Today's AI Pulse
-1.png)
Anthropic shuts down Fable 5 and Mythos 5 amid U.S. export‑control crackdown
Anthropic abruptly disabled its flagship Fable 5 and Mythos 5 models to comply with a U.S. export‑control directive, sparking a clash between the company, Washington regulators and cybersecurity experts. The shutdown triggered a rally in Chinese open‑source AI offerings such as MiniMax and Zhipu as investors shift to downloadable models that can run on‑premise.
AI and Brain-Computer Interface Allow Speechless ALS Patient to Work a Full-Time Job
University of California, Davis researchers implanted a brain‑computer interface (BCI) in ALS patient Casey Harrell in 2023, enabling him to control a cursor and synthesize speech using AI. The system, built on Blackrock Neurotech hardware and the team’s BRAND software, achieved 99% sentence‑level accuracy in lab tests and 92% in daily use. Harrell has logged more than 3,800 hours, averaging five hours per day, and has returned to full‑time work as an environmental advocate. The breakthrough demonstrates a practical, long‑term BCI solution outside the lab.
Third Wave Automation Partners with Raymond to Scale AI Across Lift Trucks
Third Wave Automation announced a partnership with Raymond, the Toyota Material Handling brand, to embed its AI‑driven shared‑autonomy platform into select Raymond lift trucks, beginning with the Swing‑Reach model. The solution lets a single remote operator supervise up to ten...
Trump’s Light AI Touch Keeps Getting Heavier
The Trump administration relaunched its "light‑touch" AI policy in early 2025, promising voluntary pre‑release notifications for frontier models. Within months, it reversed course by imposing export controls on Anthropic’s Fable and Mythos models, forcing the systems offline within 90 minutes....
Pentagon Boasts of Using AI to Write Reports Mandated by Congress
The Pentagon is leveraging its GenAI.mil platform, powered by tools like Google Gemini, to draft congressionally mandated reports, slashing production time from roughly 200 hours to five. CTO Emil Michael highlighted the rollout, noting that 1.5 million of the department’s 3.5 million...
Gusto Rolls Out Suite of AI Agents That Help Accountants Develop and Evolve Their Practice
Gusto introduced six AI agents designed to help accounting firms grow beyond traditional compliance work. The agents cover website optimization, client prospecting, sales outreach, advisory positioning, service expansion pitching, and pricing‑margin modeling. Available now through an Early Access program, firms...
PennyMac Jumps on Voice Agents with Amazon Partnership
PennyMac Financial Services is deepening its partnership with Amazon Web Services to launch borrower‑facing conversational AI agents and accelerate the modernization of its cloud‑based servicing platform, Plaisse. The new voice assistant, built on AWS’s Nova Sonic speech‑to‑speech technology, will operate...
AI Decodes Plant DNA 'Switches' To Better Predict Gene Control
An international team led by Forschungszentrum Jülich and the IPK Leibniz Institute has created a deep‑learning model that predicts where transcription‑factor proteins bind to plant DNA, effectively decoding regulatory “switches.” Trained on Arabidopsis thaliana data, the model simultaneously handles 46...
Nexla’s Express Solution Leverages Conversational Interface to Fuel Agentic AI
Nexla introduced Express, a conversational interface that lets enterprise teams build and deploy production‑grade data pipelines without writing code. The solution, now listed on AWS Marketplace, taps Nexla’s library of more than 550 bidirectional connectors and processes over 10 trillion records...
JMIR Articles Address AI Clinical Decision-Making and Health Care Worker Burnout
JMIR Publications featured two articles examining AI’s role in clinical decision‑making and the rise of digital fatigue among health‑care workers. A study found OpenAI’s o1 large language model matched or outperformed physicians in triage, first contact, and admission stages, though...
AWS Summit NYC: New AI Playbook, Agents And ROI Funds For Partners
At the AWS Summit in New York, Amazon Web Services introduced the Business Value Realization (BVR) program, a go‑to‑market motion that couples outcome‑based funding, playbooks, and benchmarks with AI deployments. The initiative also adds AI agents to the ACE co‑sell platform,...
Databricks Unifies Data Formats, Adds Real‑Time Ingestion
Awesome keynote by Ali Ghodsi at the Databricks summit. Some takeaways below. TLDR - Databricks is now a data processing platform, a data platform, an agents platform and an apps platform Architecture updates Iceberg / table formats - v3 is now GA: the...
AI Repurposes Failed Drugs Into New Medicines
How I use #AI to turn failed drugs into new medicines by Emma Ulker @Nature Learn more: https://t.co/YmS1Va9D3U #MedTech #HealthTech #Tech #TechForGood https://t.co/6fPuhjJlwC
DNAcrypt-AI: A Genome-Scale Bioinformatics Framework for Coordinate-Based Cryptography Using Artificial Intelligence
DNAcrypt-AI introduces a genome‑scale cryptographic framework that distributes encoded data across randomly selected human genome coordinates and reconstructs them via a high‑throughput pipeline. The system leverages the Covary machine‑learning engine to translate coordinate patterns into alphanumeric characters without relying on...
TwelveLabs’ Video AI Finds New Use Cases on AWS Marketplace
TwelveLabs, a 2020‑founded multimodal AI startup, offers the Marengo video‑search model, the Pegasus video‑language model, and the Rodeo AI copilot for editing. By joining AWS Marketplace, the company has widened its reach beyond telco, media, entertainment, games and sports (TMEGS)...
AI Is Reshaping AML in Australia — but at What Risk?
Australia’s AML landscape faces a paradox: AI promises stronger detection but also equips criminals with sophisticated tools. Napier AI’s AML Index estimates money‑laundering losses at roughly AU$87.4 billion ($58 billion) in 2024‑25, while AI‑driven recoveries could reclaim only AU$2.65 billion ($1.75 billion). Compliance spending...
HPE Unveils a Raft of New Networking Products for AI Workloads at Discover 2026
HPE announced a suite of networking products at Discover 2026 aimed at accelerating agentic AI deployments. The lineup includes the QFX5140 edge‑focused switch and the QFX5252 module built for AMD’s Helios AI rack‑scale platform, both marketed under the HPE AI Factory brand. HPE...
The More AI This Marketing Chief Exec Uses, the Less Scared He Gets
Nothing’s chief brand officer Charlie Smith says AI is a creativity tool, not a job threat. He’s built several low‑code “vibe‑coded” apps that automate his email, travel and daily dashboard, and uses the company’s Essential Voice dictation to eliminate typing....
Aviva Expands ChatGPT Distribution Channel
Aviva has extended its ChatGPT app to include life‑insurance quotes, allowing customers to receive an initial estimate via conversational AI before completing the purchase online. A GlobalData survey shows 41.9% of UK consumers are comfortable getting insurance quotes from chatbots,...
Checkmarx Unveils Next Generation SAST Engine with Hybrid AI Architecture
Checkmarx introduced Checkmarx SAST, the first static analysis engine built around a security‑tuned large language model, and a companion Finding Analysis Engine that filters false positives. The hybrid AI architecture can analyze any programming language, including AI‑generated and polyglot code,...
Databricks Releases Agentic Coworker Genie One
Databricks unveiled Genie One, an agentic coworker that connects to both structured and unstructured enterprise data and works across Databricks and external apps such as Google Drive, Jira, Slack, Confluence, and SharePoint. The Genie suite now includes Genie Agents, which...
Aptiv Introduces Camera-Only Occupant Detection System
Aptiv unveiled its Advanced Occupancy Classification (AOC) system, a camera‑only occupant detection solution that replaces traditional seat‑integrated sensors. Leveraging AI and computer‑vision, the software classifies adults, children, infants and objects with 100% accuracy in FMVSS 208 regulatory tests. By eliminating hardware,...
.png)
The Scam Texts Flooding Your Phone Were Built With Google's Own AI
Google has filed a lawsuit against Outsider Enterprise, a China‑based cybercrime network that leveraged Google’s Gemini AI to generate and dispatch 2.5 million phishing texts in just two weeks, stealing millions of credit‑card numbers and causing roughly $1.9 billion in losses. The...
Bing Webmaster Tools Updates AI Reporting with Intents, Topics, Citation Share and Compare
Microsoft has begun rolling out a preview of an upgraded AI performance report inside Bing Webmaster Tools, adding four new dimensions: Intents, Topics, Citation Share, and Compare. Intents categorizes the underlying queries that trigger AI citations into purpose‑based groups such...
Salesforce and Databricks Expand Partnership to Govern AI Agent Workflows
Salesforce and Databricks announced an expanded partnership aimed at unifying governed data, business context, and AI‑agent workflows across their platforms. The deal adds federated authentication, identity mapping, governance interoperability and metadata‑aware access controls, extending the existing Zero Copy integration between Salesforce...
How Spacelift Blends AI Experimentation with GitOps Governance
Spacelift’s infrastructure‑as‑code platform now blends AI experimentation with traditional GitOps governance, introducing a Model Context Protocol (MCP) server that lets AI models interact directly with cloud providers. The solution lets teams iterate rapidly in AI‑assisted pipelines and then promote stable...
When AI Chatbots Become The Trusted News Source
The 2026 Digital News Report shows weekly use of AI chatbots for news climbing to 10% globally, up from 7% the previous year. While only 1% of respondents name AI as their main news source, trust in AI‑delivered news sits...
Op-Ed: AI and the Future of International Marine Transportation
George Livingstone’s op‑ed argues that advances in artificial intelligence, quantum computing and generative design could soon enable nations to produce most consumer goods domestically. If manufacturing becomes localized, the traditional east‑west container trade that underpins global shipping would shrink dramatically,...
EMEA Firms Lack AI Visibility Despite Sovereignty Push, IBM Reports
IBM’s new EMEA AI sovereignty study reveals that 90% of executives lack full visibility into AI vendor, model and infrastructure dependencies. Nearly 75% say switching their primary AI provider would be difficult, and 81% consider a seven‑day outage critical. The...
Adoozle Launches AI Visibility Platform As Businesses Race to Adapt to AI Search
Adoozle, a ZipVentures brand, unveiled an AI Visibility Platform to help businesses be discovered by emerging AI search engines such as ChatGPT, Claude, Perplexity, Grok and Google AI. The service combines an AI Visibility Audit with structured, answer‑optimized business profiles,...
AI Adoption Accelerates in Hospitality: Hoteliers Report Operational Gains
Hospital operators are accelerating AI adoption to slash costs and boost efficiency, with multi‑year plans targeting 30%‑50% reductions. Leaders highlighted AI’s role in linking property‑management, restaurant, and housekeeping systems, enabling real‑time coordination. Hilton’s AI planner and chatbot now handle over...
Glean’s AI Platform Leverages Enterprise Data to Power Models and Agents
Glean Technologies has expanded from enterprise search into a full‑stack AI platform that powers autonomous agents and the Glean Assistant. Integrated with Amazon Bedrock and sold through AWS Marketplace, the platform lets enterprises plug in any model and keep data...
BlueZoo Announces In-Store Video Advertising Agent Built on Google’s AI Technology
BlueZoo announced an open‑source AI agent that optimizes in‑store video advertising using Google’s Gemini Enterprise technology. The solution, demonstrated at InfoComm 2026, combines real‑time, privacy‑first audience measurement with low‑cost AI‑generated video creatives. Retail media network operators can now calculate revenue‑per‑impression...
VidAU Launches AI Marketing Agents Built on a Proprietary Marketing Execution Harness
VidAU announced the launch of AI Marketing Agents, a platform that consolidates content creation, advertising, SEO, GEO, social media, and analytics under a proprietary Marketing Execution Harness. The system translates high‑level growth goals into coordinated, self‑learning workflows that plan, execute,...
Mitratech Showcases an Expanded Immigration AI Ecosystem, New Integrations Aimed at Law Firms
Mitratech Legal Solutions unveiled three new AI‑powered integrations—VisalawAI, Lawfully, and Candle AI—within its INSZoom immigration case management platform at the AILA 2026 conference. The company also previewed an AI‑enhanced client portal that automatically extracts data from uploaded documents. These additions...
Air France KLM Martinair Cargo Accelerates Commercial Transformation Through AI
Air France‑KLM Martinair Cargo has launched two AI‑driven tools within its myCargo platform: Auto‑Fill, which extracts shipment details from unstructured emails and populates booking fields, and Auto Quote, which instantly generates spot quotations. Both features aim to cut manual effort,...
Konecta Launches Kolibri , an Agentic AI Orchestration Platform
Konecta unveiled Kolibri, an agentic AI orchestration platform that bundles up to 80% pre‑built, industry‑specific use cases and the remaining 20% customized to client environments. The platform automates end‑to‑end workflows—billing, support, claims, and more—while logging every decision for real‑time auditability....
My Father Wants to Age in Place. AI Will Be Watching
A growing number of seniors, like the author’s father, are choosing to age in place while relying on AI‑driven monitoring systems to ensure safety and health. Wearable sensors, ambient cameras, and voice assistants can track vital signs, detect falls, and...
7 Business Use Cases for Agentic AI
At Data Summit 2026, Michael Vasicek of Airo Digital Labs outlined seven practical agentic AI applications, ranging from procurement advising to predictive maintenance. He highlighted two agents—contract negotiator and call‑center companion—that cut manual effort and boost customer satisfaction. The agents operate...
Race to Turn AI Compute Into a Commodity Spurs New Crypto Boom
Luxor Technology’s chief operating officer, Ethan Vera, is pivoting from Bitcoin‑miner hardware sales to building an AI‑compute infrastructure business. The venture will finance GPU acquisitions, broker hardware deals, and eventually create a marketplace for trading raw compute power. Soaring demand...
ABB Robotics Enters Partnership to Advance Physical AI
ABB Robotics has teamed up with California‑based bionics firm Psyonic to embed the Ability Hand prosthetic into its GoFa collaborative robot platform. By harvesting high‑fidelity touch, vibration and grip‑force data from human operators, the partnership seeks to teach robots human‑level...
ChurnZero Unveils Agentic Essentials, AI Platform to Boost Customer Success
ChurnZero introduced Agentic Essentials, an AI‑powered customer‑success suite that bundles more than 15 purpose‑built agents into a single annual subscription. The platform embeds deep business context into Claude and ChatGPT via a Model Context Protocol, promising tighter execution, richer intelligence...
Salesforce to Acquire Fin for $3.6 B, Bolstering Agentforce AI Agent Suite
Salesforce announced a $3.6 billion acquisition of Fin, the AI‑driven customer‑service startup formerly known as Intercom. The deal adds Fin’s Apex model and 30,000‑plus enterprise customers to Salesforce’s Agentforce platform, accelerating the company’s push into autonomous AI agents.
Databricks Pitches LTAP as a New Foundation for Agentic Applications
Databricks unveiled Lake Transactional and Analytical Processing (LTAP) at its Data + AI Summit, a lakehouse‑based architecture that merges transactional and analytical workloads on a single storage layer. By eliminating separate OLTP and OLAP systems and the associated ETL pipelines,...
Microsoft Unveils Surface Pro, Laptop With AI-Ready Snapdragon X2
Microsoft unveiled AI‑ready versions of its 13‑inch Surface Pro and new 13.8‑inch and 15‑inch Surface Laptops powered by Qualcomm Snapdragon X2. Consumer models are on sale now, starting at $1,499 for the Pro and $1,599 for the Laptops, while enterprise variants ship July 14....
Autodesk’s Neural CAD Brings AI Reasoning to Design and Engineering
Autodesk unveiled Neural CAD, an AI foundation model that reasons directly on 2D and 3D geometry rather than merely describing it. The technology builds on more than 15 years of research, datasets, and proprietary tools to generate editable CAD objects...
Teleport Adds LLM Proxy and Delegated Identity to Secure AI Agent Actions and Access
Teleport has launched two core components of its Agentic Identity Framework—LLM Proxy and Delegated Identity—in the public beta of its Beams platform. The LLM Proxy sits between AI agents and inference endpoints, inspecting every request, enforcing per‑Beam allow lists, and...
Radware AI Xploit Shield Delivers Virtual Patching for Newly Identified Application and API Flaws
Radware unveiled AI Xploit Shield, a service that automatically creates virtual patches for newly discovered application and API vulnerabilities. Leveraging AI models, the platform generates, tests, and deploys custom shields in near‑real time, protecting workloads without altering underlying code. The solution bridges...
Utterly Flocked: "We-Don't-Track-People"-Firm Deploys Nationwide Network Of Warrantless Pedestrian-Tracking Cameras
Atlanta‑based Flock Safety has built a nationwide network of tens of thousands of license‑plate‑reading and AI‑enabled cameras, despite its public claim that it does not track people. The system lets police retrieve up to 30 days of vehicle movement and...
CrowdStrike SecOps Deal With Grant Thornton Shows ‘Power Of The Platform’ For MSSPs: Execs
CrowdStrike announced that Grant Thornton Advisors is standardizing its security operations and managed detection and response services on the AI‑powered Falcon platform, replacing several legacy SOC tools. The partnership lets Grant Thornton scale its MSSP business across the U.S., U.K.,...
Mythos Proportions
Anthropic’s upcoming Claude model, codenamed Mythos, was halted after the company warned it could autonomously map corporate networks and launch attacks. The announcement sparked a sharp sell‑off in cybersecurity equities as insurers feared a new class of AI‑driven threats. Anthropic...