Please open Telegram to view this post
VIEW IN TELEGRAM
β€8π4
Please open Telegram to view this post
VIEW IN TELEGRAM
π8β€6π6π2π₯1
This media is not supported in your browser
VIEW IN TELEGRAM
Sam Altman says AI will blur the gap between passive watching and active creating
Instead of just leaning back for a movie or sweating through a game, you'll engage mid-way β shaping what unfolds. Sora shows the shift: most users are now creating, not just watching
π€ @techdaily
Instead of just leaning back for a movie or sweating through a game, you'll engage mid-way β shaping what unfolds. Sora shows the shift: most users are now creating, not just watching
Please open Telegram to view this post
VIEW IN TELEGRAM
β€8π2π₯1
Please open Telegram to view this post
VIEW IN TELEGRAM
π8π4π₯3
OpenAI and Broadcom have announced a new multi-billion dollar partnership.
The collaboration aims to develop and deploy 10 gigawatts of custom AI chips and computing systems over the next four years.
π€ @techdaily
The collaboration aims to develop and deploy 10 gigawatts of custom AI chips and computing systems over the next four years.
Please open Telegram to view this post
VIEW IN TELEGRAM
π5β€3π₯1
This media is not supported in your browser
VIEW IN TELEGRAM
Learning hacking has never been easier. Found an interactive tutorial specifically designed for web developers.
In this tutorial, youβll be given real vulnerable applications to hack in your browser. Youβll then learn how to patch these holes in the code, effectively learning both hacking and protection simultaneously. Youβll also get to see all the current methods of hacking and protection.
Get it here.
π€ @techdaily
In this tutorial, youβll be given real vulnerable applications to hack in your browser. Youβll then learn how to patch these holes in the code, effectively learning both hacking and protection simultaneously. Youβll also get to see all the current methods of hacking and protection.
Get it here.
Please open Telegram to view this post
VIEW IN TELEGRAM
π5π₯2
Media is too big
VIEW IN TELEGRAM
James Cameron believes the soul of cinema lies in human performance, not technology, and warns that generative AI must be constrained to protect that sacred human core of filmmaking.
π€ @techdaily
Please open Telegram to view this post
VIEW IN TELEGRAM
π7β€4π4π₯1π€£1
Ray Kurzweil:
"By roughly 2032 we will reach longevity escape velocity...
In the 2030s, robots the size of molecules will go into our brains, noninvasively, through the capillaries, and will connect our brains directly to the cloud...
By 2045, once we have fully merged with AI, our intelligence will no longer be constrained β¦ it will expand a millionfold, this is what we call the singularity"
Society is not ready, but the future is coming, fast we are already on the exponential
π€ @techdaily
"By roughly 2032 we will reach longevity escape velocity...
In the 2030s, robots the size of molecules will go into our brains, noninvasively, through the capillaries, and will connect our brains directly to the cloud...
By 2045, once we have fully merged with AI, our intelligence will no longer be constrained β¦ it will expand a millionfold, this is what we call the singularity"
Society is not ready, but the future is coming, fast we are already on the exponential
Please open Telegram to view this post
VIEW IN TELEGRAM
β€7π«‘7π2 2π₯1
JPMorganβs global head of sustainable solutions said the U.S canβt power its AI dreams and tech expansion without renewables - period.
Meanwhile, Trumpβs still calling wind turbines βpatheticβ and renewables a βjokeβ - as JPMorgan throws $1.5 trillion at the future he thinks is imaginary.
Turns out, you canβt drill your way to AI dominance.
Please open Telegram to view this post
VIEW IN TELEGRAM
π8β€2π₯2
Microsoft just dropped its first in-house image model and itβs already in the top 10 on LMArena.
Meet MAI-Image-1: A Fast. Photorealistic. Purpose-built model for creators.
It nails lighting, reflections, textures and beats larger, slower models in both speed and realism.
Trained with pro-creator feedback and rigorous data curation to avoid generic βAI artβ vibes. Coming soon to Copilot and Bing Image Creator.
Available now on LMArena β go try it and see why itβs already ranking with the best.
π€ @techdaily
Meet MAI-Image-1: A Fast. Photorealistic. Purpose-built model for creators.
It nails lighting, reflections, textures and beats larger, slower models in both speed and realism.
Trained with pro-creator feedback and rigorous data curation to avoid generic βAI artβ vibes. Coming soon to Copilot and Bing Image Creator.
Available now on LMArena β go try it and see why itβs already ranking with the best.
Please open Telegram to view this post
VIEW IN TELEGRAM
Please open Telegram to view this post
VIEW IN TELEGRAM
β€11π5π₯2
Media is too big
VIEW IN TELEGRAM
SpaceX has released a new video of their Starship tile factory, which currently produces over 1,000 tiles per day.
π€ @techdaily
Please open Telegram to view this post
VIEW IN TELEGRAM
π₯9π7β€3
Please open Telegram to view this post
VIEW IN TELEGRAM
β€10π5π₯3π2
Media is too big
VIEW IN TELEGRAM
German robotics company NEURA Robotics has created NEURA Gym: a large-scale, physical AI gym and training ground where hundreds of robots, including the humanoid 4NE-1, learn through real-world interactions. The company is collaborating with the Neuraverse ecosystem to create the world's largest repository of physical training data, promoting the application of humanoid robots.
π€ @techdaily
Please open Telegram to view this post
VIEW IN TELEGRAM
π5π«‘5β€4π₯3
Google is catching up with trends β YouTube is testing a Liquid Glass style interface. All this is to stay in style with the latest updates on mobile devices.
Other social networks are not far behind: Liquid Glass is being prepared for Instagram and WhatsApp. But Pavel Durov, as usual, was ahead of everyone β this feature, as you know, has already appeared in Telegram.
π€ @techdaily
Other social networks are not far behind: Liquid Glass is being prepared for Instagram and WhatsApp. But Pavel Durov, as usual, was ahead of everyone β this feature, as you know, has already appeared in Telegram.
Please open Telegram to view this post
VIEW IN TELEGRAM
π₯7π5π€£4β€2
What's new:
1. The chip has 10 GPU cores. Performance increase is about 30% compared to the M4.
2. Each block now has a built-in Neural Accelerator. Speed increase for local AI tasks is about 3.5 times.
This includes various photo processing, video generation, rendering, model launches, etc. Approximate breakdowns for different tasks (diffusion, CV, LM) can be found here.
3. Memory bandwidth is about 153 GB/s (+30% compared to the previous generation). Plus, energy efficiency has been improved, so it should no longer overheat from computations.
There is only one BUT: the price starts at $1599
Please open Telegram to view this post
VIEW IN TELEGRAM
β€11π4
This media is not supported in your browser
VIEW IN TELEGRAM
Amazing performance. Extraordinary battery life. Introducing the all-new 14-inch MacBook Pro with M5, delivering the next big leap in AI performance on the Mac.
π€ @techdaily
Please open Telegram to view this post
VIEW IN TELEGRAM
π4
A senior NATO official said about 80% of the critical parts in Russiaβs drones and modern weapons are made in China.
Since 2023, Beijing has supplied over $20 billion in components, tools, and materials.
π€ @techdaily
Since 2023, Beijing has supplied over $20 billion in components, tools, and materials.
Please open Telegram to view this post
VIEW IN TELEGRAM
π₯10π2β€1π1
Please open Telegram to view this post
VIEW IN TELEGRAM
π€£20π6π₯4β€1
Media is too big
VIEW IN TELEGRAM
The update is not revolutionary, but there are noticeable improvements in video generation control.
Main features:
β’ Control over narration and sound. The model now generates dialogues and ambient sounds better. This is promised to reduce the time needed for post-production.
β’ Flexible input data. Text, images, references, clips β can be combined and you can set style, start and end frames, continue scenes, and edit objects within the video.
Veo 3.1 is already available in Flow, Gemini API, and will soon appear in Vertex AI. Prices remain the same: 40 cents per second for the standard model and 15 for Fast. Videos are generated in 720p or 1080p, with durations up to 148 seconds when "stretching" scenes.
Please open Telegram to view this post
VIEW IN TELEGRAM
π7β€1
This media is not supported in your browser
VIEW IN TELEGRAM
The Yamaha MOTOROiD 2 is equipped with an innovative self-balancing system that allows the motorcycle to maintain stability in almost any condition. The device lacks a traditional handlebar; instead, it uses handles integrated into the body.
The bike can identify its owner by face and respond to their gestures. It also features an autonomous mode in which the MOTOROiD 2 moves, parks, and stands up from the kickstand on its own.
Please open Telegram to view this post
VIEW IN TELEGRAM
π₯22β€9 7π1