Latest publications
Tech companies worldwide laid off 73,000 people in a quarter amid restructuring for AI
According to Layoffs.fyi, in the first quarter of 2026, 95 technology companies cut more than 73,200 employees, shifting budgets from traditional teams into AI, automation, and infrastructure.
China delays HBM3 launch again: local AI accelerator makers' plans at risk
China has again postponed the launch of its own HBM3 production, along with local companies' plans to release competitive AI accelerators without relying on imported memory.
Alphabet discusses two new AI chips with Marvell to speed up inference in Google services
Alphabet has begun talks with Marvell on two specialized AI chips: one is meant to speed up data transfer, the other to update Google's TPU for fast interactive services.
Samsung unveiled Project Luna — a home companion robot for moving AI between devices
At Milan Design Week 2026, Samsung unveiled the Project Luna concept — a home robot with a round rotating screen meant to move AI workflows between devices around the home.
Futuresource: everyday AI gadgets will turn people into a "walking supercomputer"
Futuresource forecasts that in the coming years, the combined AI power of smartphones, laptops, headphones and other personal devices will reach a level that previously required a data center.
Microsoft made the Copilot agent in Word, Excel and PowerPoint available to all Microsoft 365 subscribers
Microsoft has opened access to Copilot for all Microsoft 365 subscribers: the AI agent in Word, Excel and PowerPoint can now perform multi-step actions under user control.
Elon Musk revealed how the Macrohard project will unite xAI and Tesla to create software
Musk explained that Macrohard is a joint system by xAI and Tesla: Grok will serve as the “navigator,” while Tesla’s agent will see the screen, work with video, and control the computer.
Google DeepMind admits: Genie 3 worlds lose coherence after about a minute
At GDC 2026, Google DeepMind said the AI system Genie 3 can now maintain the coherence of game worlds for about a minute, but that is still not enough to replace traditional engines.
Meta unveiled a roadmap for four MTIA AI chips for inference, but without abandoning Nvidia and AMD
Meta presented a roadmap for four MTIA accelerators: the company wants to cut the cost of AI model inference in its data centers, but it is not yet ready to fully move away from Nvidia and AMD.
Elon Musk reshapes xAI: layoffs, consultants from Tesla and SpaceX, and a bet on AI data centers
Musk has begun a tough restructuring of xAI: after an influx of money from Tesla and closer ties with SpaceX, outside consultants are joining the startup, and part of the team is facing layoffs.
US drops plan to tie AI chip sales to mandatory data center construction
Washington did not back the idea of selling American accelerators only to countries and companies willing to build data centers in the US, preserving more flexibility for the market.
OpenAI postponed the launch of erotic chats in ChatGPT over risks to teenagers
OpenAI has again delayed erotic chats in ChatGPT: the company was held back by weak age verification, the risk of teen access, and concerns about harmful scenarios and addiction.
Musk lawsuit against OpenAI and Microsoft to proceed, though $134 billion figure deemed arbitrary
A judge questioned the damages estimate in Elon Musk's lawsuit against OpenAI and Microsoft, at up to $134 billion, but did not end the dispute and left the claim for a jury to assess.
Nvidia founder expects $1 trillion in revenue from AI hardware by the end of 2027
At the opening of GTC 2026, Jensen Huang said Nvidia could generate at least $1 trillion from sales of AI hardware by the end of 2027 — a signal of the scale of demand for infrastructure.
Nvidia unveils Vera Rubin: seven chips and a full platform for AI factories
Nvidia unveiled Vera Rubin as a full stack for AI factories: new GPUs, CPUs, network switches, and inference accelerators are meant to improve performance per watt and lower token cost.
Nvidia introduced Agent Toolkit and NemoClaw for rapid and secure AI agent deployment
At GTC 2026, Nvidia announced Agent Toolkit and NemoClaw — open tools that simplify AI agent deployment and add control, security, and scalability.
OpenAI scales back side projects and bets on coding and the enterprise market
OpenAI is preparing a strategic pivot: the company wants to cut secondary initiatives to focus resources on code generation, ChatGPT, and products for enterprise customers.
Google opens “Personal Intelligence” to free users of Gemini, Chrome, and AI Mode
Google has removed the paywall from the Personal Intelligence feature: it is now available to free users of Gemini, Chrome, and AI Mode and is becoming part of the company’s mainstream AI ecosystem.
OpenAI introduces GPT-5.4 mini and nano for coding, sub-agents, and large-scale AI workloads
OpenAI released GPT-5.4 mini and nano, faster and cheaper versions of its flagship model for coding, sub-agents, UI interaction, and high-load tasks.
Jensen Huang of Nvidia called OpenClaw the 'next ChatGPT' for the era of AI agents
The Nvidia chief believes OpenClaw could become as pivotal for autonomous AI agents as ChatGPT was for the mainstream chatbot market.
Amazon bets on AI: Andy Jassy forecasts AWS at $600 billion a year in revenue
Amazon CEO Andy Jassy believes that in ten years AWS could reach $600 billion in annual revenue if demand for AI infrastructure becomes the main driver of the cloud.
Elon Musk: Tesla and SpaceX AI will continue buying Nvidia chips in bulk despite AI5
Elon Musk clarified that Tesla and SpaceX AI will not replace Nvidia with their own processors: even as AI5 is being developed, the companies will continue buying accelerators in large volumes.
Apple could make more than $1 billion from AI apps while its updated Siri stalls
Despite delays to Siri and Apple Intelligence, Apple is already making money from the boom in third-party AI services: revenue from apps in its ecosystem could exceed $1 billion.
Roblox prepares the Reality platform with AI rendering and a push for photorealistic graphics
Roblox is building Reality, a hybrid platform where a cloud engine runs the game and AI adds textures, lighting, and environments to significantly improve visual quality.
OpenAI revealed the source of the «gremlins» in ChatGPT and showed how to remove the restriction in Codex
OpenAI explained why GPT models started inserting goblins and gremlins into responses, linked it to the Nerdy personality mode, and even showed how to remove the restriction in Codex.
Google allowed ads in Gemini after previous commitments against advertising
Google no longer rules out ads in Gemini: the company is testing formats in AI Mode and allowing their transfer to the chatbot, while OpenAI already shows ads in ChatGPT.
Nvidia CEO Jensen Huang urged AI companies not to scare people with apocalyptic rhetoric
At GTC 2026, Jensen Huang criticized the grim rhetoric around AI, said the main threat is not the technology itself but fear of it, and defended Anthropic.
Nvidia explained the principle behind DLSS 5: the system only needs a 2D frame and motion vectors
Nvidia revealed the basic principle behind DLSS 5: the neural network reconstructs the image not from a full 3D scene, but from a regular 2D frame and motion vectors.
Alibaba shipped nearly 500,000 AI chips, but admitted it lags behind Nvidia
China's Alibaba said it has shipped 470,000 of its own T-Head AI chips, but openly acknowledged that they trail Nvidia in performance, and is now betting on a combination of hardware, cloud, and Qwen.
Pentagon makes Palantir Maven the foundational AI system for the U.S. armed forces
The U.S. Defense Department has decided to make Palantir's Maven its main combat AI platform: the system will be standardized across the armed forces, intelligence, and future military AI solutions.