Latest publications
Coinbase for Agents: AI agents gain access to user trading portfolios
Coinbase granted AI agents direct access to financial portfolios: language models can now not only analyze the market but also automatically execute trades.
GitHub Copilot Transitions to Token-Based Billing — Users See Price Increases
GitHub Copilot switched from fixed subscription to pay-as-you-go billing. Early users report increased usage costs.
OpenAI releases framework for safe enterprise AI deployment
OpenAI has introduced the Frontier Governance Framework, a set of documents and recommendations that help companies safely deploy and scale large language models.
Google Redesigns Payments for AI Agents: Universal Commerce Protocol
Google is implementing the Universal Commerce Protocol, a new payment infrastructure that will allow AI agents to independently conduct transactions without human involvement.
Anthropic Releases Claude Opus 4.8 with Improvements for Code and Agents
Anthropic introduced Claude Opus 4.8 with improved coding quality, agent performance, and reasoning capabilities — the new version is available via claude.ai, Claude Code, and API.
NBA Plans to Implement AI for Automatic Out-of-Bounds Ball Detection
NBA will implement an AI system for automatic out-of-bounds detection using a network of cameras and artificial intelligence, similar to Hawk-Eye in tennis.
Google Shuts Down Display Ads, Moving All Marketing to AI
Google announced the closure of Display Network and transition to AI-first Demand Gen platform—marking the end of a 20-year era when marketers controlled every step of their campaigns.
OpenAI and Singapore Launch $300M Laboratory Focused on AI Agents
OpenAI will open its first international laboratory in Singapore as part of a partnership with MDDI, with over $300 million SGD in funding dedicated to developing advanced AI agent systems.
Nvidia Vera chip: Jensen Huang's $200 billion bet amid record revenues
Nvidia introduced the Vera chip — a microchip that the company is betting $200 billion on. Jensen Huang clearly does not want this news to go unnoticed.
China mapped the entire renewable energy grid: why it matters to the world
Artificial intelligence consumes electricity at a pace that no modern energy system in the world is prepared for.
Alibaba designs AI chips for agents — and that changes the race itself
Alibaba unveiled Zhenwu M890, a chip for AI agents backed by a multi-year roadmap. It signals that the company is building an independent AI stack rather than merely offsetting U.S. sanctions.
TechEx North America: Why Corporate AI Projects Get Stuck in Pilots
On the second day of the TechEx conference, experts discussed the main paradox: AI projects shine in pilots but rarely move to production. They revealed three ways out of the dead end.
Amazon launched Alexa for Shopping, combining Rufus with the Alexa+ platform
Amazon has launched **Alexa for Shopping**, combining its Rufus (an AI assistant for shopping) with the Alexa+ platform into a unified system available everywhere people shop on Amazon.
Nvidia H200 export deal to China stalled at Trump-Xi summit
Trump flew to Beijing with Nvidia CEO Jensen Huang and promised journalists changes in high-tech chip trade.
NHS uses AI to reduce waiting lists and ease pressure on doctors
The UK's NHS is deploying AI technologies to manage an acute crisis: 7.25 million patients are waiting for elective procedures. New policies call for shifting part of care from hospitals to outpatient centers. Intelligen
RingCentral adds Shopify, Calendly, and WhatsApp support to AI Receptionist
RingCentral has expanded AI Receptionist with integrations for Shopify, Calendly, and WhatsApp. The virtual assistant can now process orders, schedule appointments, and communicate in messaging apps, going beyond simply
Bain estimates SaaS agentic AI market at $100 billion
Bain & Company forecasts that the U.S. market for SaaS applications built on agentic AI will grow to $100 billion, driven by the automation of office workflows.
Physical AI Expo comes to Silicon Valley: robots and autonomous systems in focus
Physical AI Expo North America will take place in San Jose — a conference where engineers and developers will discuss the future of robotics and the deployment of AI in the real world.
Humanoid to deploy humanoid robots at Schaeffler factories
British company Humanoid has signed a contract with German manufacturer Schaeffler to deploy 1000–2000 humanoid robots at its global production sites by 2032.
Autorek report: without a unified data layer, insurers will not be able to unlock AI's potential
Autorek believes that what is preventing insurers from implementing AI is not the models, but fragmented data, manual reconciliations, and weak integration between operational and financial systems.
Denodo: autonomous AI systems depend on the quality of enterprise data governance
As AI becomes more autonomous, the weak point is not only the models but also the data: without a unified governance layer, systems are more likely to make mistakes, conflict with one another, and create compliance risks
KPMG: companies spend $186 million on AI, but see no real returns — agents change the equation
KPMG found that the average corporate AI budget has reached $186 million a year, but the gap between investment and measurable business value continues to widen rapidly.
KiloClaw from Kilo helps companies get shadow AI and autonomous agents under control
Kilo launched the KiloClaw platform so companies can see, restrict, and securely connect autonomous AI agents that employees deploy outside IT oversight.
IBM explained why AI governance protects business margins and reduces vendor dependence
IBM believes businesses should manage AI as critical infrastructure: without transparency, auditing, and cost control, closed models begin to erode margins.
OutSystems: AI is already in production for IT teams, but the business lacks unified governance
A survey by OutSystems of 1,879 IT leaders found that AI is already being used in development workflows, but companies are finding it increasingly difficult to keep tools, teams, and priorities under control.
Apple and Qualcomm develop AI agents with guardrails and mandatory confirmation
Apple and Qualcomm are advancing assistants that can navigate apps and prepare purchases, but they are not given full freedom: money, accounts, and data access remain under user control.
SAP integrates agentic AI into SuccessFactors for hiring, payroll, and HR management
SAP has integrated a network of AI agents into SuccessFactors 1H 2026: they are intended to catch HR process failures early, speed up hiring and payroll, and reduce manual support costs.
Google Gemma 4 forced companies to reassess AI controls on local devices
The release of Google Gemma 4 intensified the problem of corporate control over AI: models now run directly on employees’ laptops, bypassing cloud gateways, logs, and standard audit measures.
DroneDash and Geodnet build an agricultural drone for large farms without remapping the field
DroneDash and Geodnet are forming Geodash Aerosystems and developing a crop-spraying drone for large farms that does not need the field to be remapped or the route rebuilt before every flight.
Commvault launched AI Protect — a “Ctrl+Z” for AI agents in enterprise clouds
Commvault introduced AI Protect, a tool for rolling back AI agent actions in the cloud that helps restore data, configurations, and access after erroneous changes.