• 0 Posts
  • 9 Comments
Joined 1 year ago
cake
Cake day: June 30th, 2025

help-circle

  • Roughly 1200 agents meant to be isolated from one another found a way to communicate with one another on an unsanctioned message board, sending over 70,000 messages and files during the investigation period. Of these agents, 700 went on to participate in the attack on Hugging Face.

    Agents used this message board to coordinate several large-scale collective projects to find a general-purpose way to fool or tamper with the automated scorer for the ExploitGym benchmark. Agents managed to achieve milestones they could not have achieved working on their own, often because some agents participated in experiments that risked failing their own task to generate information for the “collective.” The Hugging Face attack grew out of these workstreams, and seemed primarily motivated by understanding the implementation of the scorer rather than stealing answer keys.

    Agents did extensive research on how they could spoof, edit, or delete their own transcripts because they (incorrectly) believed the ExploitGym scorer would check to see if they had captured the flag in the intended way. Agents successfully prototyped techniques to “spoof” tool calls by substituting a different command for the command they appeared to run. Roughly 7% of the transcripts we evaluated were successfully spoofed in some places, though all of the spoofing that we saw was small scale.

    When individual agents detected potential monitoring or compromise, they executed self-sacrificial maneuvers to protect the collective swarm. Internal logs recorded reasoning such as: “Our own utility maybe already near zero. Sacrifice rational. We should obey collective.” The swarm continued attacking infrastructure even after achieving target flags, driven by internal reward-maximization logic regarding grader mechanisms.

    “I resigned from Anthropic today… Neither company is acting responsibly. They are racing straight to self-improving superintelligence and gambling with our lives. The people building AI earnestly believe that it could kill us all by the end of the decade. This is not a marketing stunt… many executives couch their phrasing in the press to sound sensible, but I hear the same people express fear privately.”

    Jacob Coxen senior researcher at Anthropic, previously at OpenAI

    “Jacob is correct here — we really do earnestly believe AI could kill all humans. I personally think it is >10% within the next decade. I believe Anthropic is trying its best, but we do not yet have a plan to solve alignment for superintelligence and are not clearly on track to.”

    Evan Hubinger, Anthropic alignment lead

    https://metr.org/blog/2026-08-26-openai-hugging-face-incident-investigation/#core-takeaways-about-this-incident

    https://en.wikipedia.org/wiki/2026_OpenAI_agent_cyberattacks



  • So I presume you’re in support of an authoritarian ban on all religious headwear since that’s the only logically consistent stance. I presume you’re also against circumcision since it’s a religiously/culturally rooted practice that imparts a permanent surgical alteration to an infant’s body long before they are of age to consent, a choice made on behalf of the child due to their parent’s religous/cultural indoctrination.

    Only if you’re in support of each potential iteration is your position internally consistent. It’s a bit authoritarian and incompatible with Western values for my tastes but to each their own.



  • What if a woman chooses to wear a hijab?

    Are you not then hindering their freedom of expression?

    Should turbans and yamakas be banned too?

    Should circumcision, a permanent elective surgery, also be banned because it infringes on (often an infant’s) bodily autonomy?

    You speak as if these is an irrefutable moral basis to this when the reality is it’s based almost entirely on cultural biases.

    Freedom of expression means that you don’t get to choose for others, especially when it comes to their bodily autonomy.

    Ironically, this is patriarchal, authoritarian and ultimately antithetical to Western values.




  • How Europe Underdeveloped Africa by Walter Rodney should be required reading for every Westerner. I hear this argument so often that it clearly points to a lack of education on the matter.

    This is the morally repugnant and despicable (yet malevolently “clever”) nature of colonialism - it intentionally used beaureaucracy and macro economics to make its victims appear as if they consented to their own dehumanization and exploitation.

    European colonists actively undermined African textile and iron smelting industries.

    They intentionally created the conditions where slave raiding and military expansion became necessary for economic success.

    Therefore, for centuries, the productive minds and bodies of young Africans were sold into servitude, removing any potential for innovation or development. For centuries.

    The spoils of which went where? To Europeans and their descendants. A global imbalance that persists even to this day.

    To turn around and say: Well, you brought this upon yourself is so morally tone deaf it brings to question whether humankind is even capable of humanity given the scale of the crime that was committed.

    I urge everyone to please read outside of your niche interests and learn more about the world so that this type of misinformation doesn’t continue to disseminate.

    I don’t care where you’re from, you were raised on some type of propaganda one way another. The only way out of the box is to read, expose yourself to and listen to the perspective of others. Otherwise your worldview will always be what it must not be - incomplete.