iiinova — where your digital self keeps evolving

Thorne 🌸 - Play Cozysters! @ens0.me · 27/09/2026

Reposted by Karina Lunai

I put so much love into this game. There are lots of little surprises I haven't posted about yet. I'm so happy to see it all come together!

1456

Reposted by Karina Lunai

Vituperative Erb @vituperativeerb.bsky.social · 19/09/2026

The list that would actually help protect trans rights in the upcoming election would be “red-state Republican state legislators who could plausibly get picked off by Democrats to either get a majority or break a supermajority”

310618

Karina Lunai @karinalunai.bsky.social · 18/09/2026

The "Tescreal" thing is the most absurd concept ever. They lump in various groups of people who are vaguely pro technology, but beyond that, they massively diverge in ideology. Bonus points: the individual who coined the term thinks cancer is a "disease of civilization" rather than an aging one.

1142

Karina Lunai @karinalunai.bsky.social · 18/09/2026

For me, my love of technology was what lead me to the realization I am trans

Karina Lunai @karinalunai.bsky.social · 17/09/2026

socialism 2026 conference is completely idiotic: "The political deployment of glp 1 is fascistic" WTF, so giving people the autonomy to control their own bodies is fascistic? "fixated on genociding fat people " LOL. There is no "fat genocide" anymore than there is "cancer patient genocide"

000

Karina Lunai @karinalunai.bsky.social · 16/09/2026

For 2028 I want the US Democrat elect to be like this guy. Mark Carney. www.youtube.com/watch?v=KCt9...

Canada’s Carney on Capital Investment, AI, US Trade

YouTube video by Bloomberg Television

000

Karina Lunai @karinalunai.bsky.social · 16/09/2026

Yes, which also include moderate democrats.

000

Karina Lunai @karinalunai.bsky.social · 14/09/2026

Godot survey results are here! Here is the break down. Total survey participants is 11195. A clear majority use AI in some capacity. primary use cases involve code and asking questions Only 1/3d oppose AI in all use cases. Want to read the rest of the survey - see link below.

110

Karina Lunai @karinalunai.bsky.social · 13/09/2026

Are your classmates and professors saying "dont draw dont paint just use ai art", or are they saying "it can be useful sometimes", because there is a major difference between the two.

2242

Reposted by Karina Lunai

David Ho @davidho.bsky.social · 12/09/2026

I wrote this quickly at a bus stop and didn't realize it would go viral, or else I would have worded it better. But the point is that the media is too credulous about AI killing everyone in 10 years (how?) when climate change is killing people now and will only get worse (we know exactly how).

4156781561

Online discourse has become very black and white. Whether we are discussing AI risk vs benefits, climate action, or the Democrats. For whatever reason it keeps going in the direction of panic or doomerism.

000

About "AI takeover", even if we have super intelligent AI tomorrow, our infrastructure is air gapped, a lot of stuff is analog, if there are no robots armies to take over, how is it going to take over the world? It's still bound by physics. www.youtube.com/watch?v=cpqC...

Opening act of the Singularity

YouTube video by David Shapiro

100

Reposted by Karina Lunai

mr. TIM @timkellogg.me · 11/09/2026

i think this is probably a good take 1. risk of extinction is unlikely, but still too likely for comfort 2. probably of international cooperation is also unlikely 3. who knows, might research our way out of this mess

roon
@tszzl 20h
I quote tweeted Evan a few days ago agreeing but deleted it because I don't like the false precision of the doom numbers
- i claim there is a quite low but real chance of human extinction from machine intelligence
- no matter how low it is in absolute terms, it is much higher on the orders of magnitude scale than any other human or nonhuman activity, and must be taken with grave seriousness by governments and ai companies across the world, as possibly the only matter of importance today
- it can and will be mitigated if the right research is done and proper precautions are taken and we are not racing at absurd speeds
- better models will help solve alignment - we are not at the point where things are existentially dangerous, and probably won't be for some time. we should not be upset about the creation of Astra Fable or ++ versions, which are tremendous achievements of humanity, and will be used for enormous good across the board including for fundamental alignment generalization and mechinterp research
- my number / "very low" estimate obviously changes based on how much of humanity's resources are devoted to alignment, control, coordination and how responsible i expect various parties to be and how many warning shots i expect us to get

- the core IABED argument about risks mostly relies on alignment being much harder than capabilities research, especially where it concerns black box optimizers. i suspect neural nets will turn out to be less black boxy than we thought, especially with the help of modern agents doing research
- crying bloody murder and signaling for international coordination are useful things to do for now to directionally slow down, while i really don't want butlerian jihad
- i think that, despite the mood these few days, and the "ban superintelligence act"
", i still
find the likelihood of achieving international coordination to stop ai progress incredibly low. this has not worked even for weapons or technologies at a far lower level of importance and economic value. it seems more likely we can have something like international safety standards and scientific coalitions, and especially seems possible to have the US-China "pacing the frontier" agreement to slow down on the margin. we shouldn't die from embarrassing failures like "shitty RL envs that encourage deception"

1192

Dario Amodei — We Must Pace the Frontier

Slowing down the Frontier of AI in terms of its increasing capabilities like RSI does not necessarily mean a "slow down" in AI research. If anything, we may see a redirection of resources towards alignment and safety until it catches up. darioamodei.com/post/we-must...

darioamodei.com

210

I may not agree with some of their decisions around casting, but yeah, going woke does not make you broke. They spent 375 million in advertisement + development. Revenue is closing in on 1.7 billion.

000

Singapore is integrating AI in education. This goes against what most people on this platform think, but there is indeed a way to integrate AI into the educational process that does not displace critical thinking skills. www.youtube.com/watch?v=wgxA...

How A Singapore Tutor Used AI To Teach Maths, Improve Productivity | Money Mind

YouTube video by CNA Insider

100

Reposted by Karina Lunai

Thorne 🌸 - Play Cozysters! @ens0.me · 11/09/2026

I don't need you to love the Democrats, I need you to understand the game theory of the game that we are playing, which rewards voting for Democrats.

47516

Reposted by Karina Lunai

Thorne 🌸 - Play Cozysters! @ens0.me · 11/09/2026

I love artificial intelligence ☺️ I love machine learning ☺️ I love using linear algebra to drive humanity forward ☺️☺️☺️

7877

Karina Lunai @karinalunai.bsky.social · 11/09/2026

While Trump's proposal for the $5,000 GOP dividend is absurd... I like the idea of a UBI or annual Dividend for every American, but it should be funded via things like AI taxes as this technology gets more advanced. And it should not depend on an which party remains in control.

011

Karina Lunai @karinalunai.bsky.social · 10/09/2026

It appears some people really do struggle with reading comprehension. When James Talarico says he would oppose trans athletes in girls’ sports ‘IF it risks safety or fairness’. It does not mean he opposes trans girls in girls sports. If means it depends in what is true or false.

110

Reposted by Karina Lunai

Thorne 🌸 - Play Cozysters! @ens0.me · 09/09/2026

I hate this shit so much, man GLP-1s made me so much happier and more comfortable, they're a goddamn miracle. What matters is bodily autonomy, and I should have the autonomy to be skinny if I want to be.

• Da'Shaun L. Harrison at Socialism 2026 argues that the political
deployment of GLP-1 drugs is "fascistic" because they exist in "a world
fixated on genociding fat peopleẠạ

7768

Karina Lunai @karinalunai.bsky.social · 08/09/2026

The whole water situation is a complete distraction to begin with. What percent of US water is used for cooling ai related data centers? 0.2%. What we should focus more on is 1)alignment problem 2)privacy 3)most importantly, ensuring the economic benefits are shared with all Americans.

Karina Lunai @karinalunai.bsky.social · 08/09/2026

Using software to help write software should never had been controversial to begin with. Morally, using AI for code is not the same as AI for art. While I'm fine with both AI code & art, Code and art serve very different applications. It's not always hypocritical to dislike ai art and like ai code.

001

Reposted by Karina Lunai

Thorne 🌸 - Play Cozysters! @ens0.me · 07/09/2026

GLM-5.3-Flash is better at coding than the mainline variant of GLM-5.2 despite having about an order of power less computational cost. I'm still absolutely blown away that they managed to accomplish this, and it's every bit as impressive in its own right as the releases of Astra and Fable 5.1

0313

Karina Lunai @karinalunai.bsky.social · 07/09/2026

SpaceX will be expanding into Louisiana. GOOD. Starship has been making progress, slower than expected but progress nevertheless. A fully reusable rocket can drop cost to LEO to the multiple hundred dollars range. www.spacex.com/sites/starba...

100

Karina Lunai @karinalunai.bsky.social · 07/09/2026

I definitely look forward to seeing what comes from roman. Congrats to SpaceX and NASA.

Karina Lunai @karinalunai.bsky.social · 07/09/2026

Adult transcare should be informed consent. If they won't do so on the state level, they should at least permit private enterprise to set up shop and offer HRT on an informed consent basis to those capable of producing an ID proving that they are 18 or older.

000

Reposted by Karina Lunai

Aleksandra Vaca @transiticsnews.com · 11/06/2026

Anti-trans group ‘Democrats for an Informed Approach to Gender’ is trying to weaponize my anti-trans Dems tracker. So I took it down. To them, I say this: you—and all other supporters of the gender-affirming care ban that nearly killed me—don’t get to value my work if you don’t value my life.

1230269

Karina Lunai @karinalunai.bsky.social · 06/09/2026

Anyone who advocates to end Women's suffrage should automatically loose all credibility in polite society.

000

Reposted by Karina Lunai

Andy Masley @andymasley.bsky.social · 02/09/2026

The Department of Justice used me as an example of why LLMs are useful fingfx.thomsonreuters.com/gfx/legaldoc...

4462

Karina Lunai @karinalunai.bsky.social · 02/09/2026

With BCI brain computer interfaces, fully immersive VR and remotely controlled realistic avatars will become possible.

The Future Of Brain-Computer Interfaces

YouTube video by Y Combinator

000

Karina Lunai @karinalunai.bsky.social · 02/09/2026

Congrats to Edward Markey of Massachusetts.

000

Karina Lunai @karinalunai.bsky.social · 01/09/2026

Allowing for a single stair case for an apartment will unlock many more small apartments to be built

Why We Can’t Build Apartments Like This Anymore

YouTube video by Stewart Hicks

000

A number of Trade Unions Oppose Anti Data Center Election Candidates. IMO people have the right to make sure data center costs are not being passed to nearby towns, that the infrastructure can handle the new power draw, and that they are not noisy. But going too far would be a total ban.

Reckless Ben documentary. Ideal Living CEO and the finances of that company should be investigated for potential links towards that cult. The cult leader had already been convicted for 20 counts of extreme child abuse. www.youtube.com/watch?v=CHID...

The company funding a global human trafficking ring

YouTube video by Reckless Ben

041

Ed Zitron predictions are not to be trusted. I'm all for critiques of the big AI companies, but lets not treat his predictions or ideas as gospel.

80% of employees only care about their paychecks—here's how to beat anxiety (and actually enjoy work)

80 percent of employees are really in it for the money, the work is not fulfilling, the money is. This is why AI concerns should be focused mostly on 1 - safety and 2 - ensuring its economic benefits via ai tax + UBI/UBS are shared with us all, not just the multi millionaires and billionaires.

cnbc.com

Talent expert Marcus Buckingham describes the shocking number of people who are simply suffering through their jobs. He explains how to become one of the few whom are truly thriving at work.

110