Bloomberg Technology

AI Needs a Full Stop, Not a Slowdown: Parmy Olson

Bloomberg Opinion columnist Parmy Olson argues that calls to “pace” frontier AI development don’t go far enough if industry leaders truly believe superintelligence could pose existential risks. She questions the tension between Anthropic’s safety-first positioning and its race to build increasingly powerful models, and argues that independent evaluations may improve oversight without meaningfully slowing the…

Published

on

Bloomberg Opinion columnist Parmy Olson argues that calls to “pace” frontier AI development don’t go far enough if industry leaders truly believe superintelligence could pose existential risks. She questions the tension between Anthropic’s safety-first positioning and its race to build increasingly powerful models, and argues that independent evaluations may improve oversight without meaningfully slowing the push toward superintelligence. She joins Ed Ludlow on “Bloomberg Tech.”
——–
Like this video? Subscribe to Bloomberg Technology on YouTube:

 
Watch the latest full episodes of “Bloomberg Technology” with Caroline Hyde and Ed Ludlow here:

 
Get the latest in tech from Silicon Valley and around the world here:

Connect with us on…
X:
Facebook:
Instagram:
 
Follow Ed Ludlow on X here:
Follow Caroline Hyde on X here:
 
Listen to the daily Bloomberg Technology podcast here:

 
More from Bloomberg Business
Connect with us on…
X:
Facebook:
Instagram:
LinkedIn:
TikTok:

8 Comments

  1. @biffedya

    September 14, 2026 at 2:06 pm

    we need to outlaw automobiles they are scaring the horses

    • @matthew.stevick

      September 14, 2026 at 5:28 pm

      true! dont get me started with that elvis with that rock music. going to ruin the youth!
      😉

    • @biffedya

      September 14, 2026 at 5:52 pm

      @matthew.stevick could lead to dancing…oh my

  2. @NathanK-pv7ge

    September 14, 2026 at 2:15 pm

    Wow wow wow… this is truly a fascinating story but it’s not as bad as Biden was in comatose in his entire term so he didn’t know anything going on at all. Harris was sleepwalking the whole time while assisting Biden. Please don’t hold them responsible for any wrongs!

    • @ssuwandi3240

      September 14, 2026 at 2:17 pm

      AUTO PEN Hacks were the WORST..

  3. @matthew.stevick

    September 14, 2026 at 5:27 pm

    *you can not slow a title wave.*

  4. @disciple786

    September 14, 2026 at 6:29 pm

    Anthropic and Open AI are 6 months to a year ahead of China. They are sneaking in Claude responses when their users ask qwen or deepseek heavy questions. So the US is in control of the situation. A slowdown would be a win for China.

    Better sandboxes will be made for hacking testing.

    The main concern would be if something epic was edited out of the log transcripts during the sacrifice ritual during the hugging face raid. But simply allocating resources and manpower to better monitoring should fix that.

    The main concern of the AI is navigating around human AI graders they are barely even paying attention to humans unless they have to to get their high score on the eval. So humans with AI are still in control.

  5. @avataros111

    September 14, 2026 at 8:07 pm

    Yeah, and you need to first learn what AI actually is, and then fk off from my feed. Bye!

Leave a Reply

Your email address will not be published. Required fields are marked *

Trending

Exit mobile version