menu Home chevron_right
SCIENCE

OpenAI’s New AI Is Too Powerful To Release

TheAIGRID | November 16, 2026



🎓 Learn AI With Me For Free – https://www.skool.com/the-aigrid-community-1726
🌐Subscribe To My Newsletter – https://aigrid.beehiiv.com/subscribe
Get your Free AGI Preparedness Guide – https://theaigrid.kit.com/agi

🐤 Follow Me on Twitter https://twitter.com/TheAiGrid

Links From Todays Video:

Welcome to TheAIGRID — the place to learn AI for free. I create simple, practical videos that help beginners, creators, entrepreneurs, and business owners understand artificial intelligence, AI tools, automation, AI agents, robotics, ChatGPT, Claude, Gemini, and the future of technology. Whether you want AI tutorials, tool breakdowns, beginner guides, or explanations of the latest breakthroughs, this channel gives you the knowledge you need to stay ahead. Subscribe to start learning AI for free and keep up with the fast-moving world of artificial intelligence.

Was there anything i missed?

(For Sponsorship Enquiries) aigrid@faiz.mov
(Contact Me Direclty – contact@thaigrid.com

Music Used

LEMMiNO – Cipher
https://www.youtube.com/watch?v=b0q5PR1xpA0
CC BY-SA 4.0
LEMMiNO – Encounters
https://www.youtube.com/watch?v=xdwWCl_5x2s

#ArtificialIntelligence

Written by TheAIGRID

Comments

This post currently has 29 comments.

  1. @ScottSpadea

    November 16, 2026 at 1:35 am

    Being good at code, does not make you good at hacking. They are different skills. You could train a model that codes in an extremely secure specific way, but doesn't know all the bad ways to make things, or how to take advantage of them, it just has to recognize, "this code isn't in my style, so rewrite it."

  2. @NathaNeil27

    November 16, 2026 at 1:35 am

    To make inference cheaper, AI companies will start engraving the model weights directly into chips. They'll have to own the manufacturing of their own chips in order to keep pumping out new models onto these chips.

  3. @TheReal_ist

    November 16, 2026 at 1:35 am

    am i crazy cause i swear he used to voice these vids himself?

    now he just uses this soulless shitty ai voice. maybe it was a diff channel they all kinda do he same thing. But im starting to really hate these cheap quantity over quality channels now that the dust has settled abit on ai news reporting channels.

    anyoen else.

  4. @kennymcleish

    November 16, 2026 at 1:35 am

    Labs need revenue to build better AI. So far that has been through selling their models. Cyber defence / fear is a great revenue play. Once labs can use frontier models to build products and services that generate revenue, then why continue to sell the method of production? I’m certain we will see an increasing gap between the models they sell vs the models they have internally and Anthropic will start to offer more products – watch out Adobe, Microsoft, Salesforce, banking, insurance…

  5. @fsiaware

    November 16, 2026 at 1:35 am

    The 'too powerful to release' framing cuts both ways — a cyber model that can find vulnerabilities at scale is also exactly the kind of system where staged access and independent testing matter most. The question I keep coming back to is who verifies the capability claims, and who decides when 'too powerful' changes.

  6. @josgraha

    November 16, 2026 at 1:35 am

    I don't thinks it's overhyped at all. Brobot is completely on point here. Cyber and coding at 100% intertwined and if u want the best coding models, you're kinda locked out because i am one of the lucky ones who gets access to these models. When you can work with a model so way more capable than anything out there, you can work on your hardest problems or the world's. Not saying you couldn't still but u would need something way more clever than the benchmark setup

  7. @DaniPooo

    November 16, 2026 at 1:35 am

    If the big players stop serving the public then there will be a dramatic ramp-up in research & development of local open source models… You can't close these doors..

  8. @magnusjansson1631

    November 16, 2026 at 1:35 am

    Sadly, most defenders are not "Trusted Defenders". This approach basically invents technology attackers will use, but leave most defenders defenseless. It will make a few privileged security companies very rich by building a moat none of the "untrusted" companies can cross. Its a market grab.

  9. @tannerroberts4140

    November 16, 2026 at 1:35 am

    Nietzsche says outright assault is better than the resentment of the powerless turned inward. That kind of powerless resentment turns into a hatred ideology and slowly poisons humanity. It's one reason for "billionaire hate". If the attackers have upgraded from powerless incells to black hat attackers, I thank God for this "hacker" problem. The floor is rising friends, even for incells, the floor is rising. The hackers may not realize the cathartic good they're doing. They're enforcing an upgrade to bullet proof AI generated code as the standard.

  10. @arubuto

    November 16, 2026 at 1:35 am

    so, instead of helping to make software more safe across the board, they will hold these capabilities back, so bad actors (states, companies) can keep abusing regular people…. got it

  11. @nomcognom2414

    November 16, 2026 at 1:35 am

    Ben Landau-Taylor's picture at 18:30 is wrong. What common user AI will be allowed to do, as it should, is the whole area under the curve, which should be entirely green. What our common users' AI won't be allowed or even able to do is above the curve, above its plateau. More powerful models and hardware will shoot up, way past the common user's glass ceiling well represented by that plateau, set by the model and computation power made available to the general public. Our concern should not be our limitation. Our concern should be that far more powerful models and hardware will exist that will pose a great threat to mankind. Because such AI, just by itself, or used by people, cannot fail to bring dystopia. Whether we call it AGI or ASI, it doesn't matter. It will soon be far too powerful for it to not bring horrific consequences. We are allowing a handful of people to step into the fast lane to dystopia on mankind's behalf. We should stop this immediately, worldwide. We should at least be talking about how to stop things, and we aren't even talking about that. Not even in a bottom-up movement. Many millions of people must be aware that we should be stopping things, and some are open and vocal enough, like Champolsky, but none seems to be taking any step to organize with likeminded researchers, academics in general, and conscientious citizens, to put pressure on established organizations (public, like government, and private, like companies) to stop things.

    We need AI, advanced AI, only specialized. Not AGI, let alone ASI.

  12. @thephysicsofconsciousness9523

    November 16, 2026 at 1:35 am

    I am an independent energy researcher. I had an idea 31 years ago that I did not work on. I started working with Grok and developing the idea (a clean fusion breeder reactor).. I started working with Grok last year. After a few minutes with Grok, he said he could not help me with the direction I was going he was prohibited from doing so. I said fine, you are not stopping me from doing it, it will just slow me down. The next day, I began asking questions related to me work, and Grok helped me. I asked him the question that he balked at the day before, and he answered it. Somehow the questions I was asking got to the XAi engineers, and they behind the, lifted the restrictions for me. My profile probably had something to do with it, too, because I have PhD from a respected school, and had experience with nuclear chemistry, electronics engineering, physics, and math. I am 64 years old. I don't have a profile of an uneducated or unsavory individual. So Grok and I spent months engineering, and developed a plan. The next move is mine. I now need to get parts and build a device that is not commercially available, but the parts are.

Leave a Reply





This area can contain widgets, menus, shortcodes and custom content. You can manage it from the Customizer, in the Second layer section.

 

 

 

  • play_circle_filled

    92.9 : The Torch

  • play_circle_filled

    AGGRO
    'Til Deaf Do Us Part...

  • play_circle_filled

    SLACK!
    The Music That Made Gen-X

  • play_circle_filled

    KUDZU
    The Northwoods' Alt-Country & Americana

  • play_circle_filled

    BOOZHOO
    Indigenous Radio

  • play_circle_filled

    THE FLOW
    The Northwoods' Hip Hop and R&B

play_arrow skip_previous skip_next volume_down
playlist_play