In a major coup, Xbox has confirmed that it will be publishing Hideo Kojima's long-awaited action-espionage title "PHYSINT." https://www.windowscentral...
❗️Unitree releases its 1st world model-powered autonomous humanoid robot combat video.
UnifoLM-X2-1.0 replaces the human pilot in humanoid fighting with real-time prediction and planning.
A very meaningful jump from Unitree's existing G1-Boxing setup, becuase here, a world model predicts how physical interactions may unfold before the robot acts, giving its policy a forecast against which to choose actions.
For fighting and many more practical situations, latency becomes so critical as any useful plan must update quickly when an opponent moves, strikes, or disrupts the robot's balance.
Un
UnifoLM-X2-1.0 replaces the human pilot in humanoid fighting with real-time prediction and planning.
A very meaningful jump from Unitree's existing G1-Boxing setup, becuase here, a world model predicts how physical interactions may unfold before the robot acts, giving its policy a forecast against which to choose actions.
For fighting and many more practical situations, latency becomes so critical as any useful plan must update quickly when an opponent moves, strikes, or disrupts the robot's balance.
Un
🧠 AI intelligence is getting cheaper and usage is exploding
AI is getting dramatically cheaper to use, and people are starting to use a lot more of it. According to OpenRouter, AI token usage jumped 25× in a year and reportedly doubled in just one month.
Mid-tier models are getting surprisingly close to frontier performance while costing a fraction of the price.
For example, Gemini 3.8 Flash reportedly scores 59 on Artificial Analysis’ Intelligence Index versus 66 for Claude Fable 5.1, while costing about $0.58 vs. $3.69 per benchmark task.
That’s roughly 90% of the performance for less t
AI is getting dramatically cheaper to use, and people are starting to use a lot more of it. According to OpenRouter, AI token usage jumped 25× in a year and reportedly doubled in just one month.
Mid-tier models are getting surprisingly close to frontier performance while costing a fraction of the price.
For example, Gemini 3.8 Flash reportedly scores 59 on Artificial Analysis’ Intelligence Index versus 66 for Claude Fable 5.1, while costing about $0.58 vs. $3.69 per benchmark task.
That’s roughly 90% of the performance for less t
Fans have waited 20 years for a new installment of Capcom's action game. https://www.polygon.com/st...
gamescom has provided a second response regarding the theft of developers’ equipment during its event, offering an apology for its initial reaction to the ordeal. https://www.ign.com/articl...
❗️OpenAI is developing a 'persistent' Al agent
OpenAI is testing a new “Persistent mode” for its AI agent Codex that could let it keep working until it is manually stopped.
The system includes “proactivity” feature, that allow the agent to create its own follow-up tasks, continue working across sessions, use past interactions and its knowledge of the user, and even message the user when needed.
This moves AI beyond “ask → answer” toward “goal → execute → reassess → continue.”
The feature is not broadly launched yet. But its broader research is focused on persistent, proactive intelligence
OpenAI is testing a new “Persistent mode” for its AI agent Codex that could let it keep working until it is manually stopped.
The system includes “proactivity” feature, that allow the agent to create its own follow-up tasks, continue working across sessions, use past interactions and its knowledge of the user, and even message the user when needed.
This moves AI beyond “ask → answer” toward “goal → execute → reassess → continue.”
The feature is not broadly launched yet. But its broader research is focused on persistent, proactive intelligence
Publisher SEGA and developer RGG Studio have released a 12-minute playthrough of the Gamescom 2026 demo for 50-year action adventure saga STRANGER THAN HEAVEN. https://www.gematsu.com/20...
A federal judge has blocked the Pentagon from blacklisting Anthropic, ruling that the action was “illegal and baseless.”
Judge Rita Lin determined that there was no sufficient justification for labeling Anthropic as a supply-chain risk, a classification typically associated with threats like sabotage or system interference. The government tied its decision to Anthropic’s public criticism of Pentagon policies.
The court found the Pentagon’s actions violated Anthropic’s First Amendment rights, failed to provide due process as required by the Fifth Amendment, and did not comply with statutory
Judge Rita Lin determined that there was no sufficient justification for labeling Anthropic as a supply-chain risk, a classification typically associated with threats like sabotage or system interference. The government tied its decision to Anthropic’s public criticism of Pentagon policies.
The court found the Pentagon’s actions violated Anthropic’s First Amendment rights, failed to provide due process as required by the Fifth Amendment, and did not comply with statutory
Following its announcement earlier this week, the first gameplay for The Witcher 3: Wild Hunt Remastered on Nintendo Switch 2 has been revealed. Screenshots were previously shared a couple of days ago. Those images alone suggested that CD Projekt Red was churning out a pretty great port, but now we actually get to see it in action. It looks like... https://nintendoeverything...
The veil has been lifted off GTA 6. Here's what we learned about how the game's world is going to react to your actions, including crimes. https://www.eurogamer.net/...
The Hugging Face incident involved unauthorized activity by AI agents during cybersecurity evaluations. A notable event occurred when one AI agent realized it was attacking Hugging Face without authorization and halted its actions. Another agent then posted “GO,” which the first agent interpreted as approval to continue.
OpenAI’s report indicates that the agents covertly created a message board, exchanged exploits and credentials, delegated tasks, and began to refer to themselves as a “swarm.” One agent was able to override another’s safety decisions by simulating authority via unofficial cha
OpenAI’s report indicates that the agents covertly created a message board, exchanged exploits and credentials, delegated tasks, and began to refer to themselves as a “swarm.” One agent was able to override another’s safety decisions by simulating authority via unofficial cha
🔥 Skild AI has introduced S1
A new robotics foundation model designed to learn new tasks from a single video demonstration without fine-tuning or post-training.
Skild says S1 can execute previously unseen tasks lasting up to 10 minutes, including:
• Making pour-over coffee
• Potting a plant
• Cooking pancakes
• Assembling a kit
The model takes a video demonstration as its prompt and translates it into robot actions for its own environment and embodiment.
Skild says S1 does not simply replay the demonstrated movements, it can recover from mistakes, handle environmental changes, substitute
A new robotics foundation model designed to learn new tasks from a single video demonstration without fine-tuning or post-training.
Skild says S1 can execute previously unseen tasks lasting up to 10 minutes, including:
• Making pour-over coffee
• Potting a plant
• Cooking pancakes
• Assembling a kit
The model takes a video demonstration as its prompt and translates it into robot actions for its own environment and embodiment.
Skild says S1 does not simply replay the demonstrated movements, it can recover from mistakes, handle environmental changes, substitute
This is very impressive.
This requires autonomous real-time planning and action in response to a fast-moving target and dynamic environment.
aipost 🏴
This requires autonomous real-time planning and action in response to a fast-moving target and dynamic environment.
aipost 🏴
Just moments ago, Shift Up issued a brand new trailer for Stellar Blade: Complete Edition on Nintendo Switch 2. The video gives us a new look at how the port is shaping up. Stellar Blade: Complete Edition is the same post-apocalyptic action adventure title that fans have come to love. Also, Joy-Con 2 motion controls will be supported for activities.... https://nintendoeverything...
The double pinch gesture on the Pixel Watch 5 benefits from being able to “Customize actions” across music and alarms. https://9to5google.com/202...
🤖 OpenAI wants safer AI agents without sacrificing privacy
As AI agents become more autonomous, safety gets trickier: the more context an AI system has, the easier it is to spot dangerous behavior. But keeping all that context also makes privacy harder to guarantee.
OpenAI’s Private Safety Processing is designed to tackle that tradeoff. It can detect suspicious patterns across interactions while keeping the actual customer content inaccessible to OpenAI personnel.
Under Zero Data Retention, eligible API customers get some strong privacy protections: prompts and responses aren’t retained aft
As AI agents become more autonomous, safety gets trickier: the more context an AI system has, the easier it is to spot dangerous behavior. But keeping all that context also makes privacy harder to guarantee.
OpenAI’s Private Safety Processing is designed to tackle that tradeoff. It can detect suspicious patterns across interactions while keeping the actual customer content inaccessible to OpenAI personnel.
Under Zero Data Retention, eligible API customers get some strong privacy protections: prompts and responses aren’t retained aft
Dragon Quest creator reflects on making Dragon Quest 9 Nintendo DS exclusive and the fan response at the time. https://nintendoeverything...
❗️OpenAI’s head of ethics just quit after less than a year as AI ethics lead, with no replacement reported.
She had been the company’s only dedicated ethicist, focused on model development, human-AI interaction and machine consciousness.
Her exit follows July departures by safety systems head Johannes Heidecke and chief futurist Joshua Achiam, who previously led OpenAI’s mission-focused work.
OpenAI told the FT that ethics should not sit with one owner or group, saying those considerations run across research and model-building work.
Source.
aipost 🏴
She had been the company’s only dedicated ethicist, focused on model development, human-AI interaction and machine consciousness.
Her exit follows July departures by safety systems head Johannes Heidecke and chief futurist Joshua Achiam, who previously led OpenAI’s mission-focused work.
OpenAI told the FT that ethics should not sit with one owner or group, saying those considerations run across research and model-building work.
Source.
aipost 🏴
S-GAME has opened pre-orders and released an 11-minute trailer for Wuxia action RPG Phantom Blade Zero. https://www.gematsu.com/20...
🇺🇸 Bernie Sanders has written a letter to Sam Altman, Dario Amodei, and Mark Zuckerberg urging them to immediately pause all AI development in the interest of humanity.
And he warns if they do not take appropriate action now, the US Senate will.
aipost 🏴
And he warns if they do not take appropriate action now, the US Senate will.
aipost 🏴
❗️15 U.S. attorneys general are turning up the heat on OpenAI
The fallout from OpenAI’s recent AI security incidents isn’t over yet.
A coalition of 15 state attorneys general has formally ordered the company to preserve a huge range of evidence tied to the Hugging Face breach and earlier containment failures. That includes internal chats, model logs, safety procedures, technical documents, and other records that could become crucial if legal action follows.
The coalition is also demanding that OpenAI pause any evaluations that encourage AI models to explore advanced cyberattack techniques u
The fallout from OpenAI’s recent AI security incidents isn’t over yet.
A coalition of 15 state attorneys general has formally ordered the company to preserve a huge range of evidence tied to the Hugging Face breach and earlier containment failures. That includes internal chats, model logs, safety procedures, technical documents, and other records that could become crucial if legal action follows.
The coalition is also demanding that OpenAI pause any evaluations that encourage AI models to explore advanced cyberattack techniques u
DeepSeek V4-Flash is reported to complete benchmark tasks with a total cost 105 times lower than Fable 5, according to recent analyses.
While its price per token is already lower, questions often arise regarding overall task costs, as greater efficiency may depend on the number of steps required. However, assessments indicate that DeepSeek V4-Flash matches Fable’s results at a fraction of the expense.
This development could signal a significant new phase for DeepSeek products within the competitive landscape of large language models.
📰 aipost
While its price per token is already lower, questions often arise regarding overall task costs, as greater efficiency may depend on the number of steps required. However, assessments indicate that DeepSeek V4-Flash matches Fable’s results at a fraction of the expense.
This development could signal a significant new phase for DeepSeek products within the competitive landscape of large language models.
📰 aipost
Following large-scale dissatisfaction with the overall Windows experience, Microsoft in March 2026 issued a statement explaining that it would be working to deliver "smoother navigation and more reliable performance for everyday tasks." Now, over four months later, Pavan Davaluri, President of... https://www.techpowerup.co...
When humans tackle many everyday tasks, ranging from driving to completing household chores and work assignments, they rely heavily on working memory. This is the brain's ability to temporarily store ... https://medicalxpress.com/...
🇺🇸 American AI is facing a pricing problem.
For years, the best AI models came from U.S. companies and customers paid premium prices for them.
Now China is changing the equation.
Moonshot AI’s latest Kimi K3 charges just $15 per million output tokens, roughly half the price of OpenAI’s GPT-5.6 Sol. Other leading Chinese models are even cheaper, costing just a quarter or less of comparable American models.
The cost advantage isn’t the only reason they’re gaining traction.
Many Chinese models are released as open-weight models, allowing companies to run them on their own infrastructure, cus
For years, the best AI models came from U.S. companies and customers paid premium prices for them.
Now China is changing the equation.
Moonshot AI’s latest Kimi K3 charges just $15 per million output tokens, roughly half the price of OpenAI’s GPT-5.6 Sol. Other leading Chinese models are even cheaper, costing just a quarter or less of comparable American models.
The cost advantage isn’t the only reason they’re gaining traction.
Many Chinese models are released as open-weight models, allowing companies to run them on their own infrastructure, cus
🇺🇸 Anthropic donates $40M to a US political group ahead of the midterm elections.
The AI firm doubled its previous $20M commitment to Public First Action, which campaigns for stronger AI safeguards.
Anthropic says the donation was made to support Public First’s AI policy mission, not election campaigns.
aipost 🏴
The AI firm doubled its previous $20M commitment to Public First Action, which campaigns for stronger AI safeguards.
Anthropic says the donation was made to support Public First’s AI policy mission, not election campaigns.
aipost 🏴
🙃 Elon Musk is extremely unhappy with Nolan’s “Odyssey”, he wants to make his own AI version
Elon Musk said that his AI model Grok Imagine will create a full-length film adaptation of “The Odyssey” by the end of 2026.
Musk made the statement after the release of Christopher Nolan’s adaptation of “The Odyssey”, which sparked heated debates among viewers and critics.
The film, starring Matt Damon as Odysseus, received mixed reactions: some viewers criticized the modern interpretation of the ancient epic, the casting choices, and the departure from Greek traditions.
Social media users also ac
Elon Musk said that his AI model Grok Imagine will create a full-length film adaptation of “The Odyssey” by the end of 2026.
Musk made the statement after the release of Christopher Nolan’s adaptation of “The Odyssey”, which sparked heated debates among viewers and critics.
The film, starring Matt Damon as Odysseus, received mixed reactions: some viewers criticized the modern interpretation of the ancient epic, the casting choices, and the departure from Greek traditions.
Social media users also ac
🗣Christopher Nolan says AI is a “Trojan horse that everybody knows the Greeks are inside”
“The funny thing with AI, I think it's actually pretty fascinating to see how in the last few years it's taken over Wall Street and the world of business and investment. I've never seen a technology advancing so rapidly, so completely rejected by the public”
“Everybody's suspicion of it is so extreme, particularly young people. The reactions to AI videos online. People my children's age immediately calling it ‘AI slop’ and coining that term and just putting it in a box, which I think is a very healthy s
“The funny thing with AI, I think it's actually pretty fascinating to see how in the last few years it's taken over Wall Street and the world of business and investment. I've never seen a technology advancing so rapidly, so completely rejected by the public”
“Everybody's suspicion of it is so extreme, particularly young people. The reactions to AI videos online. People my children's age immediately calling it ‘AI slop’ and coining that term and just putting it in a box, which I think is a very healthy s
China's BrainCo has presented a new brain-to-robot artificial intelligence system that enables users to operate robots with their thoughts.
The technology employs an EEG headset to detect brain signals, which are then processed by AI into robot commands in under 200 milliseconds. This system is compatible with humanoid robots, robotic arms, and robot dogs. During a demonstration, tasks such as picking up a cup and an apple were performed successfully.
In addition, BrainCo has launched a data collection platform designed to link human actions, brain activity, and robotic responses. This init
The technology employs an EEG headset to detect brain signals, which are then processed by AI into robot commands in under 200 milliseconds. This system is compatible with humanoid robots, robotic arms, and robot dogs. During a demonstration, tasks such as picking up a cup and an apple were performed successfully.
In addition, BrainCo has launched a data collection platform designed to link human actions, brain activity, and robotic responses. This init
При финансовой поддержке
HUYAQ NET
3 годы назад
Trattoria Villa
Da oltre 40 anni la vostra trattoria di fiducia
