Full back catalogue
Every AI and data science story we've published, newest first. 32,953 articles across 660 pages. For the live feed, head to the newsroom or browse by topic hub.
Higharc is a useful data point for anyone skeptical that generative AI transfers cleanly into heavy, physical industries: its wedge is not chat but structured 3D spatial data. On June 30, 2026 the company said it raised…
For AI teams, the hardest enterprise data to use is not documents or CRM records; it is the fast, unstructured conversation traffic that has migrated off email and into WhatsApp, iMessage, Signal, and WeChat, where…
Search Engine Journal contributor **Alexander Kesler** published a four-step verification protocol on **July 1, 2026** designed to catch AI errors, what he calls a **"cognitive mirage,"** before they shape business…
Infobip published its **India Digital CX Report 2026**, finding that Indian enterprises lead Asia on channel adoption but lag on the integration layers that agentic CX requires. ANI press coverage of the release cites…
Bank of England Deputy Governor **Sarah Breeden** warned at the ECB's Sintra Forum on June 30, 2026 that autonomous **AI agents** could "amplify volatility in stress" and trigger a "market meltdown," and said existing…
UBS analysts led by **Karl Keirstead** found that roughly **60%** of enterprises are now throttling AI spending in some way, based on conversations with about a dozen enterprise IT executives, according to a UBS…
The **Godot Foundation** announced on **June 30, 2026** that it is tightening contribution rules for AI-assisted code, banning "autonomous AI agent use or vibe coding" and prohibiting AI from generating "substantial…
**KPMG Australia** confirmed in February 2026 that **28** staff had used AI tools to cheat on mandatory internal AI-ethics training exams since July 2025, with the most senior case - a registered company auditor at…
**Jamf** announced general availability of **AI Governance**, a new capability in Jamf for Mac that discovers, enforces policy on, and produces audit-ready reports for AI tools running on managed Macs, with initial…
For AI infrastructure teams, Blue Energy's gas-to-nuclear model addresses nuclear's financing gridlock directly: natural gas turbines deliver firm power and early revenue while BWRX-300 SMRs move through a parallel…
For AI/DS practitioners: Samsung and SK Hynix supply most of the world's HBM memory for AI training. Political pressure on chipmaker profit distribution adds macro risk to that supply chain. South Korean political…
For AI/DS practitioners building procurement, payments, or spend-analytics systems, the shift of B2B purchasing into cart-and-card experiences changes transaction data shape and latency, concentrating richer line-item,…
For practitioners, moving edge AI from prototype to production increasingly depends on coordinated hardware, software, and supply-chain ecosystems rather than isolated model improvements. According to PR Newswire and…
For practitioners, sustained AI-driven demand for chips, servers, and data-centre equipment is reshaping procurement and capacity planning across Asia, creating pockets of excess demand even as input-cost shocks…
For AI/ML practitioners: the LSST will stream roughly 7 million sky-change alerts per night, all classified by ML broker networks using CNNs and gradient-boosted forests in real-time - one of the largest production ML…
For engineering teams that depend on open-source components, the security math just changed: frontier AI models now surface vulnerabilities faster than volunteer maintainers can triage them, turning the software supply…
**Microsoft (MSFT)** shares closed at **$373.02** on June 30, 2026, roughly 50% below the average Wall Street 12-month price target of **$561** from 56 analysts, according to a July 1, 2026 FinanceFeeds scenario…
Samsung SDI CEO Choi Joo-sun's explicit call to become a 'true AI-native company' signals agentic AI adoption spreading into battery manufacturing - not just software and semiconductor sectors. At the company's 56th…
For AI and ML practitioners, reliably priced, large-scale clean power is becoming an infrastructure input as much as compute capacity; projects that aggregate utility-scale renewables can change the procurement and…
Japan's METI has formally committed up to 1 trillion yen ($6.2 billion) over five years to back Noetra -- a physical AI consortium anchored by SoftBank, Sony, NEC, and Honda, with approximately 44 companies expected to…
A **July 1, 2026** RealClearMarkets op-ed marking the 70th anniversary of the **1956 Dartmouth workshop** - where the term "artificial intelligence" was coined - argues regulators should be cautious about repeating…
Production reality is forcing AI-driven workforce decisions into reverse: Ford has rehired roughly 350 veteran engineers to address quality gaps that automated systems could not close, while Commonwealth Bank of…
For AI and data practitioners, public-security institutions signalling an intent to "master AI" usually increases demand for applied analytics, operational ML tools, and ethics-compliance capabilities across procurement…
**South Korea's** newly appointed Prime Minister **Han Seong-sook**, a former Naver CEO, vowed on her first day in office on **July 1, 2026** to accelerate the country's AI push, saying "we will make bolder investments…
For AI infrastructure planners: India is emerging as a priority renewable-energy market precisely because AI-driven data-center buildout demands vast clean-power capacity. Brookfield Asset Management says India's data…
Panasonic Holdings' stock has more than doubled in 2026, pushing its market valuation to a record **JPY11.5 trillion (about $71 billion)**, as Bloomberg and The Japan Times report investors are pricing in rising demand…
Meta published research and open-sourced code and data for **Brain2Qwerty v2**, a non-invasive brain-computer interface that decodes typed sentences from magnetoencephalography (MEG) recordings at an average **61%**…
South Korea's **National Growth Fund**, a state-backed vehicle set to reach **KRW150 trillion** (about $96.3 billion) over five years, will scale up investment in the emerging physical AI sector, **Financial Services…
Second-quarter investor rotation pushed roughly **$2 trillion** in combined market-cap gains into **Intel**, **Micron**, and **Advanced Micro Devices**, as CNBC reports Wall Street chases AI-infrastructure exposure…
Samsung Electronics and SK Hynix pledged **3,200 trillion won ($2.07 trillion)** in combined investment tied to South Korea's national semiconductor push, including a new **800 trillion won** chip cluster of two fabs in…
Siteline, which tracks AI agent traffic, ran a simulated **Claude** agent through 534 attempts to find pricing and features across **100** top B2B software products and found the agent regularly failed to extract prices…
**John Ratcliffe**, director of the **CIA**, said at the **AWS Summit** in Washington on **June 30, 2026** that it would be "not misplaced to refer to" frontier AI capabilities as "akin to digital nuclear weapons," per…
Korea's Ministry of Trade, Industry and Resources formally denied online rumors that Seoul sent letters to Samsung Electronics and SK Hynix urging them to form a government-led think tank for channeling chipmaker…
For practitioners, this is a bet that the real constraint on enterprise AI is deployment capacity, not model access. On June 30, 2026, AWS announced a dedicated Forward Deployed Engineering organization backed by a $1…
For healthcare ML teams, this is the clearest signal yet that patient-facing large language models can clear FDA review when the task is narrowly scoped. On June 25, 2026, UpDoc publicly debuted what it calls the first…
Google released **Nano Banana 2 Lite** (`gemini-3.1-flash-lite-image`) on June 30, its fastest and cheapest Gemini image model, generating text-to-image outputs in about **4 seconds** at **$0.034 per 1K-resolution…
Anthropic introduced **Claude Sonnet 5** in a **June 30, 2026** blog post, calling it "the most agentic Sonnet model yet" and an upgrade to the Sonnet family. For AI practitioners building agents, lower-cost models that…
A March 2026 opinion essay by Cincinnatus Hibbard, syndicated across Bay Area alt-weeklies (East Bay Express, Bohemian, Pacific Sun), proposes using Claude Projects or Google NotebookLM to create domain-specific AI…
Overclocking team **Team OGS** pushed a GALAX **HOF OC LAB** GeForce **RTX 5090D** to **4002 MHz**, becoming the first NVIDIA Blackwell-architecture GPU to break the symbolic 4 GHz core-clock barrier, according to…
For teams building AI-for-science systems, the bottleneck is no longer recalling facts or running a fixed pipeline; it is the higher-order judgment of deciding which analysis a messy dataset can actually support.…
The strategic question for government and critical-infrastructure AI has shifted from which model is smartest to who owns the weights and where they run. Palantir and Nvidia answered on June 29, 2026, launching a joint…
**Microsoft** held a neighborhood meeting in **Person County, North Carolina** on Tuesday to discuss its planned data center campus and gather community input. The project involves a **1,385-acre** site near Roxboro.…
OpenCV.org and AWS launched **COOL (Cloud-Optimized OpenCV Library)** on the AWS Marketplace for Graviton 2, 3, and 4 instances, delivering roughly **30% performance gains** and proportional cost reductions over a…
Zvi Mowshowitz's June 30 weekly roundup synthesizes the most consequential AI policy and access developments in months. The core story: US export controls halted Claude Fable 5 and Mythos 5 on June 12 for national…
The LG C6H OLED Evo AI (2026) review from Digital Trends asks whether this is the first meaningful C-Series upgrade in years. The C6H's main differentiator is its RGB Tandem 2.0 OLED panel - the same used in the…
AMD announced the **Versal Premium Gen 2 Memory on Package** adaptive SoC family, integrating up to **32 GB** of `LPDDR5X` on-package memory with up to **288 GB/s** of bandwidth, according to AMD's blog post.…
A working paper from **Ramp** and workforce-analytics firm Revelio Labs, linking Ramp corporate card and bill-pay data to Revelio workforce records for **21,559** U.S. firms from 2021 to early 2026, finds that the…
**Vibe Island**, a native macOS Swift app, turns the MacBook notch into a persistent control panel for terminal-based AI coding agents, surfacing permission requests, tool-use approvals, and agent status without…
**MDOTM Ltd.** closed a **$27 million** growth equity round led by **Expedition Growth Capital**, bringing the AI investment-platform maker's total funding to **$36.5 million**, according to Business Wire and Axios.…
CoreWeave became the first AI cloud provider to bring up and fully validate NVIDIA's Vera Rubin NVL72 rack platform on its cloud, the company announced on **June 1, 2026**. Each NVL72 rack packs **72 Rubin GPUs** and 36…
