The Figure 01 robot can now have normal conversations

OpenAI models provide high-level visual and linguistic intelligence.

Figure's neural networks provide fast, low-level, dexterous robot actions  

OpenAI previously announced a partnership with startup Figure and implemented its models into their robots.

What powerful paws he has! Just like the terminator.

#news

⚡️ Nvidia has made an AI revolution - Uncle Huang has just triumphantly unveiled the world's most powerful Blackwell graphics card. It will be used to create a new generation of giant neural networks.

Here's everything you need to know right now:

🔸 Scale - one Blackwell is 1945 (!) times more powerful than a PS5! Huang suggests assembling the cards into clusters of 100,000 (!!!) gas pedals.

🔸 Blackwell is 5 times faster than the previous Hopper chip. It was released only a year ago.

🔸 To train a conditional GPT-4, it will take only 90 days and 2000 chips - previously it took 8000 chips.

🔸 The chip will be used by Google, Microsoft, OpenAI and Tesla.

Bottom line: in a couple of years ChatGPT will seem like an underdeveloped programm. If you want to know what the future will be like, imagine your wildest fantasy, you won't be wrong.

A paralyzed man demonstrates how Elon Musk's Neuralink chip allows his mind to play chess on a laptop

He simply uses his mind to control the cursor on the screen. In addition to chess Nolan Arbaugh played Civilization VI for about 8 hours.

In the future, the chip will help cure autism, schizophrenia and even obesity. And Elon Musk promises to implant it in his brain!

#news

Sam Altman discussed GPT-5, Sora, Ilon Musk, AGI, and aliens

In a recent discussion between Lex Friedman, Sam openly discussed a wide range of topics that are of great interest to many.

Let's walk through the important topics:

🤖 GPT-5: Sam Altman discusses the evolution and future of AI models, specifically looking forward to GPT-5. He expresses excitement about AI getting smarter as a whole, rather than by individual capabilities. Altman hints at work in progress and suggests that OpenAI will release several significant updates before officially moving to what could be called GPT-5, emphasizing a trajectory of continuous improvement rather than single leaps.

🎥 Sora: Altman shares his thoughts on Sora, an OpenAI project capable of generating video. He reflects on how Sora could change the content creation process and hints at broader implications for AI's ability to create increasingly complex simulations. Despite the technological breakthroughs, Altman remains cautiously optimistic, recognizing the challenges and ethical considerations involved in developing such technologies.

📑 Legal Dispute with Elon Musk: The conversation touches on the complex relationship between Sam Altman and Musk, especially in light of Musk's lawsuit against OpenAI and his criticism of it. Altman expresses respect for Musk's contributions to technology and hopes for a friendly relationship in the future, despite the current tensions and differences in their approaches to AI development and security.

💸 The $7 Trillion for AI Chip Project: Sam clarified that he did not tweet about wanting to raise $7 trillion for the chip project. However, he recognized that there is a significant need for more computing power. He noted that meeting this demand is a big challenge because of energy requirements, data center expansion, supply chain development, and chip manufacturing.

🧠 Predictions about AGI: Altman predicts that by 2030, and possibly sooner, we will see the development of extremely advanced AI systems. He believes that the emergence of AGI is not the end, but rather the beginning.

👽 Aliens: In a lighter moment, Altman and the host discuss the possibility of alien civilizations and how advances in AI can help humanity explore such deep questions. This speculative conversation highlights the broad curiosity and ambition that drives Altman's vision of AI's potential.

#news

All the people in this video aren't real

Arcads — a new AI that can generate short videos with astonishing realism.

Neural network is trained on facial expressions and gestures of actors, so that an unaware person will not notice the trick at all.

The text for the voiceover is made up by the user.

Interior design AI: Collov will update your interior design

Collov transforms the overall look of a room, or replace only the floor or walls. Also knows how to change all fittings, chandeliers etc and any item you highlight.

#service

Do you enjoy reading this channel?

Perhaps you have thought about placing ads on it?

To do this, follow three simple steps:

1) Sign up: https://telega.io/c/learn_ai_artificial_intelligence
2) Top up the balance in a convenient way
3) Create an advertising post

If the topic of your post fits our channel, we will publish it with pleasure.

HeyGen has an interesting update - Avatar in Motion.

Now the model can work not only with static videos, but also with those where you are walking down the street and actively desticulating.

By the way, you definitely didn't notice this, but the original video was in German, and the English you hear is the translation.

This AI generates songs based on your cringe (and beyond) lyrics. The tracks are powerful!

Instead of a prompt, you can write anything you can think of - one generation takes a couple of minutes. For fine-tuning you can specify style, voice timbre and reference - neuronka will understand you.

Eminem, look for a new job, we have a AI here 👈

#service

👍Rows is the only spreadsheet with the capabilities of an AI Analyst and a native OpenAI integration. Both are available to everyone and free to use. What makes the AI features in Rows special is: The combination of the spreadsheet interface with the AI capabilities. This means that you can leverage AI to drag automations across ranges of cells, combine the AI actions with any other spreadsheet formula, and use cell references to create dynamic AI experiences. The ability to use AI to summarize and help you make sense of your data.

#Rows⚡️ #OpenAI⚡️ #Feature

SocialClone is a neural network that will make a human clone and create content with its participation

In order to create a "doppelganger", only one video is needed. The service can add images, music, subtitles, etc.

Unfortunately the neural network is paid.

#service

🚀MIT launches Working Group on Generative AI and the Work of the Future
New initiative is convening leading companies and nonprofits with support from Google’s Community Grants Fund.

The world counts on MIT to turn sophisticated ideas into positive impact for the good of society,” says MIT President Sally Kornbluth. “This working group is focused on doing exactly that: In the face of broad public concern about AI’s potential to eliminate jobs, they are developing practical strategies for how to use generative AI to make existing jobs better and improve people’s lives.

The working group has already interviewed or engaged with more than 40 companies. Working group members include Amsted Automotive, Cushman and Wakefield, Cytiva, Emeritus, Fujitsu, GlobalFoundries, Google Inc., IBM, Liberty Mutual, Mass General Brigham, MFS, Michelin, PwC, Randstad, Raytheon, and Xerox Corp.

#Network⚡️ #AI⚡️ #AInews

Create a beautiful landing page with AI in seconds.
🫶Sitekick simplifies the process, empowering you to craft visually stunning pages that convert - no coding, design, or copywriting skills required.

#Network⚡️ #AI⚡️ #Feature

DIKTATORIAL Suite - AI edits music

A neural network for mastering audio using your promts. Simply put, a virtual sound engineer who brings "raw" material to life by stylizing it for musical genres and adjusting it.

Results based on your inputs are generated in a minute. Everyone has 10 credits available at the start for free audio generation. If you need more, there's a subscription, now discounted (you can grab it for $3 a month).

#service

A new AI: DBRX, which beats all free competitors, has been dropped into open access

See for yourself on the graph: even Musk's unapologetic Grok loses.

DBRX can write code, solve math problems and more. You can run it for free from your browser without registration - here

#service

CreatieAI - doing design with AI

A convenient and comfortable space for working with AI: select areas of design you have problems with, and the neural network itself will suggest design ideas suitable for that particular fragment.

Also here you can upscale images and do a lot of things for free that are not available in Figma, for example. This is a case where the integration of neural networks into a platform for designers has created a really cool service.

#service

It looks like Sora finally has a competitor - Higgsfield AI announced a new model.

Of course, it's still a long way from Sora, but it's better than available models like GEN-2 and Pika.

The waitlist is here.