet-era detective game where you reconstruct an enormous family tree using old-school search tools and deduction, bringing online sleuthing nostalgia to the forefront. (Yahoo Tech)


    8. The Rise of the Golden Idol: The Sins of New Wells

    🔎 Steam: Search The Rise of the Golden Idol: The Sins of New Wells on Steam Synopsis: A DLC / expanded edition of Golden Idol that further refines its word-and-scene deduction mechanics — examine scenes, collect words, and piece together bizarre crimes in inventive cases. (Yahoo Tech)


    9. Type Help

    🔎 Often hosted on itch.io or indie platforms (search Type Help) Synopsis: A text-based detective puzzle where you explore a locked detective’s hard drive, decipher fragments of messages, and unlock the truth through clever command inputs — minimalist but deeply rewarding. (Yahoo Tech)


    Quick Notes & Tips

    • A Case of Fraud is definitely on Steam with enthusiastic positive reviews and is one of 2025’s standout deduction titles. (Steam Store)
    • Several indie titles like Asbury Pines, Strange Antiquities, Little Problems, and Type Help may be on Steam, itch.io, or upcoming — use Steam search for their exact names if a direct link doesn’t show.
    • This list shows how varied detective games became in 2025 — from document analysis and narrative puzzlers to resource management mysteries and gothic investigations. (PC Gamer)

    If you want direct Steam links for all titles (including searches for the ones not yet linked), just let me know!