Today's AI Pulse
Anthropic shuts down flagship models after U.S. export‑control order
Anthropic abruptly disabled its Fable 5 and Mythos 5 models to comply with a U.S. export‑control directive, highlighting the fragility of businesses that rely on closed‑source AI services. The shutdown triggered a sharp rally in Chinese open‑source offerings such as MiniMax and Zhipu as investors bet on downloadable models that can run on‑premise.
Also developing:
By the numbers: SpaceX to acquire Cursor in $60B all‑stock deal
Level AI Launches AI Workers, Purpose-Built AI Agents for Customer-Centric Teams
Level AI introduced AI Workers, a suite of purpose‑built AI agents that automate research, analysis, and planning for contact‑center teams. The launch includes eight specialized workers—coaching, conversation research, executive research, and more—leveraging the platform’s existing customer‑intelligence data. Over 100 enterprise contact centers, such as Smartsheet and VistaPrint, have already run more than 25,000 Worker executions with strong user feedback. By linking QA, CRM, and sentiment signals in a single layer, the solution eliminates parallel data pipelines and speeds insight delivery.
China Targets Clean-Energy AI Compute by 2030
NEA with NDRC, MIIT, and the National Data Administration issue plan: sets 29 tasks around reliable energy supply for AI compute, green/low-carbon data centers, compute-power coordination, energy-sector AI scenarios, data value, and model innovation, with a 2030 goal of substantially...
Generative AI System Could Cut Animal Testing by Up to 50% in Preclinical Research
Researchers at Goethe University Frankfurt, Philipps University of Marburg and the Fraunhofer Institute for Translational Medicine and Pharmacology have unveiled genESOM, a generative AI tool that creates synthetic preclinical data. Trained on small experimental sets, it can mimic real laboratory...
AI Scraping Has Become Its Own Media Business
AI companies rely on a shadow industry of data scrapers that harvest publisher content without permission and sell it to large language model providers. A recent Digiday‑cited report identified at least 21 such firms, including Parallel AI, Exa and Bright...
BluSky AI Inc. Announces Launch of Regulation A+ Offering to Accelerate U.S. AI Infrastructure Expansion
BluSky AI Inc. announced a Regulation A+ offering that lets individual and accredited investors participate with as little as $1,000. The capital will fund the rollout of more than ten modular AI Factory sites across the United States, targeting a combined...
Bad Data + Cognitive Surrender: The Emerging Enterprise AI Threat
Enterprise adoption of generative AI has surged, with over 65% of firms using it in at least one function. However, the article warns that bad data paired with AI’s persuasive confidence—termed cognitive surrender—creates a hidden threat. Flawed data leads to...
Google Cloud Is Hiring an Army of AI Deployment Engineers
Google Cloud announced 59 new forward‑deployed AI engineering roles across the U.S. and key global hubs, offering base salaries between $127,000 and $183,000. The move is backed by a $750 million investment to expand a high‑touch engineering model for enterprise customers....
Nitro Launches Nitro Automate, Integrating Intelligent Document Automation Into Any Workflow, System, and AI Agent
Nitro Software announced early‑access to Nitro Automate, an intelligent document automation platform that embeds Nitro’s enterprise‑grade processing capabilities directly into AI agents, CRM, ERP, HR, and custom applications. The solution can programmatically handle thousands of documents per second, supporting conversion,...
Infoblox and GoDaddy Support Open Standards for AI Agent Discovery, Identity and Verification
Infoblox and GoDaddy announced joint support for two open standards—DNS‑AID and Agent Name Service (ANS)—that enable AI agents to be discovered, identified, and verified using the existing Domain Name System and public‑key infrastructure. DNS‑AID, advancing as an IETF draft, lets...
AI to Infest Eight in Ten Premium Phones Within Two Years
Counterpoint Research predicts that over 80% of premium smartphones will embed agentic AI capabilities by 2027, while a similar share of wearables will be AI‑enabled by 2032. The shift is driven by new chipsets such as MediaTek’s Dimensity 9400 and...
Real-World Evidence Shows Generative AI Is Making Human Creative Output More Uniform
A new systematic review and meta‑analysis of 19 empirical studies shows that using generative AI tools makes human creative output more uniform across users. While AI can boost individual performance, the pooled data reveal a statistically significant homogenization effect, especially...
Understanding AI Realities for Leaders at Data Summit 2026
At Data Summit 2026 in Boston, the new Data + AI Leadership Forum hosted Ryan Frederick’s session on practical AI implementation. Frederick warned that many firms rush adoption without a data foundation, conflate pilots with production, and lack clear ownership, leading to costly...
AI Risks and Mitigation Strategies at Data Summit 2026
At Data Summit 2026 in Boston, Nicole Janeway Bills highlighted that only 43 % of firms have formal AI governance policies, underscoring a widening risk gap as AI adoption accelerates. She identified ten critical AI risks, from data privacy and bias...
Bloomberg Vault Integrates BSpeech, Unlocking Multi-Language Voice Transcription Capabilities Across 50+ Languages for Compliance Professionals
Bloomberg announced that its BSpeech multi‑language voice transcription engine is now integrated into Bloomberg Vault, the firm’s communications governance platform. The add‑on converts recorded calls into searchable text in more than 50 languages, embedding transcripts directly into Vault’s surveillance and...
Exadel Launches Exadel Colleague, an Autonomous AI Delivery Peer Built to Solve Problems AI Coding Tools Leave Behind
Exadel unveiled Exadel Colleague, an autonomous AI delivery peer that works inside existing engineering workflows. Unlike traditional AI coding assistants, the platform accepts tickets, plans work, writes tests, builds code, and submits validated pull requests without human intervention between assignments....
8×8 AI Studio Adds OpenAI’s GPT Realtime 2 to Support Production Voice Agents
8×8 AI Studio announced early‑availability support for OpenAI’s GPT Realtime 2, the latest voice‑AI model. The upgrade brings GPT‑5‑class reasoning, a 128K context window, and automatic Realtime‑Whisper transcription to live voice agents. Customers can opt‑in per agent, with a new reasoning‑effort...
Twilio CFO Viggiano Says AI Will Turn Platform Into Enterprise ‘Nervous System’
Twilio chief financial officer Katherine Viggiano told Fortune that the company is moving beyond its communications roots to become the connective tissue for AI agents. She highlighted a 20% jump in Q1 revenue to $1.41 billion and a raised full‑year outlook...
Quisitive Becomes Microsoft Frontier Partner to Drive Secure AI Transformation
Quisitive announced it has achieved Microsoft Frontier Partner status, joining an elite group of firms with deep AI expertise on the Microsoft Cloud. The milestone follows its recent Microsoft Copilot Specialization and aligns with the launch of Microsoft 365 E7,...
US, China to Coordinate on AI Threats
U.S. and Chinese delegations met in Beijing to draft a protocol of AI guardrails aimed at preventing non‑state actors from exploiting powerful models such as Anthropic’s Mythos. The talks underscore Washington’s desire to keep its technological lead while avoiding stifling...
Claude Goes Proactive: AI Acts Before You Ask
Anthropic says proactive AI in six months. Claude learns your workflow, starts doing things before you ask. No prompt. It just acts. Three hard problems for anyone building this.
Meibel Launches Document Intelligence: The Grounding Layer That Turns Complex Enterprise Documents Into Trusted, Structured Knowledge
Meibel unveiled Document Intelligence, a foundational data layer that converts unstructured enterprise documents into structured, confidence‑scored knowledge with full provenance and cross‑document relationship mapping. The platform preserves layout, tables, handwriting, and charts, allowing a single PDF to generate both searchable...
SUSE Targets AI, VMware Alternatives, and Open Source Infrastructure at SUSECON 2026
At SUSECON 2026 in Prague, SUSE unveiled AI Factory, an Industrial Edge IoT platform, and new VMware migration tools, signaling a shift from pure Linux distribution to a full‑stack open‑source infrastructure provider. The AI Factory bundles pre‑integrated AI stacks with...
Docker Launches Governed AI Runtime for Developer Laptops
Docker announced a governed AI runtime on May 12, giving security teams a single console to set policy for AI agents on developer laptops. The move targets the most exposed node in enterprise environments and pits Docker against a crowded...
Google Launches Premium Googlebook Laptops, Positioning AI‑first Hardware Against Chromebooks
Google announced the Googlebook, a new high‑end laptop series that runs on a proprietary AI‑focused OS and integrates Gemini Intelligence across the desktop. The line, built with five major OEMs, aims to offer a premium alternative to Chromebooks and expand...
Codex /Goal Lets AI Bots Work Nonstop Autonomously
Codex new `/goal` is literally INSANE You can easily transform your AI bots into a tireless 24/7 workforce that grinds for HOURS completely unguided 🤯 Codex desktop → Settings → Config, set goals = true Check out the ace cheat sheet + article...
To Tame Enterprise AI Chaos, Open Source Rallies Around a Standard Execution Layer
Enterprises are grappling with AI trust, governance, and cost as they deploy agentic models. Red Hat announced it will champion vLLM as a standard open‑source inference layer, leveraging its recent acquisition of Neural Magic for quantization and performance engineering. The...
Satya Nadella’s $190 B AI Infrastructure Bet Signals a New Era for Microsoft
Microsoft announced a $190 billion capital‑expenditure plan for 2026, anchored by a $625 billion commercial backlog. The spend, focused on GPUs, data‑center land and renewable power, reflects CEO Satya Nadella’s strategy to lock in AI revenue for years to come.
Atlassian Team ’26: The New Logic Of Work
Atlassian’s Team ’26 showcased its expanding Teamwork Graph, a context‑rich knowledge layer that links Jira, Confluence, and other tools, delivering a 70% cut in defect‑resolution time. The company highlighted the $1 billion annualized recurring revenue and 30% YoY growth of its...
Shield AI Teams with Taiwan’s Thunder Tiger to Deploy Hivemind on USVs
Shield AI and Taiwan’s Thunder Tiger Corp. have signed a memorandum of understanding to integrate Shield AI’s Hivemind autonomy software onto Thunder Tiger’s unmanned surface vessels. A live demonstration is scheduled for this summer, marking the first maritime deployment of...
How to Build and Deploy AI Agents in Under an Hour
AI agents have moved beyond chatbots in 2026, offering autonomous, multi‑step capabilities through reasoning, memory, and goal‑setting. The guide explains how to craft prompt contracts, leverage memory files, and select platforms such as Claude Code, Codex, OpenClaw or Antigravity. By...
Analysis: Verizon Digital Twin Changes How Operator Sees Itself
Verizon has completed a nation‑wide digital twin of its tens of thousands of cell sites, combining 3D models, drone‑captured imagery, and AI analytics. The system reduces storm‑damage assessments from several hours of manual climbs to roughly 15 minutes of automated...
Infobip Unveils AgentOS, an AI-Powered Platform for Customer Experience Automation
Infobip launched AgentOS, an AI orchestration platform that lets autonomous agents act across systems and messaging channels. The platform unifies customer data, communication tools, and execution logic to enable real‑time, goal‑driven interactions. At a Johannesburg event, Infobip highlighted use cases...
Gallagher Introduces New AI Tool to Advance the Future of Employer Benefits Decision‑Making
Gallagher announced an AI‑enabled benefits solution embedded in its Benefits & HR Consulting practice. The tool, built on the Avante platform and linked to Gallagher Drive, offers employees conversational guidance and employers real‑time analytics on utilization, cost drivers, and plan...
How Penn Medicine Plans to Use AI to Train Their Doctors
Penn Medicine secured a $1.1 million American Medical Association education grant to develop AI‑driven tools that give medical trainees real‑time feedback on clinical reasoning. The project will deploy ambient artificial intelligence to record and analyze both peer‑to‑peer and patient‑facing conversations, using...
Dan Marsh: AI Won’t Replace Advisers, but It Could ‘Supercharge’ Them
Dan Marsh, CEO of Octopus Money, argues that AI will not replace financial advisers but will "supercharge" them by automating back‑office tasks. He highlights that 30 million UK adults already use chatbots for money advice, yet the human element remains crucial...
Ep 777: No, Anthropic Isn’t Leading In Enterprise AI Adoption. Separating AI Facts From Fiction and How Companies Should Choose...
In this episode, Jordan Wilson debunks the viral claim that Anthropic has overtaken OpenAI in enterprise AI adoption, exposing flaws in the Ramp AI Index's methodology and data interpretation. He explains that Ramp’s metric—based on paid transactions—does not accurately measure...
Data Readiness for Agentic AI in Financial Services
Financial services firms see rapid adoption of agentic AI, but its success hinges on data quality, security, and accessibility rather than model sophistication. Elastic’s Steve Mayzak emphasizes that searchable, governed data stores are essential for autonomous AI to operate reliably...
Fiserv Picks OpenAI for Agent Tech and Other Digital Transactions News Briefs From 5/14/26
Fiserv announced a partnership with OpenAI to embed generative‑AI agents and AI‑driven cybersecurity into its banking solutions, marking the first major AI collaboration for a payments processor. The company also disclosed a pending joint venture with Bridgeport Partners to spin...
Mythos Proves Potent in Vulnerability Discovery, Less Convincing Elsewhere
Anthropic’s Mythos AI model demonstrates a clear advantage in uncovering software vulnerabilities when evaluated on live environments combined with source code, confirming the company’s core claim. XBOW’s independent testing shows the model’s strength wanes on static source‑code analysis alone and...
Enterprises Can Now Train Custom AI Models From Production Workflows — No ML Team Required
Empromptu AI unveiled Alchemy Models, a platform that captures validated outputs from enterprise AI applications and automatically fine‑tunes custom models without a dedicated ML team. The system uses Golden Data Pipelines to clean, enrich, and label production data, turning every...
Untitled
Cognex announced the general availability of OneVision™, a collaborative AI vision development platform that unifies cloud‑based model management with edge‑based inspection. Since its beta in June 2025, more than 100 manufacturers have moved from single‑line pilots to multi‑site rollouts, cutting deployment...
The Strange Japanese Companies Minting Money From AI
Ajinomoto, famed for MSG, now profits from Ajinomoto Build‑up Film, an insulation material for AI processors made from MSG by‑products. The company commands over 95% of this niche market. Surging AI chip demand has made the film scarce, lifting Ajinomoto’s...
The AI that Transformed American Warfare
The MAVEN Smart System is an AI‑driven weapon platform that not only pinpoints targets but also advises commanders on how to engage them. It has intercepted Iranian missile trajectories bound for Israel, located rocket launchers in Yemen, and tracked drug...
How Bechtel’s Going Big with AI and Automation
Bechtel has appointed a Senior Vice President of EPC Transformation to embed AI, robotics and digital tools across its global projects. The new leader, John Platt, says the firm will identify real‑world challenges, pilot solutions and scale successful technologies. Early...
Machine Learning System Design Interview #26 - The Inference Bottleneck Illusion
In a Meta senior ML engineer interview, candidates are asked to cut a recommendation system’s latency from 400 ms to a 100 ms SLA. Most immediately propose model‑level tricks such as INT8 quantization or pruning, assuming the ranking inference is the bottleneck....
End of the Check Engine Light Era
Suppliers and OEMs are replacing traditional check‑engine light alerts with AI‑driven diagnostics that automatically diagnose faults, book repairs and reroute trucks. Connected platforms from DTNA and Volvo deliver real‑time data, enabling predictive maintenance and over‑the‑air updates that keep trucks on...
How AI Is Helping Municipal Clerks Speed up Public Records Reporting
AI‑driven clerkware is reshaping how U.S. municipalities record public meetings. In Long View, North Carolina, the HeyGov ClerkMinutes platform slashed transcription time from four‑to‑eight hours down to 30‑60 minutes while eliminating errors, especially in budget figures. The tool timestamps speakers,...
AI Events Around the World in 2026: The Main Annual Meetings To Know
AI events in 2026 span research‑heavy conferences such as NeurIPS, ICML, CVPR and AAAI, as well as industry‑focused summits like the AI Summit London, World Summit AI, NVIDIA GTC and SuperAI Singapore. Attendance figures illustrate the scale—NeurIPS recorded over 13,000...
States Rush to Ban AI-Generated ‘Surveillance Pricing’ That Uses Shoppers’ Personal Data
Retailers are increasingly using AI-driven algorithms to set individualized, real‑time prices—a practice dubbed “surveillance pricing.” Consumer Reports uncovered price gaps of up to 23% for identical groceries on Instacart, potentially adding roughly $1,200 to a four‑person household’s annual bill. State...
PR Firm News: Supreme Group Unveils Supreme Intelligence FDE+
Supreme Group launched Supreme Intelligence FDE+, embedding AI‑native experts within client teams to build production‑grade healthcare solutions. The service taps a bench of more than 350 specialists and expands the firm’s proprietary AI operating system. Sprout Social announced the July rollout...