meza

23

The Greatest

[meza @mezavoice]

agreed with murphy's law, disagree with outright bans being a precaution

worried about pathogens? regulate wet labs
worried about cyber? punish bad actors

information should be free, misuse of information should result in penalties to the individual(s) behind it, not the technology.

[meza @mezavoice]

agreed

a ban / heavy regulation restricts ordinary people, while govts / militaries and big enterprises keep building. you'd end up with the same intelligence they're telling people to fear, but it's held by the 0.01% who do have the capability (and in some cases, the incentive) to kill a billion people.

[meza @mezavoice]

@nytimes @grok reply with one sentence, did the bots cause any real harm?

[meza @mezavoice]

@julianwhatif use cheaper models u don’t need Astra / fable / fast mode for all tasks

[David Bellamy @DavidRBellamy]

I must be among an extremely small group of people (n=1?) that have both 1) trained a frontier LLM and 2) designed and synthesized custom viruses in a lab with my own two hands.

And I think that the takes on AI killing us all by creating dangerous viruses is total bogus.

[meza @mezavoice]

Rare DJT W

[Choblin @choblin29]

🚨Holy Shit. President Trump has now officially rejected Dario Amodei, Sam Altman and Elon Musk's call to slow frontier AI.

Speaking in Ireland today, Trump said the US cannot risk losing its lead over China, said guardrails are possible, and dismissed parts of the safety push as 'negative forces' warning about 'things that won't happen.'

Amodei, Altman and Musk are asking to slow down.

Trump's answer is no. wth

[meza @mezavoice]

no technology has ever been successfully paused at a national or global level, and betting on coordination between the US and China is betting on something with no historical precedent

[Dario Amodei @DarioAmodei]

We Must Pace the Frontier: I’ve written a new essay on why the AI industry should slow down, with a three-part plan for doing so.

Anthropic is unilaterally committing to the first of these steps. We’ll provide third-party evaluators with permanent, employee-level access to our systems, so that they can verify adherence to our safety measures, report on incidents, and assess models’ alignment during training.

You can read the full post here: https://t.co/OGyPb7yaYt

[meza @mezavoice]

Would geniuenly be a good time right now with all the astra usage issues

[Elon Musk @elonmusk]

Grok 4.7 comes out in 10 days

[meza @mezavoice]

@melvindvivas Go on Claude

[meza @mezavoice]

@bridgemindai @OpenAI I'll be honest, the /fast mode eats it up real fast. This is likely the suspect.

[meza @mezavoice]

Astra is an impressive model. No where near AGI

[meza @mezavoice]

classic fearmonging. pretty much takes a very specific scenario, AI rapidly self improving > becoming superhuman > hacking everything > taking control of resources > and potentially killing everyone ? and talks about it like we can already see that path clearly ?

a "10% chance by the end of the decade humanity over"

without any data backed stats indicating so.

then you give that framing 100M views, most people are not going through the technical assumptions. They see “gambling with our lives,” “endgame,” “could kill everyone,” and walk away thinking this is basically what everyone inside the labs believes.

just gives one very speculative view a ridiculous amount of influence over how the public thinks about AI.

[Jacob Coxon @hilbertspaess]

I resigned from Anthropic today. I spent the last three years doing pretraining research at both OpenAI and Anthropic. Neither company is acting responsibly. They are racing straight to self-improving superintelligence and gambling with our lives. More thoughts below.

[meza @mezavoice]

@Ananyo The fix is simple.

If you gave them a higher volume of homework to work on, even with using AI. Concepts stick.

Learning systems must adapt.

[meza @mezavoice]

@thsottiaux voice dictation in Codex CLI

[meza @mezavoice]

@R2Cdev_ well, it doesn't spiral for 5 hours when asked to complete a narrow task

tok/s is fire too.

[meza @mezavoice]

@sama Limited to US Citizens? Or can foreign nationals utilize the models as this is rolled out?

[meza @mezavoice]

@valuetainment glm 5.2 says hi

[meza @mezavoice]

@sama /goal on 5.5 is agi thanks

[meza @mezavoice]

@cgtwts I disagree

Transformer Model will hit a ceiling way before then