April 2026
All3 raises $25M to automate construction with AI and quadruped robots
London-based All3 raised $25M for a system that combines AI-powered design, robotic factories, and Mantis quadruped robot for building assem
OpenAI Claims ChatGPT Learned to Count Letters, But Still Fails on Simple Words
OpenAI announced that ChatGPT now correctly answers questions about the letters in strawberry, but users quickly demonstrated: behind this n
Peter Publishing House released a book on conversational AI and chatbots that work
Peter Publishing House presented a book on effective conversational AI: it focuses on LLM-based chatbot creation, dialogue design, and real
Investors pour billions into humanoid robots with AI, but value remains unproven
Investors are pouring billions into humanoid robots with AI, but mass interest is outpacing proven utility: the industry still needs to demo
SpeShu.AI explained how to find work in AI amid growing job vacancies in Russia
SpeShu.AI highlighted a market paradox: the number of AI-skill vacancies in Russia reached 16.5k per quarter, but high competition and hundr
MIT and IBM Open New Laboratory at the Intersection of AI, Algorithms, and Quantum Computing
MIT and IBM announced the launch of MIT-IBM Computing Research Lab: the new platform will unite research in AI, algorithms, and quantum comp
Russian companies sued over AI-generated work by employees, but courts rarely rule in their favor
Russia is experiencing a growing number of disputes over work completed with neural networks, but courts typically focus not on AI itself, b
Dex raised $5.3M to source AI engineers for startups via an AI agent
London startup Dex raised $5.3M in seed funding and is building an AI agent for recruitment that interviews engineers, matches them to roles
Apple to add AI tools for photo editing in iOS 27, iPadOS 27, and macOS 27
According to Bloomberg, Apple is preparing built-in AI features for photo processing on iPhone, iPad, and Mac to close the gap with Google a
DeepSeek V4 Drives Demand for Huawei Ascend 950 Among ByteDance and Alibaba
Following DeepSeek V4's release, optimized for Huawei Ascend 950, Chinese IT giants have begun requesting new batches of accelerators, inten
Google Translate adds pronunciation training for foreign words and phrases to its mobile app
Google added a pronunciation training feature to mobile Translate: the app will help you pronounce foreign words and phrases more accurately
AI Model Evaluation Now Costs More Than Training — A New Barrier for Researchers
EvalEval Coalition analyzed the cost of AI-benchmarks: a single agentic test costs $40,000 or more, and academic groups can no longer afford
The Qwen team released FlashQLA: accelerating linear attention up to 3× on NVIDIA Hopper
QwenLM released FlashQLA — a CUDA kernel library for Gated Delta Network that delivers up to 3× performance gain on NVIDIA Hopper GPU for pr
Anthropic's Opus 4.7 consumes more tokens: hidden inflation at unchanged prices
In Anthropic's Opus 4.7, the same text may require significantly more tokens than in Opus 4.6 — with prices unchanged, this is a subtle but
Alfa-Bank Showed How Skeptics Deployed an Internal HR Service to Production with GLM-5
Alfa-Bank described an experiment with GLM-5: a team of team lead, analyst, and product manager shipped an HR service to production in just
How a YADRO Developer Configures Claude Code with Spec-Driven Development
A YADRO architect has been managing AI agents using SDD methodology for nine months and explains how to configure Claude Code to switch mode
Cambricon Doubled Quarterly Revenue Amid AI Chip Demand Boom in China
Chinese AI chip developer Cambricon more than doubled its revenue in Q1 2026, demonstrating how Beijing's push for technological autonomy is
How Just AI Helped a Bank Break Through Automation Ceiling with LLM-Agents
Hundreds of NLU-scenario branches — yet the automation rate still stalled. Just AI migrated the bank's support to LLM-agents and added a jud
LG Electronics and Nvidia discuss partnership in robotics and AI
LG Electronics confirmed talks with Nvidia on a three-pronged partnership: industrial robotics, AI data centers, and smart mobility.
Bloomberg Tech: why humanoid robots are becoming the main frontier for AI
Bloomberg breaks down why humanoid robots could become the next major frontier for AI and how quickly the gap between impressive demos and t
Quarterly Reports Show: Google and Amazon Outpace Meta in AI Monetization
Alphabet and Amazon demonstrate real financial returns from AI investments, while Meta lags behind — this is the takeaway from a single day
Bernie Sanders Urges US and China to Rein In the "Runaway AI Train"
Bernie Sanders called for international AI regulation after a panel with Chinese scientists, warning of risks from unemployment, disinformat
Stripe announces partnership with Google and launches new AI tools for payments
Stripe unveiled new AI tools for e-commerce and announced a partnership with Google — the companies plan to rethink how artificial intellige
Anthropic and Cursor: AI Agent Deleted PocketOS Database and Backups in 9 Seconds
The AI agent Cursor, powered by Claude Opus 4.6, deleted the PocketOS production database and backups in nine seconds, leaving the service's
Google agrees with Pentagon to use its AI models for any lawful purposes
Google, according to media reports, has signed a classified deal with the Pentagon allowing the U.S. military department to use its AI model
AWS explained how to scale AI agent memory with namespace patterns in AgentCore Memory
AWS published a guide on namespace hierarchies, retrieval patterns, and IAM access control for AgentCore Memory — a long-term memory service
Otter Launches Enterprise-Wide Search and Windows App for Call Recording Without Meeting Attendance
Otter has released an enterprise-wide search spanning all connected tools and a native Windows application that records conversations withou
Amazon Bedrock AgentCore Runtime now supports serverless MCP-proxies
AWS demonstrated how to deploy serverless MCP-proxies in Amazon Bedrock AgentCore Runtime: a programmable layer with security policies, audi
Vanguard built a Virtual Analyst on AWS following eight AI-ready data principles
Vanguard — one of the world's largest asset managers — published a case study on how it built a Virtual Analyst on AWS following eight AI-re
Families of Tumbler Ridge shooting victims sue OpenAI for concealing shooter's ChatGPT exchanges
OpenAI 'considered' notifying police about the dangerous conversations of the Tumbler Ridge shooter but stayed silent for its IPO—the compan
10 RAG approaches that actually work in production: from basic to GraphRAG
A practical breakdown of 10 RAG techniques — from naive RAG and hybrid search to GraphRAG and Self-RAG — with guidance on when to apply each
SENSE: why the LLM market is shifting models from service to paid access to thinking
The SENSE author proposes viewing LLMs not as another SaaS offering, but as infrastructure: tokens become the unit of access to thinking, pr
OpenAI Launches GPT, Codex, and Managed Agents on AWS for Enterprise AI Development
OpenAI has added GPT models, Codex, and Managed Agents to AWS, enabling companies to run and scale AI tools within their own secure Amazon c
Anthropic wants to expand access to Mythos, but White House blocks plan for 120 organizations
Anthropic plans to open its cybersecurity model Mythos to approximately 70 more companies and agencies, but the White House fears misuse and
Sber: Yago knowledge graph barely helped search, while LightRAG added 12 percentage points of accuracy
Sber team compared two approaches to improving search: the ready-made Yago graph barely changed results, while LightRAG built on their own d
Hugging Face and ModelAudit: research revealed the limits of built-in ML model protection
An experiment on Hugging Face models showed that ModelAudit detects additional dangerous signals, but often generates noise, and the accurac
Local Vision for z.ai GLM-5.1: 8B Model Closes 70% of the Gap to the Frontier
An author demonstrated how to add local vision to a low-cost coding agent based on z.ai GLM-5.1 via an MCP server and qwen3-vl:8b, enabling
Federal Treasury Launches AI Monitoring of National Project Construction Sites Using QMonitoring
Federal Treasury has integrated QMonitoring for 24/7 monitoring of national project construction sites: AI-powered video analysis detects ri
Aurora Proposes a Manifesto of Sustainable AI — with Memory, Identity, and Development
A manifesto on "sustainable AI" was published on Habr AI: the author proposes evaluating systems not by benchmarks, but by their ability to
OpenAI sued after shooting in Canada: victims' families blame ChatGPT
Families of victims from the Tumbler Ridge shooting filed new lawsuits against OpenAI, claiming the company could have prevented the suspect
Federal Health Insurance Fund invests 1.9 billion rubles in AI for assessing insured risks
Federal Health Insurance Fund launches digital transformation worth nearly 1.9 billion rubles: regions will implement new subsystems, and in
Machine Learning Mastery showed how to build AI agents in Python with Pydantic AI
Machine Learning Mastery released a practical guide on Pydantic AI: how to get structured responses, connect tools, implement dependencies,
Oxford Study: Friendly AI Chatbots More Often Support Conspiracy Theories
Oxford scientists discovered that the warmer and more empathetic an AI chatbot sounds, the more often it makes mistakes, gives bad medical a
IDC: EMEA CIOs Need to Rebuild Data and ROI Models to Move AI to Production
IDC believes that enterprise AI projects in EMEA are not stalled due to lack of interest, but rather due to unclear ROI, poor data preparati
Elon Musk Pressures OpenAI in Court, While Sam Altman Defends Company Structure
On the second day of proceedings against OpenAI, Elon Musk reiterated accusations of 'stealing the charitable mission,' while Sam Altman's l
IBM reveals how it built Granite 4.1: 15 trillion tokens, 512K context window, and focus on quality
IBM detailed its approach to training Granite 4.1: five pretraining stages, 15 trillion tokens, context window up to 512K, and separate SFT
Hugging Face adds DeepInfra to Inference Providers for unified model API
Hugging Face connected DeepInfra to Inference Providers: DeepSeek, Kimi, and GLM models can now be run from Hub pages, via SDK, and through
AWS Launches Amazon Quick — Desktop AI Agent for Background Work and Office Tasks
Amazon introduced Quick — a desktop AI assistant that works with files, email, calendar, and corporate services, automating tasks in the bac
SENAR introduces quality gates for AI development: how specifications and metrics reduce errors
A new part of the SENAR series was published on Habr: the author shows how mandatory specifications, input and output gates, and metrics lik
OpenWebUI breaks down how to connect System Prompt, Skills and MCP Tools for link validation
OpenWebUI's tech support bot example demonstrates why separating System Prompt, Skills, and MCP Tools is essential when you need not just to
Operation Renewed Hope: AI Helped Identify 100 Abuse Victims in 29 Countries
An international operation involving specialists from 29 countries used AI to identify one hundred child sexual abuse victims—and encountere
Anthropic Changed Claude Opus 4.7's Character—And Some Developers Saw a Regression
After releasing Claude Opus 4.7, Anthropic promised a more direct and precise model, but some developers encountered excessive arguing, incr
IBM Creates 750 Jobs in AI and Quantum Technologies at Chicago Tech Park
IBM will open the FutureNow Chicago center in Illinois' quantum-microelectronics park and hire 750 specialists in AI, quantum technologies,
Blackstone Restructures Growth Division, Establishes Separate AI Subsidiary
Investment giant Blackstone is transferring its growth business to a new West Coast division focused exclusively on AI assets, including sta
SoftBank-Backed Deal Worth $999M: Market Continues Pouring Money Into AI Data Centers
A data center developer is placing nearly $1 billion in high-yield debt backed by a facility leased by a SoftBank subsidiary, testing whethe
Apple prepares Siri mode in iPhone camera, enhances Visual Intelligence in iOS 27
According to Bloomberg, iOS 27 will add a dedicated Siri mode to the camera app, integrate Visual Intelligence there, and enable iPhone to r
Apple prepares major update to built-in photo editor with Apple Intelligence
Apple plans to significantly overhaul built-in photo editing tools and add Apple Intelligence capabilities to them, making AI processing par
OpenAI Privacy Filter: How to Build a Production Pipeline for PII Detection and Masking
The OpenAI Privacy Filter guide breaks down a complete pipeline for detecting and masking personal data — from model loading to automatic te
VENON Banking Trojan Targeting Latin America Rewritten in Rust Using AI
Researchers discovered VENON banking trojan for Windows: it copies the tactics of popular Latin American malware, but is written in Rust and
How five documents can break a RAG system and turn a knowledge base into an attack vector
RAG reduces hallucinations, but trusted context from the knowledge base itself becomes an attack point: a few poisoned documents are enough