We only have 2 years…

31 thoughts on “We only have 2 years…”

1. "Distillation attacks" Yeah, right. Just as any big US AI company has done against small time IP owners. If you don't have billions of dollars to sue these companies for copyright violation, they don't give a shit and will simply steal your work. Stop pretending this is actual reasonable argument.

   Anthropic is running out of money and wants to be bailed out by the US taxpayer.

2. I think you're mistaken about winner-takes-all. The only way whoever gets to superintelligence and self-improvement is unstoppable is if they also have unlimited computational resources. Assuming they don't everybody else that comes in second or third will be able to catch up as long as the leader does not have unlimited computation.

3. What i don't understand is the implication that there is an endpoint or a finish as if there is some expiry date. for instance if china is ahead in 2028 does this mean usa will just throw their hands up in the air, walk away and give up? maybe i'm missing something.

4. Monopoly is the only goal in capitalism, not free market or competition.

5. sure, they should share their models openly just like chinese do, then we support us models

6. Problem is, the CCP won't follow any security standards, at least not behind closed doors.

7. “America and it’s allies…”
   Do you mean Russia? Because every one else is decoupling as fast as possible from the US.
   “…authoritarian governments…”
   Buddy, wake up, look around and smell the facisim 🤦‍♂️

8. The government that spies on all other governments, bomb any country at will and export weapons, contamination and poverty, wants us to be scared or AI falling in the hands of bad actors 🙄

9. I think this is typical paranoid American projection. China doesn't want to dominate the world with AI, nor does it want to dictate to the US what to do with American AI. China wants to use their own AI for their own needs and, and if people want to use it or not is up to them. It's in fact the US mindset that unless they dominate everything and everyone then the world is falling. This is due to their desire to try and preserve their hegemony. China is all about multipolarity and multilateralism and very much against hegemony, and bullying by the US or any other country.

   You'd do better to invite some people knowledgeable on China who actually live/lived there to explain their system to you because there is a lot of western ignorance due to propaganda. I mean just take your own words which just echo Trump's mindset: "you have to have leverage to dictate to the other party what to do". Who's the authoritarian here? It's not China, but the US. This notion clearly stems from an attitude of American exceptionalism, which is baloney. The US needs to move past their hegemonic inclinations, and try diplomacy based on mutual interest and respect.

10. Do you really believe the US is a democracy, the "good guys" 😅

11. We already lost that race. Also the US winning is no better than China. Heard of Palantier?

12. Anthropic’s more right than the cynical commenters

13. Genuine question here! I am a business administration major, but I am considering switching to a computer science career (I am taking a few edX courses and will take more legit courses at a university). I am aiming toward a master in computer science. Do you think it is worth considering in 2026 with AI that will soon automate research across all major STEM fields? My motivation is to not only vibe-code, but also to understand how AI systems create and manage code and I think this knowledge is crucial. However, given the fast pace at which AI is advancing, I fear that having this kind of knowledge will be obsolete by 2028. Does someone have an idea?

14. China's desire for tech self-sufficiency is an act of aggression simply because it challenges U.S. dominance? 🤔

15. The solution is the second amendment. Everyone must be able to run AI in their own PC.

16. Loved this one, Matt!

17. Oh yes, sure, let’s all just use closed-source American AI. Let's put our startups completely at the mercy of a few corporate gatekeepers. Build your entire business on their tech, only to get randomly banned by Anthropic or OpenAI because some automated bot or people decided you violated their arbitrary terms of use, because you are not using it the way, they want you to; or at the rate at which they allow you to. What’s more authoritarian than that?
    Open-source AI is the only stack you can actually rely on. Go ahead and use closed models to get the job done for now, but keep an open-source stack active and ready. It's your only real insurance policy. And China has greatly helped open source, stay in the competition, not America.

18. If you give me more money, I can build a bigger hot air balloon to get to the moon.
    I don't believe LLMs will become intelligent.
    For the super-intelligence chart, a toddler is self-improving.
    LLMs are not self-improving. They do not adjust their weights dynamically. They do not continually learn.

    Anthropic trying to scare people about China or Mythos is to get more money.
    Similarly, for Nvidea saying they need to supply more chips. It is just to make more money.

    The AI investment is making things horrible. Not specifically the job losses due to new technology, but the secondary impact of taking investment from everything else to AI and the loss in jobs due to investment in AI. The higher cost to power as AI is consuming so much. The higher cost to all electronics as resources are focused on AI for profit.
    The large investors do not care who wins the AI race. They just get in and get out.

    If AI dominance is truly a national security issue, then why are all these US based companies not collectively working for the security interests of the US.

19. I like AI, I can't wait till I don't have to work, or even read a book, AI does everything for me. Then I can finally enjoy life and everything will be free

20. We only have 1 year until we see a major warning about us having only 2 years until we see a huge huge change that will revolutionize everything.

21. I believe WWIII is unavoidable.

22. AI is all about getting a foothold in nations to do massive surveillance. We've all been fooled.

23. Was that written by Claude? Looks like Anthropic is trying very hard to not be a "supply-chain risk" 😂

    AI self-improvement loops are completely overrated. The possibility space is too large for AI to navigate effectively (think DAG). Yes, we should strive for dominance. No, there's no 2028 deadline.

24. Hey you can say China without saying CCP, communist party Regime etc etc – we refer to other countries by their name!!
    Second- as for surveillance- USA is spreading Palantir around its vassal states and their surveillance is predictive and lethal.
    China uses surveillance for the public good – safe streets and traffic flow etc….

25. That is pathetic… Current architecture will never be AGI. To human brain you compare the current AI as paper plane to the space shuttle. The simplest no no is the dimensionaluity. Binery system you compare to a complex mnultidimensional human organism. It's just sad that so clever pelpe say such dumb things. But I understand you have to earn some money with the fud….

26. Could Anthropic simply be fearing competition from Chinese AI, and want the US govt to thwart Chinese companies advances, not at all for security reason but to continue commercial supremacy? I don't know, just wondering.

27. No disrespect to any of the brilliant individuals that have helped attempt to Shepherd in this technology, but one quick correction.z America is not nor has it ever in its entire existence Ben democracy… it was not one in 1776 and it is not one in 2026. Maybe somebody should actually go read the constitution we are a constitutional federal republic. That has never once changed.

28. I couldn't see the link to the video on what your exact thoughts on open source are at the end of the video…please send it to me

29. Blah, blah, blah. It's the same mindless AI slop and PR spin every day, making easy money from our "eyes on the screen". If AI is so good why aren’t the AI “experts” and influencers on their super yachts ? LOL. We know the truth. AI companies are shonks and grifters. AI can’t “do” maths, it doesn’t store knowledge, it can’t be used in law, it’s only right 50% of the time, its results can’t be verified, and it can hallucinate. The AI fraudsters and charlatans and their clueless PR fanboys and influencers will disappear. Hooray !

30. Are you dumb? All the stuff they are fear mongering about China is what our government is already doing to us. Also Anthropic is a drama queen, it's always them warning this and that

31. Trump's regime same as Chinese. No difference.

Comments are closed.