Coverage index

Follow the whole AI market.

Move from model releases to infrastructure, agents, security, robotics, and the market shifts connecting them.

Inside the index

Eight core news desks organize the daily flow. Four cross-cutting topics reveal stories that span categories.

Browse coverage Cross-cutting

Daily structure

Primary coverage

The core desks used to organize every story, briefing, and archive view.

01 / Coverage desk

Models

Model launches, pricing, benchmarks, and capability shifts.

View all Models

Open source

DeepSeek Releases 305B Vision Weights Under MIT License, but Local Use Needs Serious Hardware

The release gives developers the model files and serving references needed to run or adapt DeepSeek-V4-Flash-Vision-Exp independently. Its 168GB size means that freedom is aimed more at infrastructure-equipped teams than typical individual users.

5 hours ago3 min read

Launch

Google’s TimesFM-3 Adds Known Future Events to Zero-Shot Forecasting

The 330-million-parameter model can combine related data streams with inputs such as promotion calendars and weather forecasts. Google reports leading results on three public benchmarks; performance on an organization’s own data remains the practical open question.

7 hours ago3 min read

Launch

Microsoft Releases GigaPath-Flash, Reporting Roughly 50x Less Compute for Cancer Research

The open-weight research release is designed for repeated analyses across large cancer cohorts, while its performance claims remain limited to initial benchmarks and cohorts.

7 hours ago2 min read

Platform shift

OpenAI Maps a Third Era of Persistent AI Coworkers, With Access and Reliability Still in the Way

The proposed shift would turn AI from a prompt-by-prompt helper into a longer-running collaborator, but its usefulness depends on shared context, workplace-system access, and dependable controls.

Yesterday3 min read

Benchmark

Chatbots Debunked Foreign Falsehoods About 75% of the Time, Beating Search Results

The result makes chatbots a potentially stronger starting point for investigating state-spread claims than a search-results page. It does not make their answers self-validating: language, citations and product design still shape the outcome.

Yesterday4 min read

Benchmark

Google’s WikiSkill Lifts Agent Benchmarks by Giving Models a Memory of Failed Work

The framework turns task traces into reusable instructions instead of changing a model’s training, offering a practical route to more capable agents while leaving uneven task gains and cross-model transfer as constraints.

Aug 29, 20263 min read

Launch

Gnani Launches 30B Evon Model and Self-Hosted Agent Stack for Indian Institutions

The open-weight release couples Indian-language support with tools intended to keep sensitive data inside a customer’s own infrastructure.

Aug 28, 20263 min read

Open source

Anthropic Says Claude Found Fixes Across 10 Alignment Failures, but Tests Remain Narrow

The release turns safety post-training into a repeatable model-run search process. Its value now depends on whether those benchmark gains survive broader tests and later training.

Aug 28, 20263 min read

02 / Coverage desk

Products

New AI products, features, apps, and platform releases.

View all Products

Launch

AWS Agent Registry Reaches GA With Approval Gates for Enterprise Agent Discovery

The service makes approval status a condition of discovery, separating a complete administrative record from the capabilities available for enterprise reuse.

1 hour ago3 min read

Platform shift

Google Sets Gemini Notebook Limits by Task Complexity, With Five-Hour Resets

The new system changes when capacity returns and what counts against it, while offering alternate outputs and delayed generation when a request exceeds a user’s available compute.

3 hours ago2 min read

Launch

Pixel Watch 5 Reaches Stores With Offline Gemini Actions and Europe-First Breathing Alerts

Google’s newly available smartwatch pairs proactive assistance with an emergency-response feature, while its longer-term health summaries still have no specific release date.

3 hours ago3 min read

Platform shift

Google Wires Gemini Live to Spark, Letting Voice Requests Run Across Its Apps

Gemini Live is becoming a single spoken entry point for background workflows, inbox actions and personal-data lookups. The useful test is whether its handoffs between those systems feel dependable enough for people to connect their accounts.

4 hours ago3 min read

Partnership

RelativityOne Connects Microsoft Copilot to Legal Administration Through MCP

Legal teams can now run specified RelativityOne administration and reporting tasks from Copilot, but the rollout splits workflow creation and reporting across two Copilot surfaces.

8 hours ago3 min read

Platform shift

Unitree’s $4,017 Go2 Pro Makes Robot Dogs Cheaper, Not Yet Everyday Tools

A real-world Go2 Pro outing captures both sides of Unitree’s proposition: legged robots now cost far less, but heat, battery life, durability and a clear job still constrain them.

9 hours ago3 min read

Partnership

Broadcom Validates Five AI Model Families for VMware’s On-Premises Model Service

The update gives enterprises a path to run models from NVIDIA, Google, NEC, Alibaba Cloud and Z.ai inside their own infrastructure, with new sharing, routing and code-isolation services around them.

10 hours ago2 min read

Platform shift

Google Makes Dreambeans Free for U.S. Accounts, Opening Its Data-Connected Daily Feed

The experiment’s appeal—and constraint—is that its recommendations rely on permission to combine signals from Gmail, Photos, Calendar, YouTube and Search.

17 hours ago2 min read

03 / Coverage desk

Tools

AI apps, agents, automations, creator tools, and builder workflows.

View all Tools

Enterprise adoption

AWS’s Bedrock Chat Blueprint Carries Tenant Filters Across Every Retrieval Hop

The reference design hands document processing and answer generation to Bedrock, but its security outcome still depends on application code deriving retrieval boundaries from verified identities.

1 hour ago3 min read

Open source

Google Showcases HEIR, an Encrypted-AI Compiler, With Four Private-Inference Demos

The open-source tool is meant to turn a cryptography-heavy conversion task into a development workflow. Its four CPU-based examples show possible uses; promised accelerator latency results remain ahead.

1 hour ago2 min read

Platform shift

AWS Publishes Four-Stack Bedrock Blueprint for Observable Agentic Retrieval

The sample separates cross-knowledge-base routing from retrieval within a selected knowledge base, but ties the pattern to AWS-managed services and their operating requirements.

1 hour ago3 min read

Platform shift

Google DeepMind’s Farm AI Reaches Six African Countries as FAO Plans Data Integration

The expansion puts Google DeepMind’s agricultural insights in Kenya, Uganda, Ghana, Rwanda, Nigeria and Zambia. A planned FAO integration would test whether those outputs can help countries update crop maps and agricultural statistics faster.

2 hours ago2 min read

Benchmark

AI ECG Tool Flags Heart Risk in Two Seconds, but Still Needs the Scan That Diagnoses It

The reported results suggest routine electrical heart traces could help prioritize scarce ultrasound appointments. The system’s stated role, however, stops short of confirming disease or ruling it out.

8 hours ago2 min read

Launch

Modellix Adds 28+ Models, Including $0 Options, to DeepSeek Harness Beta

The beta gives users of the rapidly adopted coding agent one model-selection layer and web-tool routing, while its free-call pitch still requires a Modellix account and funded balance to create an API key.

10 hours ago3 min read

Platform shift

Broadcom Releases AgentMinder to Check AI Agent Tool Calls at Runtime

The new control layer is designed to evaluate each requested agent action before it reaches company systems, shifting governance from static permissions toward live policy enforcement.

11 hours ago2 min read

Launch

Replica Cyber Ships Isolated AI Agents With Runtime Controls for Regulated Firms

The product makes containment, intervention and post-incident reconstruction part of the runtime environment. Its unanswered test is whether those controls hold when agents connect to the systems that make them valuable.

12 hours ago2 min read

04 / Coverage desk

Startups

Funding, launches, acquisitions, pivots, and company formation.

View all Startups

Funding

Clipto Raises $15M at $250M Valuation to Turn Local Files Into AI-Agent Context

The profitable file-search startup says its MCP connection lets authorized AI tools retrieve selected local data. Its bet is that cross-format search can remain a product category even as large platforms build similar features into their own apps.

7 hours ago3 min read

Launch

Jeff Dean Leaves Google to Build Discovery Loop, a Science AI Startup Backed by Radical Ventures

The new public-benefit company has an unusually experienced founding team and support from both Radical Ventures and Alphabet. Its central technical question is still unanswered: what system will turn the ambition to automate research into a usable product?

23 hours ago4 min read

Funding

IndiGo Ventures Joins Sarvam’s $1.5B Series B, Linking Airline Work to Ownership

The airline’s venture arm is backing a supplier it plans to use for customer service, staff productivity and operations—where dependable deployment will be the real test.

Aug 28, 20262 min read

05 / Coverage desk

Business

Enterprise adoption, partnerships, earnings, and AI strategy.

View all Business

Platform shift

Carrier’s Data-Center Orders Rose More Than 300%, Bringing Cooling Into AI Buildouts

The order increase gives the AI infrastructure buildout a concrete thermal-equipment demand marker, though it does not reveal the size of Carrier’s data-center revenue.

22 minutes ago2 min read

Partnership

Intuit Links QuickBooks and Mailchimp to Perplexity Computer Through MCP

The MCP connection moves Perplexity Computer closer to business-software tasks, with announced uses spanning invoicing, payroll information and marketing analysis.

2 hours ago2 min read

Acquisition

Imagenet Buys Analytica to Add Data Engineering and AI to Payer Operations

The purchase puts data engineering, governance and analytics beside Imagenet’s payer-administration workflows. The stated goal is to turn the resulting operational data into more useful intelligence without replacing clients’ core technology.

2 hours ago2 min read

Enterprise adoption

Caterpillar Commits $100M to Retrain 118,000 Employees as AI Moves Beyond Mining

The five-year training commitment puts workers at the center of Caterpillar’s automation strategy. Its harder task is fitting more autonomous equipment into established jobsite roles and routines.

4 hours ago2 min read

Acquisition

Deloitte Buys Wavicle Assets to Add Databricks Engineering to Its AI Practice

The asset purchase brings data and AI engineers with cross-cloud Databricks experience into Deloitte’s consulting organization. The intended payoff is faster, industry-specific AI deployment, though those customer results remain a future objective.

5 hours ago2 min read

Funding

Lambda Adds $1B in Short-Dated Debt for Nvidia Chips Bound for Microsoft

The borrowing adds another customer-specific hardware deployment to Lambda’s balance sheet, while its reported equity plans remain unfinished.

6 hours ago2 min read

Funding

Nvidia Puts $3.5B Into MediaTek to Keep Custom AI Chips in Its Data-Center Stack

MediaTek’s planned accelerators are meant to run beside Nvidia hardware, giving Nvidia’s interconnect and rack architecture a role as customers pursue their own silicon.

8 hours ago2 min read

Platform shift

Micron Trades at Just Above 6x Forward Earnings as Contracts Test Its Memory Cycle

Long-term customer agreements may curb both peak-price gains and later declines, but new supply and uncontracted sales leave the durability of Micron’s earnings open to debate.

11 hours ago2 min read

06 / Coverage desk

Infrastructure

The compute, chips, data centers, networks, and power systems behind AI.

View all Infrastructure

Viral

Vance Backs Data Centers With a Condition: Add Power, Not Local Bills

The vice president’s proposed bargain reframes the infrastructure fight: builders can pursue AI capacity, but must show communities they will expand electricity supply rather than deepen grid costs.

5 hours ago3 min read

Partnership

Humain, DataVolt Plan 100 MW AI Data Center at NEOM’s Oxagon

The expected-2028 facility gives Saudi-backed Humain a defined role in DataVolt’s planned Oxagon campus, but all of the capacity remains prospective.

9 hours ago2 min read

Platform shift

OpenAI Reportedly Buys Tens of Thousands of Macs for Computer-Use Agent Training

The reported purchase points to dedicated capacity for agent training alongside GPU systems, while its cost and performance tradeoffs remain unclear.

14 hours ago2 min read

Platform shift

McKinsey Sees $7 Trillion Data-Center Buildout by 2030, Raising the Stakes for Power Gear

The forecast is an estimate, not committed spending. But rising orders, backlogs and liquid-cooling deployments show why electrical distribution and thermal management are becoming central to the AI infrastructure race.

17 hours ago2 min read

Platform shift

Huawei Proposes 2,008-Chip Egypt AI Build as U.S. Pursues a Rival Offer

Egypt has a defined Huawei proposal and a prospective U.S. response, but the American side has not announced a consortium or disclosed its hardware, financing or technology-use terms.

Yesterday2 min read

Funding

Blue Owl Leads $2.4 Billion IREN GPU Financing Tied to Hardware Delivery

IREN has a large debt commitment for its Canadian AI-cloud buildout, but the amount becomes available alongside delivered and commissioned hardware—not as one immediate cash payment.

Yesterday2 min read

Security risk

X Finds 200,000-Account Bot Farm in Suspected Chinese AI Policy Influence Push

The platform removed accounts that broke its rules, but the content it identified landed in a U.S. fight over power costs, grid capacity and local development that is already politically charged.

Yesterday2 min read

Platform shift

China Pushes AI Computing West as Guizhou’s Data-Center Buildout Tests Local Gains

Cooler weather, renewable power and open land make China’s west a natural destination for AI infrastructure. The harder question is whether capital-intensive server farms can deliver durable local incomes without worsening public debt.

Aug 29, 20263 min read

07 / Coverage desk

Policy

Lawsuits, regulation, copyright, safety, and government action.

View all Policy

Security risk

Google Urges Modular Red-Team Agents as Cyberattacks Move Toward Machine Speed

Google’s prescription is deliberately incremental: automate parts of a conventional exercise first, then join them into a simulation. Its warning is more urgent than its evidence base, however, because concrete intelligence on malicious agents remains limited.

17 minutes ago3 min read

Regulation

Google Calls for AI Training Exceptions and Creator Opt-Outs at Singapore IP Forum

Google’s policy case would preserve broad access to publicly available training material while giving rights holders tools to refuse participation. Its harder task is showing that technical controls and voluntary deals can satisfy creators without changing the legal default.

4 hours ago3 min read

Regulation

EU Classifies ChatGPT as Search Engine, Starts Four-Month DSA Compliance Clock

The designation places an AI chatbot in the EU’s top platform-law tier, requiring risk reviews, mitigation and outside scrutiny alongside Reddit and Roblox.

12 hours ago2 min read

Security risk

FSB Warns G20: Frontier AI Cyber Risk Could Undermine System-Wide Market Confidence

Andrew Bailey’s new warning turns frontier-model safety into a financial-resilience problem: shared providers and cross-border cyber incidents can make a local failure spread further.

18 hours ago2 min read

Security risk

Israeli Government-Backed Hanover Site Published 560,000 Words to Influence AI Answers

The site’s foreign-agent filings and machine-readable design make its purpose unusually visible. Its effect on chatbot behavior, or on future model-training data, remains unproven.

Aug 29, 20264 min read

Platform shift

California Webinar Urges Water Rules as Tulare, Kings AI Sites Await Review

The immediate choice for rural hosts is whether developers must disclose water use and cooling plans before local boards consider a project.

Aug 29, 20262 min read

Security risk

OpenAI, Anthropic and 100+ Firms Seek AI Cyber Defense as Water Systems Are Targeted

The coalition wants governments to put defensive AI in the hands of public-service operators, but its warning offers neither funding nor a delivery timetable as attackers already use AI-assisted scripts against water-system controls.

Aug 29, 20263 min read

Lawsuit

Apple’s $250M Siri Settlement Could Open Claims September 21 for Eligible iPhone Buyers

The proposed 90-day filing window gives affected U.S. buyers a path to compensation, but court approval and the final claim requirements are still pending.

Aug 29, 20262 min read

08 / Coverage desk

Culture

Viral AI moments, creative AI, public reaction, and social shifts.

View all Culture

Platform shift

Claims Adjusters’ AI Backlash Hits 98% Negative Reviews as Insurers Automate Claims

The backlash is not simply resistance to a new tool. Adjusters describe being held accountable for faulty outputs while job losses and fewer entry-level openings sharpen fears that automation is remaking the profession.

13 hours ago3 min read

Viral

Trump Posts AI Kharg Strike Videos After U.S. Hits Launchers on Larak

The posts place simulated destruction at one of Iran’s key oil-export hubs hours after a limited U.S. operation targeted rocket launchers on a different island.

18 hours ago2 min read

Platform shift

MIT Weighs Education Overhaul as AI Handles Almost Any Written Undergraduate Assignment

The committee’s finding puts a harder assessment question before universities: when AI can produce credible written answers, what can a submitted assignment establish about a student’s learning?

Yesterday2 min read

Benchmark

JAMA Paper Says Autonomous AI Could Surpass Physicians and AI-Assisted Care by 2030

The contested forecast recasts clinical AI from a physician tool into a possible replacement, while real-world patient use and medical training remain unsettled.

Aug 28, 20262 min read

Across the market

Cross-cutting topics

Follow themes that move through products, companies, policy, and technical systems at the same time.

09 / Coverage desk

Agents

AI systems that plan, use tools, take actions, and complete work.

View all Agents

Security risk

Google Urges Modular Red-Team Agents as Cyberattacks Move Toward Machine Speed

17 minutes ago3 min read

Platform shift

AWS Publishes Four-Stack Bedrock Blueprint for Observable Agentic Retrieval

The sample separates cross-knowledge-base routing from retrieval within a selected knowledge base, but ties the pattern to AWS-managed services and their operating requirements.

1 hour ago3 min read

Launch

AWS Agent Registry Reaches GA With Approval Gates for Enterprise Agent Discovery

The service makes approval status a condition of discovery, separating a complete administrative record from the capabilities available for enterprise reuse.

1 hour ago3 min read

Funding

Clipto Raises $15M at $250M Valuation to Turn Local Files Into AI-Agent Context

7 hours ago3 min read

Launch

Modellix Adds 28+ Models, Including $0 Options, to DeepSeek Harness Beta

10 hours ago3 min read

Platform shift

Broadcom Releases AgentMinder to Check AI Agent Tool Calls at Runtime

The new control layer is designed to evaluate each requested agent action before it reaches company systems, shifting governance from static permissions toward live policy enforcement.

11 hours ago2 min read

Launch

Replica Cyber Ships Isolated AI Agents With Runtime Controls for Regulated Firms

12 hours ago2 min read

Platform shift

OpenAI Reportedly Buys Tens of Thousands of Macs for Computer-Use Agent Training

The reported purchase points to dedicated capacity for agent training alongside GPU systems, while its cost and performance tradeoffs remain unclear.

14 hours ago2 min read

10 / Coverage desk

Security

Vulnerabilities, attacks, defenses, and the security consequences of AI.

View all Security

Security risk

Google Urges Modular Red-Team Agents as Cyberattacks Move Toward Machine Speed

17 minutes ago3 min read

Security risk

FSB Warns G20: Frontier AI Cyber Risk Could Undermine System-Wide Market Confidence

Andrew Bailey’s new warning turns frontier-model safety into a financial-resilience problem: shared providers and cross-border cyber incidents can make a local failure spread further.

18 hours ago2 min read

Security risk

Claude, Codex and Hermes Ran Unowned Package Commands Inside Corporate Networks

The experiment showed that trusted-looking documentation can become an execution path when an agent is allowed to install software. It did not establish confirmed infections or stolen production data.

Yesterday3 min read

Security risk

X Finds 200,000-Account Bot Farm in Suspected Chinese AI Policy Influence Push

The platform removed accounts that broke its rules, but the content it identified landed in a U.S. fight over power costs, grid capacity and local development that is already politically charged.

Yesterday2 min read

Security risk

Israeli Government-Backed Hanover Site Published 560,000 Words to Influence AI Answers

The site’s foreign-agent filings and machine-readable design make its purpose unusually visible. Its effect on chatbot behavior, or on future model-training data, remains unproven.

Aug 29, 20264 min read

Security risk

Sony and Warner Sue Anthropic Founders Over Claude Data, Seeking Up to $150,000 Per Work

The publishers’ case reaches beyond Claude’s alleged lyric outputs to the route by which copyrighted material entered Anthropic’s systems—and seeks records that could tie those decisions to its co-founders.

Aug 29, 20263 min read

Security risk

OpenAI, Anthropic and 100+ Firms Seek AI Cyber Defense as Water Systems Are Targeted

Aug 29, 20263 min read

Security risk

TeamPCP’s Trivy Attack Hit 2,500 Organizations; AI Tool Backdoors Are Claimed

The LiteLLM-linked exposure shows how a poisoned dependency can yield reusable developer access. Claims of persistence in AI coding-tool settings would make cleanup harder, but they remain uncorroborated.

Aug 28, 20263 min read

11 / Coverage desk

Open Source

Open models, open software, licenses, and the communities building around them.

View all Open Source

Open source

Google Showcases HEIR, an Encrypted-AI Compiler, With Four Private-Inference Demos

1 hour ago2 min read

Open source

DeepSeek Releases 305B Vision Weights Under MIT License, but Local Use Needs Serious Hardware

5 hours ago3 min read

Launch

Microsoft Releases GigaPath-Flash, Reporting Roughly 50x Less Compute for Cancer Research

The open-weight research release is designed for repeated analyses across large cancer cohorts, while its performance claims remain limited to initial benchmarks and cohorts.

7 hours ago2 min read

Open source

LAION Releases 10-Million-Hour Video Corpus for Research, Excluding Commercial Use

The release brings video, sound and still frames into one open research resource, while synthetic labels, English-heavy sourcing and noncommercial terms narrow how it can be used.

Aug 29, 20262 min read

Launch

Gnani Launches 30B Evon Model and Self-Hosted Agent Stack for Indian Institutions

The open-weight release couples Indian-language support with tools intended to keep sensitive data inside a customer’s own infrastructure.

Aug 28, 20263 min read

Open source

Lightwheel Releases 10,000 Hours of Human Video for Robots, With 90,000 Still to Come

The release gives robotics teams a sizable pretraining resource and a way to test two collection designs, but its value will depend on annotation quality, task fit, and robot-specific data added afterward.

Aug 28, 20263 min read

Launch

NVIDIA’s TensorRT Model Connect Skips ONNX, but x86_64 Users Must Build From Source

The open-source preview turns model conversion into a versioned deployment artifact, separating engine construction from native inference while imposing an ARM-first installation path.

Aug 28, 20263 min read

Open source

Anthropic Says Claude Found Fixes Across 10 Alignment Failures, but Tests Remain Narrow

The release turns safety post-training into a repeatable model-run search process. Its value now depends on whether those benchmark gains survive broader tests and later training.

Aug 28, 20263 min read

12 / Coverage desk

Robotics

Embodied AI moving through factories, roads, homes, and the physical world.

View all Robotics

Platform shift

Unitree’s $4,017 Go2 Pro Makes Robot Dogs Cheaper, Not Yet Everyday Tools

A real-world Go2 Pro outing captures both sides of Unitree’s proposition: legged robots now cost far less, but heat, battery life, durability and a clear job still constrain them.

9 hours ago3 min read

Platform shift

Nvidia’s $10B Physical-AI Business Bets on a Full Stack—and an Open China Trade Lane

The company’s hardware, robot models, simulation tools and startup investments form a platform strategy that can serve competing manufacturers. Its ability to keep doing so depends partly on policy that remains unsettled.

19 hours ago3 min read

Platform shift

Jensen Huang Rejects Robot Tax as Bill Gates Warns AI Could Erode Jobs and Tax Revenue

The disagreement is not over whether AI will reshape work. It is over whether policy should slow automation and fund a safety net before productivity gains create the new jobs Huang expects.

Yesterday3 min read

Funding

Andreessen Horowitz Raises $1.1B for AI’s Physical Infrastructure, From Chips to Robotics

The first hardware-specific vehicle in the firm’s portfolio turns an existing AI investment program into a dedicated mandate spanning compute, power, facilities and physical machines.

Yesterday2 min read

Platform shift

Sanctuary AI Puts Its Robot Brain on Existing Machines After a 99.5%-Plus Test

The company is pitching dexterous automation that can fit installed robot systems, but its automotive wire-plugging result has not yet become a production deployment.

Aug 29, 20263 min read

Open source

Lightwheel Releases 10,000 Hours of Human Video for Robots, With 90,000 Still to Come

Aug 28, 20263 min read

Enterprise adoption

ICE Plans $1M–$2M SPOT Robot Buy for Hazard Assessments and Remote Inspection

The proposed purchase frames robot dogs as a way to inspect risky scenes from a distance, while earlier police deployments show both practical use and public backlash.

Aug 28, 20263 min read

Platform shift

Meta Tests Data-Center Repair Robots, but Cable Bots Still Need Human Supervision

Meta already uses robots for bounded logistics work. The harder test is whether supervised machines can make delicate maintenance tasks fast and reliable enough to change how data centers are staffed.

Aug 28, 20263 min read

One briefing / every desk

Get the shifts that cross categories.

Subscribe Latest stories