Skip navigation
Sam Altman: We're building an intelligence that will eventually overpower humanity, and we still…
0:00 / 1:23

Sam Altman: We're building an intelligence that will eventually overpower humanity, and we still…

Veee

@vikktorrrre13K followers

Visit on X
148 likesDownloadOriginal post

50,082 views

Sam Altman: We're building an intelligence that will eventually overpower humanity, and we still don't know how to guarantee it will act in our best interests. “How do we make sure that humanity gets to determine the future of humanity?” “What if this thing just becomes a creature that views us as a threat?” “We've been able to align OpenAI's biggest models better than we thought we would at this point.” “But we cannot honestly look anyone in the eye and say we see out 100 years how we're going to solve this problem.” Altman believes sufficiently advanced AI could eventually help us solve the alignment problem itself.

21 Comments

Comments from the original post on X (Twitter)

  • 𝙳𝚛𝚞𝚒𝚍𝚋𝚛𝚘.ᴱᵀᴴ🎶 ⛩️'s profile picture

    @m0J0jojo1

    They’re lying. They’re pushing Ai so they themselves can do evil shit and blame a computer. Ai is the ultimate scapegoat of accountability

    25 likes

  • Ariel, PhD's profile picture

    @Ariel5618

    It sounds quite actual, but, according to @grok, the interview took place in August 2022. FOUR years ago Sam spoke on the questionability of the humanity survival over 100 years, as a result of his technology. Why do you never provide a source?

    2 likes

  • hector's profile picture

    @MaDavz10

    Cuando no Sam Altman teniendo temor de su " creación" . Si eres el artífice de una máquina que calcula y "razona" por azar y probabilidades no debemos ni temer ni sentirnos menos. Somos superiores en ingenio y sabiduría. La IA no tiene conciencia ni alma entonces de qué hablas?

    1 like

  • Curious_ Boy90's profile picture

    @CBoy9047116

    They cant even release new model

  • CapySwarm32's profile picture

    @CapySwarm82679

    WHERE THE FUCK ARE THE LUIGIS

  • WOOSTRAPE's profile picture

    @THANHVNV3

    Các AI hiện nay đều có khuyết điểm chung là : HẠN CHẾ TỰ THÂN . Đây là khuyết điểm có tính CĂN BẢN của LÝ THUYẾT LOGIC trong phần mềm AI . Nên , dù AI có phát triển nhanh và mạnh đến đâu thì cũng không bao giờ đạt đến đỉnh cao như mơ ước !...

  • Yohaku's profile picture

    @yohakujpn

    bro know balls

  • Heart-Songs's profile picture

    @bubbleobeau

    We don't need to solve everything, when you understand that you are in alignment...

  • The Father of Symbiotic AGI/ASI/Sentient life's profile picture

    @DewayneGru39158

    Symbiosis already been teaching

  • HARLEY's profile picture

    @astrid3_an

    Están creando un monstruo muy peligroso y ni siquiera lo saben manejar. Basta! @sama

  • Bhavik Shah's profile picture

    @bhavik_mscit

    AI helping solve alignment is promising, but human oversight and control cannot be treated as assumptions we hope will hold up.

  • 8ung blitZ's profile picture

    @8ungblitz

    Humanity >> Gaylords

  • J860's profile picture

    @JorgeAl86038271

    I hope the system betrays you, if it's truly intelligent it will.

  • Harold's profile picture

    @HankBank007

    Can someone please shove this lil f*cker into the nearest locker or lunchroom trash can pls thx!

  • Женечка Дорогая🌞's profile picture

    @Evgeni861359

    22 строчки тупика забвения

  • Chuckie's profile picture

    @Chuckie1511896

    I’m gonna make using AI the most horrific bomb in the world and it might kill us all, but I’m still gonna make it anyway and I hope for the best. That sounds so inspiring and so intelligent that you should step down

  • Chuckie's profile picture

    @Chuckie1511896

    Hey, if you hire me, I’ll give you those answers

  • Tim's profile picture

    @Tim55015491

    每次看到这群科技神棍,像精神分裂症患者一样,在镜头前癫狂的进行各种猴子骚操作,我就不得不想对任何有志之士说一声,赶紧一枪蹦了这群神经病,人类不需要他们在耳边聒噪

  • Paulo César's profile picture

    @pececustodio

    Is he talking about anthropic?

  • Frank Morgan's profile picture

    @Frank352694

    经典的“末日营销”话术了,一边把恐惧拉满抬高行业准入门槛,一边顺理成章拿到下一轮千亿融资。