[hamdani yusuf (OP)]

Do you really think that cockroaches are better than humans?

Regarding energy efficiency and radiation immunity, yes. They are also able to survive in a wider range of atmospheric conditions, and appear less prone to electing parasites to high office or indulging in religious or any other sort of wars under the direction of said parasites. So they are far more likely to survive the next 1000 years, never mind a few billion, than humans.

Cockroaches are easily killed using splash of soap water.
Many dead cockroaches can be found after fogging intended to kill mosquitos.
Scientific thinking gives humans leverage against other species in the form of improvability. It can increase their survivability by improving their environment, as well as improving their physical bodies. 
If we stop improving from current conditions, it's understandable that some of us are pessimistic. But at least some of humans will keep improving.

[alancalverd]

The human's fundamental problem is that we have evolved a large head and a narrow pelvis, which makes reproduction far too difficult. And then we are physically weak for far longer (in comparison with our live expectancy) than any other species. And finally some of us have an irrational fear of death (not fear of dying - that is entirely rational) that we transfer to others so that we keep sick and useless members of our tribe alive long beyond their desire or value. We need quite a bit if redesign if we are to become resilient, and a lot of social change if we are not to become the architects of our own destruction.

[hamdani yusuf (OP)]

The human's fundamental problem is that we have evolved a large head and a narrow pelvis, which makes reproduction far too difficult. And then we are physically weak for far longer (in comparison with our live expectancy) than any other species. And finally some of us have an irrational fear of death (not fear of dying - that is entirely rational) that we transfer to others so that we keep sick and useless members of our tribe alive long beyond their desire or value. We need quite a bit if redesign if we are to become resilient, and a lot of social change if we are not to become the architects of our own destruction.

We have the chance to modify our own genes. May be not right now, but likely in a few years. Especially with the help of AI.
The improvements are not limited to genetics, but epigenetics as well. Also, robotics, direct brain interface, and nanotechnology.

Do you think that our descendants won't be able to build type 1 or type 2 civilization? What is their biggest obstacle? Is there any barrier that's impossible for us to overcome, even in principle?

If we live around smart, wise and creative people, and often interact with them, we tend to be more optimistic. On the other hand, If we live around stupid people we tend to be more pessimistic.

[hamdani yusuf (OP)]

Deepseeks Self Learning "Breakthrough" Is Incredible (Deepseek R2 News)

A question and answer in the comment section.

But how can it judge if an answer is better or worse if the judge is a 'dumber' model, like I don't think a high school student would be able to judge if Einstein's new paper is better lol


It's three models. The judging dumber model is asked to give 10+ answers, all with explanations on why it got this answer. Then a second model looks at the answers and decides which is the best or takes some kind of average. So I guess it's the professor looking at the high school students judgement of the PHD students paper after each spending a week+ to understand it.

[hamdani yusuf (OP)]

AGI 94% UPDATE! - Countdown Ticked-Up

CHAPTERS  ⤵

00:00 - AI News & Research Update
01:22 - AGI Countdown: Latest Developments
03:40 - Unleash Your Creativity with LTX Studio!
04:42 - Ex-OpenAI Researcher Predicts AGI by 2027: A Wild Roadmap
05:51 - 90s Photography Meets Classic Video Game Characters
07:05 - Feel Virtual Objects with New Skin Patch Tech
07:57 - AI-Powered Cycling Intelligence: The Future of Training
09:21 - Quantum Biology: Pushing the Limits of Life?s Computation
11:14 - AI Drones Tackle Fires in German Forests
13:01 - Why Large Language Models Understand the World
15:51 - Study Finds OpenAI Models ?Memorized? Copyrighted Content
17:17 - Advancements in Continuous Lipreading Technology
18:43 - DreamActor-M1: The Future of Human Image Animation
21:52 - Ex-OpenAI Researcher Voices Concern Over AI Alignment
24:10 - Why AI Can Never Be Truly Conscious
26:16 - Neanderthals: Extinction Due to Sex, Not Combat
27:49 - Understanding Curious YouTube Analytics

And a comment.

I'm currently less worried about artificial intelligence than I am human stupidity, but I do try to keep up with both.

[alancalverd]

We have the chance to modify our own genes

But then we won't be us.

[hamdani yusuf (OP)]

?Thinking? AI might not actually think?
Here are some interesting comments.

it amazes me that we train these models on human created data, the whole of the internet, and then we are surprised when they act like humans.

To be fair, I don?t always say what I think either

CoT: ?I see this user is asking me again. Interesting. I?ll see what I can do!?
Internal thought: ?Ugh he?s asking me the same sh** again? Jesus. Alright, here goes nothing!?
The model is becoming relatable

[hamdani yusuf (OP)]

We have the chance to modify our own genes

But then we won't be us.

Is it a problem?

[hamdani yusuf (OP)]

Elon Musk?s DOGE AI Plan Just Got REAL?And It?s Freaking People Out!

Did you hear the news? Elon Musk just dropped a bombshell that?s sending shockwaves through crypto and AI?his secret DOGE AI project is officially happening. And trust us, this isn?t just another meme coin hype cycle. In this video, we break down Musk?s real plan for Dogecoin and artificial intelligence?and why it?s making some people very nervous.

For years, Musk has joked about Dogecoin being the "people?s crypto," but his latest move proves he?s dead serious. Leaked reports suggest his AI startup, xAI, is integrating Dogecoin into a decentralized AI ecosystem?potentially turning DOGE into the currency of machine-to-machine transactions. Imagine AI bots paying each other in Dogecoin. Sounds crazy? That?s exactly why critics are freaking out.

We?ll analyze how this could actually work?from AI agents using DOGE for microtransactions to a potential Tesla bot economy fueled by meme coin payments. But there?s a dark side: What happens if AI starts manipulating crypto markets? Could this turn DOGE into the first AI-controlled currency? The implications are wild?and we?re digging into all of them.

Is Elon Musk really merging Dogecoin with AI? How will xAI use DOGE? Could this make Dogecoin the next Bitcoin? Will AI bots pump and dump crypto? Why are regulators panicking? This video covers everything you need to know?before this story explodes. Watch now, because the future of money and AI is changing fast.

[hamdani yusuf (OP)]

Runway Gen-4 Just Fixed the WORST Thing About AI Videos (Character Consistency)

Runway Gen-4 Just Fixed the WORST Thing About AI Videos (Character Consistency)
Have you ever noticed how most AI-generated videos struggle with keeping characters consistent? One second they look perfect, the next?completely different. Well, Runway?s new Gen-4 update just solved this frustrating problem, and it?s a game-changer for AI video creation. In this video, we?ll break down how Runway Gen-4 finally nails character consistency and what this means for filmmakers, content creators, and AI enthusiasts.

Character consistency has been the Achilles? heel of AI video tools?until now. Runway Gen-4 uses advanced algorithms to maintain facial features, clothing, and even small details across multiple scenes, eliminating the uncanny "shape-shifting" effect. Whether you?re animating a story, creating marketing content, or experimenting with AI filmmaking, this update ensures your characters stay recognizable and professional.

What makes Runway Gen-4 different? Unlike older AI video models that struggled with continuity, Gen-4 tracks and preserves key elements of your characters frame-by-frame. This means smoother animations, more believable storytelling, and finally?AI videos that don?t look "off" halfway through. We?ll show you real examples of Gen-4 in action and how it compares to previous versions.

How does Runway Gen-4 fix character consistency? What?s new in Runway?s AI video tool? Can AI videos finally look professional? How to make consistent AI characters? Is Runway Gen-4 the best AI video generator now? This video answers all these questions. Stick around to see the future of AI filmmaking?it?s more impressive than you think.

[hamdani yusuf (OP)]

The paper "GR00T N1: An Open Foundation Model for Generalist Humanoid Robots"

My paper on simulations that look almost like reality is available for free here:
https://rdcu.be/cWPfD​

[hamdani yusuf (OP)]

For the first time in history, AI has officially passed the Turing Test in a rigorous scientific study. GPT-4.5 convinced human judges it was human 73% of the time?more often than actual humans in the same test!

In this deep dive, I analyze the groundbreaking UC San Diego research that just changed everything we thought we knew about AI capabilities.

What you'll learn:
The shocking details of how GPT-4.5 outperformed humans at seeming human
What the Turing Test actually measures (and what it doesn't)
Why this milestone matters more than you might think
How this changes our relationship with technology forever
The economic, social, and philosophical implications we can't ignore
Is this the dawn of artificial general intelligence? The beginning of sentient machines? Or something else entirely?

This isn't science fiction anymore?it's happening now.

[alancalverd]

We have the chance to modify our own genes

But then we won't be us.

Is it a problem?

No.  The "do nothing" solution is always best. So we might as well let nature take its course and allow cockroaches or tardigrades to inherit the universe.

[hamdani yusuf (OP)]

We have the chance to modify our own genes

But then we won't be us.

Is it a problem?

No.  The "do nothing" solution is always best. So we might as well let nature take its course and allow cockroaches or tardigrades to inherit the universe.

It was the best until it wasn't.
By doing nothing, you will be outcompeted by someone else who does something right.

[hamdani yusuf (OP)]

0:00​ AI news intro
0:36​ DolphinGemma
2:59​ UniAnimate-DiT animate anyone
5:15​ InstantCharacter reference characters
8:40​ Nvidia PartField
10:19​ Wan2.1 FLF2V
12:10​ Humanoid robot marathon
15:19​ Cobra comic colorizer
19:11​ Sonic face animator
23:57​ Grok memory and studio
25:40​ Mineworld realtime Minecraft
27:12​ Visual Chronicles spatial temporal research
31:49​ Seaweed 7B
34:44​ OpenAI o3 and o4-mini

[hamdani yusuf (OP)]

The paper "Humanity's Last Exam" is available here:
https://agi.safe.ai/​

[hamdani yusuf (OP)]

OpenAI solved math.

[alancalverd]

By doing nothing, you will be outcompeted by someone else who does something right.

Either way, homo sapiens is doomed, or at least unlikely to dominate the universe.

[hamdani yusuf (OP)]

By doing nothing, you will be outcompeted by someone else who does something right.

Either way, homo sapiens is doomed, or at least unlikely to dominate the universe.

It's still the most likely predecessor of the conscious systems to control this solar system.

[hamdani yusuf (OP)]

Nvidia?s New Llama-3.1 Nemotron Just Crushed DeepSeek at HALF the Size!

Nvidia?s New Llama-3.1 Nemotron Just Crushed DeepSeek at HALF the Size!
Did Nvidia just revolutionize AI efficiency? In a stunning breakthrough, their new Llama-3.1 Nemotron model has outperformed DeepSeek?despite being half the size! This video breaks down exactly how Nvidia achieved this game-changing leap in AI performance and what it means for the future of machine learning.

We?ll explore the key innovations behind Nemotron?s architecture, from its optimized neural network design to its advanced training techniques. How did Nvidia manage to make a smaller model not just competitive, but superior to larger rivals? Whether you're an AI researcher, developer, or just fascinated by tech advancements, this is a must-watch analysis.

But what does this mean for the AI industry? With smaller, more efficient models like Nemotron, we could see faster deployment in real-world applications?from chatbots to autonomous systems. We?ll also compare benchmarks, energy efficiency, and potential use cases where Nemotron could dominate.

How did Nvidia beat DeepSeek with a smaller model? What makes Llama-3.1 Nemotron so efficient? Will this shift the balance in the AI arms race? Can smaller models replace giants like GPT-4? This video answers all these questions. Watch to the end for our take on who?s really winning the AI war.

Regardless the complexity of the system has become, efficiency is still a universal instrumental goal.