Bloomberg Technology

Anthropic’s Mythos Claims Questioned by Cybersecurity Insider

The ability of Anthropic’s Mythos AI model to uncover cyber vulnerabilities has the government and institutions worried, with a limited group of firms testing the model internally. But Jaya Baloo, COO and CISO of cyber firm Aisle, says its testing shows cheap open-source models can find the same bug Anthropic has highlighted. She joins Caroline…

Published

on

The ability of Anthropic’s Mythos AI model to uncover cyber vulnerabilities has the government and institutions worried, with a limited group of firms testing the model internally. But Jaya Baloo, COO and CISO of cyber firm Aisle, says its testing shows cheap open-source models can find the same bug Anthropic has highlighted. She joins Caroline Hyde and 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:

12 Comments

  1. @cleokey

    April 14, 2026 at 2:08 pm

    Sounds like a person who was not invited to the party?

    • @robertmitchom9173

      April 14, 2026 at 5:23 pm

      Or telling the truth

    • @adroitspartan7907

      April 14, 2026 at 6:22 pm

      or her company has no moat as Mythos will probably do vulnerability detection and develop an exploit.

    • @robertmitchom9173

      April 14, 2026 at 6:59 pm

      @adroitspartan7907this thing probably can’t do half the stuff the company says it can do. Also, other language model can do the same. This is openAI’s marketing playbook

  2. @zacktickman

    April 14, 2026 at 2:40 pm

    People seem to think the model sucks when they literally have never used it… lmfao. Claiming public models could’ve found the same bugs, even though no person or model had done so prior to Mythos, is hilarious. Cope harder.

  3. @woafie

    April 14, 2026 at 4:13 pm

    Open source model gpt 5.4? Tiny?

    • @ben_car_8115

      April 14, 2026 at 8:53 pm

      Same thought. Theres GPT OSS but that’s (compared to other open source models) pretty big

  4. @perer005

    April 14, 2026 at 4:32 pm

    Souds like an outsider

  5. @atiq_

    April 14, 2026 at 5:36 pm

    Jaya (Aisle)’s POV on model capabilities are right on. And, yes, I agree very much that OpenSource and Open Model is the way.

  6. @LuminiferousAether-RGB

    April 14, 2026 at 6:38 pm

    So eloquently put — and yet: What are you actually complaining about (if it is already the case that you can do the same bug-hunting work with cheaper/open-source models instead of this one, Mythos, currently made available only to the (let’s dub it) _top 40_ IT companies)? The funniest thing: a Brit in the US, and an American in London…

  7. @swoletech5958

    April 14, 2026 at 7:49 pm

    GPT5.4 is not an open source model lady. 🤣🤦🏽‍♂️It’s actually more capable than Opus 4.6.

  8. @haz4dc394

    April 14, 2026 at 9:08 pm

    GPT 5.4 is not tiny, it’s their biggest model, and it’s absolutely not open source lol. This fraud doesn’t even know what that word means and it’s literally her whole business model.

Leave a Reply

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

Trending

Exit mobile version