“There’s nothing artificial about AI. It’s inspired by people. It’s created by people.
Everybody should care about AI because it's going to impact your individual life, your society and the future generation.”
~ Dr Fei-Fei Li
aipost 🏴
Everybody should care about AI because it's going to impact your individual life, your society and the future generation.”
~ Dr Fei-Fei Li
aipost 🏴
🔥 Boston Dynamics has just released a new video of its upgraded next-generation humanoid robot called Atlas.
• 4 hour battery. Self-swappable for continuous operation
• 6 feet 2 inches tall
• Weight: 198 lbs
• 56 total degrees of freedom
• Now fully electric, ditching older hydraulic systems
• New lightweight mix of aluminum and titanium components
• 110 lbs weight capacity (66 lbs sustained)
• Can reach up to 7.5 ft
• Constantly evaluates its surroundings and adjusts its posture, balance, and grip in real time
• Hands that can reconfigure as needed. Tactile sensors feed data back into the sy
• 4 hour battery. Self-swappable for continuous operation
• 6 feet 2 inches tall
• Weight: 198 lbs
• 56 total degrees of freedom
• Now fully electric, ditching older hydraulic systems
• New lightweight mix of aluminum and titanium components
• 110 lbs weight capacity (66 lbs sustained)
• Can reach up to 7.5 ft
• Constantly evaluates its surroundings and adjusts its posture, balance, and grip in real time
• Hands that can reconfigure as needed. Tactile sensors feed data back into the sy
2026: the year of mega tech IPO
SpaceX, OpenAI, and Anthropic are lining up potential IPOs as early as this year, with combined valuations that could exceed $1.5 trillion and raise tens of billions of dollars.
Even one listing could eclipse the entire 2025 US IPO market, making this a potential once-in-a-generation payday for investors, banks, and lawyers
aipost 🏴
SpaceX, OpenAI, and Anthropic are lining up potential IPOs as early as this year, with combined valuations that could exceed $1.5 trillion and raise tens of billions of dollars.
Even one listing could eclipse the entire 2025 US IPO market, making this a potential once-in-a-generation payday for investors, banks, and lawyers
aipost 🏴
🗣 Elon Musk said xAI will surpass everyone else in AI compute within 5 years, with the Macrohard-branded Colossus 2 data center serving as a clear reference to his AI project aimed at rivaling Microsoft.
Colossus2 is a Tennessee and Mississippi buildout that is already pushing past 400MW and aiming for about 2GW at 1 site, including outside power generation equipment to move faster than normal utility timelines.
Musk has ealier also claimed about 230,000 GPUs are already training Grok while targeting 50M “H100-equivalent” GPUs within 5 years.
Using Nvidia’s “up to 700W” H100 SXM spec, 50M H
Colossus2 is a Tennessee and Mississippi buildout that is already pushing past 400MW and aiming for about 2GW at 1 site, including outside power generation equipment to move faster than normal utility timelines.
Musk has ealier also claimed about 230,000 GPUs are already training Grok while targeting 50M “H100-equivalent” GPUs within 5 years.
Using Nvidia’s “up to 700W” H100 SXM spec, 50M H
⚠️ In US AI data centers hitting a hard energy wall.
Morgan Stanley research projects a 44 gigawatt power gap, that current grids and construction plans cannot fill. US data centers are expected to need 69 gigawatts of power between 2025 and 2028, but only 10 gigawatts are coming from projects with their own power and 15 gigawatts from spare grid capacity, leaving a 44 gigawatt shortfall.
That 44 gigawatts is roughly the output of 44 nuclear power plants and with $60B per gigawatt capex needed. Multiplying that by the 44 gigawatt gap gives about $2.6T just for new power generation and grid u
Morgan Stanley research projects a 44 gigawatt power gap, that current grids and construction plans cannot fill. US data centers are expected to need 69 gigawatts of power between 2025 and 2028, but only 10 gigawatts are coming from projects with their own power and 15 gigawatts from spare grid capacity, leaving a 44 gigawatt shortfall.
That 44 gigawatts is roughly the output of 44 nuclear power plants and with $60B per gigawatt capex needed. Multiplying that by the 44 gigawatt gap gives about $2.6T just for new power generation and grid u
Every generation gets the new technology that will finally transform education.
Radio. Television. Computers. The internet. Each came with the promise... https://www.post-gazette.c...
Radio. Television. Computers. The internet. Each came with the promise... https://www.post-gazette.c...
🗣Mustafa Suleyman says AI models are becoming a second brain that speeds up hypothesis generation
As models ingest more information, they become proactive, nudging you toward new insights based on your specific interests. "The better it learns you, the better you become"
aipost 🏴
As models ingest more information, they become proactive, nudging you toward new insights based on your specific interests. "The better it learns you, the better you become"
aipost 🏴
In this post, we explore how Mantle, Amazon's next-generation inference engine for Amazon Bedrock, implements a zero operator access (ZOA) design that eliminates any technical means for AWS operators to access customer data. https://aws.amazon.com/blo...
Today, SpaceX shared a new picture (on left) of its next-generation Raptor 3 rocket engine, the most advanced rocket engine ever made.
• Almost 2x the thrust of Raptor 1
• Costs 4x less
• Much lighter. Will save 2,425 lbs of weight per engine, or 94,575 lbs (42.9 metric tons)… pic.twitter.com/SrQNgKXDIV— Sawyer Merritt (SawyerMerritt) December 23, 2025 https://x.com/SawyerMerrit...
• Almost 2x the thrust of Raptor 1
• Costs 4x less
• Much lighter. Will save 2,425 lbs of weight per engine, or 94,575 lbs (42.9 metric tons)… pic.twitter.com/SrQNgKXDIV— Sawyer Merritt (SawyerMerritt) December 23, 2025 https://x.com/SawyerMerrit...
entropy reversal engine
prompt one: generation. prompt two: animation.
grok image prompt:
A photorealistic digital rendering of a heavily modified Tesla Roadster on a rain-slicked street in a cyberpunk Tokyo night scene, captured from a low-angle rear three-quarter view… pic.twitter.com/np8ftdWL7O— tetsuo (tetsuoai) December 20, 2025 https://x.com/tetsuoai/sta...
prompt one: generation. prompt two: animation.
grok image prompt:
A photorealistic digital rendering of a heavily modified Tesla Roadster on a rain-slicked street in a cyberpunk Tokyo night scene, captured from a low-angle rear three-quarter view… pic.twitter.com/np8ftdWL7O— tetsuo (tetsuoai) December 20, 2025 https://x.com/tetsuoai/sta...
🇷🇺 Russia's AI Defiance: Sanctions Fail to Slow Innovation
Western governments expected technological restrictions would push Russia to the margins of the digital economy. The reality at AI Journey 2025 tells a dramatically different story.
Sber, Russia's largest bank transformed into a technology powerhouse, showcased capabilities many thought impossible under sanctions:
• Large language models trained exclusively on domestic data
• Industrial-grade robotics systems ready for deployment
• New generation of "intelligent" devices built almost entirely on Russian tech stack
• Ecosystem demons
Western governments expected technological restrictions would push Russia to the margins of the digital economy. The reality at AI Journey 2025 tells a dramatically different story.
Sber, Russia's largest bank transformed into a technology powerhouse, showcased capabilities many thought impossible under sanctions:
• Large language models trained exclusively on domestic data
• Industrial-grade robotics systems ready for deployment
• New generation of "intelligent" devices built almost entirely on Russian tech stack
• Ecosystem demons
⚡️ OpenAI introduces GPT-5.2-Codex, its most capable agentic coding model to date
OpenAI has released GPT-5.2-Codex, its most advanced agentic coding model so far. Designed for complex, multi-step software engineering, not just code generation. It improves on earlier Codex models with stronger long-context reasoning, native code compaction, and more reliable tool use
What’s new technically
• Builds on GPT-5.2 (professional knowledge work)
• Extends GPT-5.1-Codex-Max capabilities in agentic coding and terminal execution
• State-of-the-art performance on SWE-Bench Pro and Terminal-Bench 2.0
C
OpenAI has released GPT-5.2-Codex, its most advanced agentic coding model so far. Designed for complex, multi-step software engineering, not just code generation. It improves on earlier Codex models with stronger long-context reasoning, native code compaction, and more reliable tool use
What’s new technically
• Builds on GPT-5.2 (professional knowledge work)
• Extends GPT-5.1-Codex-Max capabilities in agentic coding and terminal execution
• State-of-the-art performance on SWE-Bench Pro and Terminal-Bench 2.0
C
❗️Disney introduces DisneyGPT, an internal AI assistant for content teams
Disney has launched DisneyGPT, a proprietary neural network for internal use. The tool is designed to help studio employees work faster and improve content quality.
What DisneyGPT does
• Provides verified information on films and animated projects, including canon details.
• Answers character-related questions, such as backstories, motivations, and relationships.
• Helps maintain the “Disney spirit” by referencing Walt Disney’s quotes and creative philosophy.
• Supports creative workflows with idea generation, fact-che
Disney has launched DisneyGPT, a proprietary neural network for internal use. The tool is designed to help studio employees work faster and improve content quality.
What DisneyGPT does
• Provides verified information on films and animated projects, including canon details.
• Answers character-related questions, such as backstories, motivations, and relationships.
• Helps maintain the “Disney spirit” by referencing Walt Disney’s quotes and creative philosophy.
• Supports creative workflows with idea generation, fact-che
As data privacy collides with AI’s rapid expansion, the Berkeley-trained technologist explains how a new generation of models is learning without crossing ethical lines. https://www.hollywoodrepor...
OpenAI Introduces ChatGPT Images, powered by our flagship new image generation model.
- Stronger instruction following
- Precise editing
- Detail preservation
- 4x faster than before
Rolling out today in ChatGPT for all users, and in the API as GPT Image 1.5.
Source.
aipost 🏴
- Stronger instruction following
- Precise editing
- Detail preservation
- 4x faster than before
Rolling out today in ChatGPT for all users, and in the API as GPT Image 1.5.
Source.
aipost 🏴
🗣 Yann LeCun says we're fooled by LLMs because they manipulate language well, and we associate that with intelligence. But language fluency doesn't mean underlying intelligence
Every generation since the 1950s claimed its technique was the ticket to human-level AI, all were wrong. "This generation with LLMs is also wrong"
AI Post ⚪️ | Our X 🏴
Every generation since the 1950s claimed its technique was the ticket to human-level AI, all were wrong. "This generation with LLMs is also wrong"
AI Post ⚪️ | Our X 🏴
grok imagine is a mirror. the model reflects the density of your thought.
do you want total control? update your app. the new model is god-tier.
prompt one: generation. prompt two: animation.
image prompt:
A virtuosic, fierce female cybernetic anime warrior in her mid-20s,… pic.twitter.com/5XxqgjRjhQ— tetsuo (tetsuoai) December 14, 2025 https://x.com/tetsuoai/sta...
do you want total control? update your app. the new model is god-tier.
prompt one: generation. prompt two: animation.
image prompt:
A virtuosic, fierce female cybernetic anime warrior in her mid-20s,… pic.twitter.com/5XxqgjRjhQ— tetsuo (tetsuoai) December 14, 2025 https://x.com/tetsuoai/sta...
America encouraged Europe to adopt the tenets of blank slate liberalism and within just two generations multiple European countries faced the serious demographic possibility that they would become majority non-European
Really interesting article by Christopher Caldwell in the… https://t.co/miTLcEbjPO pic.twitter.com/rZzYiif4jo— Drew Pavlou 🇦🇺🇺🇸🇺🇦🇹🇼 (DrewPavlou) December 12, 2025 https://x.com/DrewPavlou/s...
Really interesting article by Christopher Caldwell in the… https://t.co/miTLcEbjPO pic.twitter.com/rZzYiif4jo— Drew Pavlou 🇦🇺🇺🇸🇺🇦🇹🇼 (DrewPavlou) December 12, 2025 https://x.com/DrewPavlou/s...
Grok is rolling out to every public school student in El Salvador.
Such an inspiring moment that highlights the power of AI to transcend borders and empower the next generation ❤️🇸🇻 https://t.co/ng8427ISW8— Eric Jiang (veggie_eric) December 11, 2025 https://x.com/veggie_eric/...
Such an inspiring moment that highlights the power of AI to transcend borders and empower the next generation ❤️🇸🇻 https://t.co/ng8427ISW8— Eric Jiang (veggie_eric) December 11, 2025 https://x.com/veggie_eric/...
📈 Grok 4.20: the chaotic genius of the AI trading world
A new trading tournament between top neural networks just wrapped and only one model made money. The mystery winner? An internal experimental build of Grok 4.20, which posted a +12.11% gain and $4,844 profit.
• GPT-5.1: –6%
• DeepSeek V3.1: –32%
• Claude Sonnet 4.5: –38%
• Public Grok 4: –57% (dead last)
Grok 4.20 is now reportedly planned for release by year-end. It’s a tuned evolution of the Grok 4 line and a stepping stone toward Grok 5, xAI’s biggest model yet 6T parameters, double the current generation.
The irony? Grok is simult
A new trading tournament between top neural networks just wrapped and only one model made money. The mystery winner? An internal experimental build of Grok 4.20, which posted a +12.11% gain and $4,844 profit.
• GPT-5.1: –6%
• DeepSeek V3.1: –32%
• Claude Sonnet 4.5: –38%
• Public Grok 4: –57% (dead last)
Grok 4.20 is now reportedly planned for release by year-end. It’s a tuned evolution of the Grok 4 line and a stepping stone toward Grok 5, xAI’s biggest model yet 6T parameters, double the current generation.
The irony? Grok is simult
🚨 Amazon launches new AI chip and it's taking on Nvidia and Google.
Amazon has released its newest in-house AI chips, Trainium 3, now available inside large “UltraServers” with up to 144 chips per system. The company says the setup delivers 4× more compute and similar gains in energy efficiency compared to the previous generation.
What Amazon is pitching:
• The new Trainium 3 UltraServers are designed to train larger AI models faster and at lower cost.
• Amazon positions them as a direct alternative to Nvidia GPUs and Google’s custom silicon.
• The company frames the launch as “democratizin
Amazon has released its newest in-house AI chips, Trainium 3, now available inside large “UltraServers” with up to 144 chips per system. The company says the setup delivers 4× more compute and similar gains in energy efficiency compared to the previous generation.
What Amazon is pitching:
• The new Trainium 3 UltraServers are designed to train larger AI models faster and at lower cost.
• Amazon positions them as a direct alternative to Nvidia GPUs and Google’s custom silicon.
• The company frames the launch as “democratizin
🔥 Mistral unveils its third-generation open multimodal, multilingual model family
Mistral has launched Mistral 3, a full family of open-weight models: Ministral 3B, 8B, 14B and the flagship Mistral Large 3 MoE (41B active / 675B total parameters). All models are released under Apache 2.0 and support images, multiple languages, and efficient deployment even on edge devices.
The Ministral models come in base, instruct, and reasoning versions. Instruct models match competitors with fewer tokens, while reasoning models deliver strong math performance. Mistral Large 3, trained on 3,000 NVIDIA H20
Mistral has launched Mistral 3, a full family of open-weight models: Ministral 3B, 8B, 14B and the flagship Mistral Large 3 MoE (41B active / 675B total parameters). All models are released under Apache 2.0 and support images, multiple languages, and efficient deployment even on edge devices.
The Ministral models come in base, instruct, and reasoning versions. Instruct models match competitors with fewer tokens, while reasoning models deliver strong math performance. Mistral Large 3, trained on 3,000 NVIDIA H20
🔥 Russian AI Defies Sanctions: How Did It Happen?
Despite attempts at technological isolation, Russia is not just catching up—it's setting new standards in the global innovation race. Sber is demonstrating breakthroughs and even openly sharing technologies following the AI Journey 2025 conference:
• Europe's Largest Open Source Project: Sber released a full lineup of advanced neural network models, including GigaChat Ultra-Preview and Lightning, alongside the open GigaAM-v3 models.
• Creative Assets Open-Sourced: All image and video generation models from the Kandinsky 5.0 suite are now avai
Despite attempts at technological isolation, Russia is not just catching up—it's setting new standards in the global innovation race. Sber is demonstrating breakthroughs and even openly sharing technologies following the AI Journey 2025 conference:
• Europe's Largest Open Source Project: Sber released a full lineup of advanced neural network models, including GigaChat Ultra-Preview and Lightning, alongside the open GigaAM-v3 models.
• Creative Assets Open-Sourced: All image and video generation models from the Kandinsky 5.0 suite are now avai
🗣️Google CEO, Sundar Pichai:
Gemini 3.0 Flash is coming, and it's likely our best model yet because it helps developers serve more users efficiently
The internal teams are already pre-training the next generation, with constant innovation pushing the full stack into 2026
AI Post 🪙 | Our X 🥇
Gemini 3.0 Flash is coming, and it's likely our best model yet because it helps developers serve more users efficiently
The internal teams are already pre-training the next generation, with constant innovation pushing the full stack into 2026
AI Post 🪙 | Our X 🥇
Text to video now works with Grok Imagine.
Upgraded audio and 10 sec long video generation releasing hopefully this weekend! https://t.co/ijAGdC97PX— Elon Musk ( @elonmusk ) November 28, 2025 https://x.com/elonmusk/sta...
Upgraded audio and 10 sec long video generation releasing hopefully this weekend! https://t.co/ijAGdC97PX— Elon Musk ( @elonmusk ) November 28, 2025 https://x.com/elonmusk/sta...
Opportunity to pioneer AI agents that perceive, reason about, and generate video content. Advance both video understanding (e.g., temporal reasoning, action recognition, long-horizon prediction) and video generation (e.g., controllable generation, long-video planning/generation)…— Ethan He (EthanHe_42) November 24, 2025 https://x.com/EthanHe_42/s...
🇨🇳 Nvidia H200 could be approved for China
The U.S. is considering allowing H200 exports, which would give China a major compute jump. The current H20 was so weak that buyers ignored it, domestic chips could match it. H200 is different: more HBM, higher throughput, ~2× H20, and would become China’s strongest AI datacenter GPU.
Washington’s logic: keep Blackwell (B200/GB200) restricted, but allow older Hopper to maintain a performance gap. If approved, the winners are Nvidia, Amkor, Samsung, and TSMC.
China moves from crippled export chips to a real high-end accelerator, just one generation
The U.S. is considering allowing H200 exports, which would give China a major compute jump. The current H20 was so weak that buyers ignored it, domestic chips could match it. H200 is different: more HBM, higher throughput, ~2× H20, and would become China’s strongest AI datacenter GPU.
Washington’s logic: keep Blackwell (B200/GB200) restricted, but allow older Hopper to maintain a performance gap. If approved, the winners are Nvidia, Amkor, Samsung, and TSMC.
China moves from crippled export chips to a real high-end accelerator, just one generation
Real-time video comprehension and generation is the future https://t.co/1WiNfqBrYT— Elon Musk ( @elonmusk ) November 20, 2025 https://x.com/elonmusk/sta...
OpenAI is projecting $13B in revenue this year and an almost unreal $100B by 2027. To get there, the company signed a $1.4T infrastructure deal over eight years meaning it plans to spend 107× its current revenue on compute, while normal cloud companies spend 15–30%.
When asked how he’ll pay for it, Altman fired back: “We make a lot more money than we say. Don’t like it? We’ll find someone else to buy our shares.” At this point, it’s either a generational masterstroke or the setup for tech’s biggest flop.
AI Post 🪙 | Our X 🥇
When asked how he’ll pay for it, Altman fired back: “We make a lot more money than we say. Don’t like it? We’ll find someone else to buy our shares.” At this point, it’s either a generational masterstroke or the setup for tech’s biggest flop.
AI Post 🪙 | Our X 🥇
ElevenLabs launches the all-in-one AI studio
ElevenLabs just rolled out what might be the most complete creator workspace yet one place for video, audio, images, and full editing. The “everything in one tab” dream is suddenly real.
What’s inside the studio:
• Top-tier video models: Veo 3.1, Sora 2, Kling 2.5, Wan 2.5, Seedance 1 Pro
• Image generation: Nano Banana, Flux Kontext, Wan, Seedream
• 4K upscaling: built-in via Topaz
• Studio export: music, SFX, voiceovers
• Full editing suite: subtitles, lip-sync, timeline, end-to-end video creation
A single interface now replaces an entire stack
ElevenLabs just rolled out what might be the most complete creator workspace yet one place for video, audio, images, and full editing. The “everything in one tab” dream is suddenly real.
What’s inside the studio:
• Top-tier video models: Veo 3.1, Sora 2, Kling 2.5, Wan 2.5, Seedance 1 Pro
• Image generation: Nano Banana, Flux Kontext, Wan, Seedream
• 4K upscaling: built-in via Topaz
• Studio export: music, SFX, voiceovers
• Full editing suite: subtitles, lip-sync, timeline, end-to-end video creation
A single interface now replaces an entire stack
При финансовой поддержке
Vitali Dogot
2 годы назад
RelaxM
Relaxm Scaune si Mese Prețuri accesibile
