Login
You're viewing the f.duriansoftware.com public feed.
  • DrALJONESDrALJONES@mastodon.social
    Oct 3, 2026, 5:04 AM

    Interview: AI is not the threat. It's the people building it "that are the existential risks to humanity."

    "If a bridge collapses, we don’t ever ask, “Was the bridge uncontrollable? Was the bridge unethical?” And really, AI is no different than that."

    When we describe these systems as if they have their own agency, we’re helping the companies market themselves. "We’re also helping them sidestep accountability."

    ~Timnit Gebru

    democracynow.org/2026/10/1/tim

    #USPol #EUPol #News #AI #TechFascism ...2

    This is not models going rogue. This is gross incompetence & negligence ~Timnit Gebru Instead of discussing faulty, unreliable systems, the tech bros are marketing their systems as all-knowing, super- powerful, superintelligent. But any cybersecurity expert would tell you their security practices were laughable. For example, OpenAl was testing how well their models performed at finding & exploiting vulnerabilities in other systems. That's what they trained their models to do. But while testing this capability, you don't want to hack other organizations, so you should do it where there's no internet access. OpenAl didn't take this precaution, so the models were able to access the internet. This is not models going rogue. This is gross incompetence & negligence by a particular company, which actually they could be held criminally liable for. These men - who | believe are the actual real existential risks to humanity - are supposed self-regulate, self-police, & we have to rely on their morality. Oct 1, 2026 Abridged from https:/www.democracynow.org/2026/10/1/timnit_gebru_ai
    💬 1🔄 25⭐ 8

Replies

  • 💬 0🔄 1⭐ 4
  • 💬 1🔄 0⭐ 0
  • Adamadamsaidsomething@mastodon.social
    Oct 3, 2026, 7:14 AM

    @DrALJONES

    Its Munroe's diagram and it suggests future A.I. agents for humorous affect but the over all message is that A.I. is a tool and we need to be more worried about who it gives power to, at least that is the way I see it, so I think it reinforces the point Gebru is making, but we can all interpret the diagram differently. FWIW Munroe made this diagram well before the current sudden interest in this topic, and I think it pre-bunks the tech bros current spin rather then buy into it

    💬 0🔄 0⭐ 3
  • Adamadamsaidsomething@mastodon.social
    Oct 3, 2026, 7:18 AM

    @DrALJONES

    You have to ask yourself the question, if the A.I. is not self-aware, who is controlling (or not controlling) the A.I. that is controlling the swarms of killer robots?

    💬 0🔄 0⭐ 1
  • 💬 0🔄 0⭐ 0
  • 💬 1🔄 0⭐ 0
  • 💬 0🔄 0⭐ 0
  • A Bunch of Draculasdecay@todayiwilllaunchmyinfantsonintoorbit.com
    Oct 3, 2026, 9:28 PM

    @DrALJONES @adamsaidsomething I did read the piece and I did not miss the point. You're still still anthropomorphizing in exactly the way Gebru talks about:

    OpenAI could have been held liable for unleashing malware around the world, hacking Hugging Face, hacking all sorts of government agencies, but instead of doing that and holding them accountable for unsafe security practices that they should have had, basic security practices that cybersecurity experts would tell you they could have implemented, instead of that, they managed to redirect the conversation to some type of unprecedented rogue machines that don’t exist

    They turned on a problems-causing machine and are not being held accountable for the problems being caused and even extending the idea of "well, robots following their training" no, it's an industrial process, just an extremely complex one

    💬 0🔄 0⭐ 0
  • mccmcc@mastodon.social
    Oct 3, 2026, 6:02 PM

    @DrALJONES @adamsaidsomething IMO once we agree to call it "AI" we've already ceded that point (yes, I'm fully aware "AI" has a distinct historical meaning in computer science, but policy/moral discussions aren't taking place exclusively in computer science spheres, and anyway the computer science community meaning has shifted dramatically since 2018)

    💬 0🔄 0⭐ 0