TMT Expands Digital Trust and Safety with Microsoft Publishers
The Media Trust (TMT) announced a partnership to bolster digital trust and safety within Microsoft’s advertising ecosystem. Leveraging TMT’s proprietary AI detection, global infrastructure, and malware‑analysis teams, the collaboration will deliver real‑time threat detection and mitigation for malware, redirects, and cloaked ads across Microsoft‑owned and operated publishers. The solution aims to protect users while preserving legitimate advertiser revenue, enhancing the overall quality of Microsoft’s ad environment. This move underscores TMT’s two‑decade leadership in securing digital ad inventory for global brands.
Oligo Appoints Shira Bendkowski as VP of Product
Oligo Security announced Shira Bendkowski as its new Vice President of Product. Bendkowski, formerly VP of Product at Aqua Security and head of product at XM Cyber, will steer Oligo’s product vision for runtime security across applications, cloud, workloads, and...
Show HN: AskUCP – UCP Protocol Explorer Showing All Products on Shopify
AskUCP, launched on January 11 2026, is a global AI shopping search built on the Universal Commerce Protocol (UCP). UCP, an open standard co‑created by Google and Shopify, lets applications query product data across e‑commerce platforms without traditional APIs or integrations. At...
CoreSite Chicago Adds AWS Native 400G Direct Connect for Cloud and AI
CoreSite, an American Tower subsidiary, has launched native 400 Gbps Amazon Web Services Direct Connect at its Chicago data‑center campus, delivering ultra‑high bandwidth and low‑latency connectivity for AI, enterprise and GPU‑optimized workloads. Early adopters, including a cybersecurity firm, are using the...
OpenCog Hyperon and AGI: Beyond Large Language Models
OpenCog Hyperon, an open‑source neural‑symbolic framework from SingularityNET, positions itself as a bridge between narrow large language models and artificial general intelligence. Its core combines probabilistic logic, symbolic reasoning, evolutionary program synthesis, and a metagraph called Atomspace to enable dynamic,...
BOSGAME Launches Next-Generation AI PCs Built on AMD and Intel Platforms
BOSGAME unveiled two AI‑focused mini PCs, the VTA‑439 and VTI‑490, slated for global release in Q1 2026. The VTA‑439 runs on AMD’s Ryzen AI 9 HX 470 with Radeon 890M graphics, while the VTI‑490 leverages Intel’s Core Ultra X9 388H and an Arc B390 GPU....
SEON Identity Verification Combines KYC Checks with Real-Time Fraud Intelligence
SEON introduced an AI‑powered Identity Verification solution that combines document validation, biometric liveness detection, proof‑of‑address checks, and optional government database queries within its unified risk platform. The service draws on more than 900 real‑time fraud signals to evaluate both the...
SIOS Technology VP of CX Cassius Rhue Shares 2026 IT Predictions
SIOS Technology’s Vice President of Customer Experience, Cassius Rhue, outlined a forward‑looking vision for high‑availability (HA) and disaster‑recovery (DR) solutions through 2026. He predicts HA will evolve from pure uptime guarantees to a strategic pillar for hybrid‑cloud resilience, cybersecurity, AI...
How Target Is Pioneering Agentic Commerce in Retail
Network International has signed a processing agreement with Saudi Sudanese Bank to expand digital payment services across Sudan. The partnership will provide acquiring, issuing, and fintech solutions, enabling merchants to accept card payments and facilitating cross‑border transactions. By leveraging Network...
The Right to Be Forgotten: The Emerging Science of Machine Unlearning
Machine unlearning is emerging as a scientific discipline that lets organizations delete specific data points from trained AI models without full retraining. The approach addresses GDPR’s Right to Erasure and growing copyright demands, which treat model outputs as personal or...
ParaScript® Named Product Innovation Winner in the 2026 BIG Innovation Awards
ParaScript won the 2026 BIG Innovation Award for Product Innovation with its patented embedded signature verification solution that enables real‑time check fraud detection. The technology embeds an encrypted, invisible signature at issuance, allowing instant verification across ATMs, teller lines, remote...
Review: AI Strategy and Security
AI Strategy and Security, authored by Dr. Donnie W. Wendt, is a practical guide for technology leaders and security professionals designing enterprise AI programs. The book maps AI adoption to business objectives, outlines readiness assessments, and defines a comprehensive team...
The Race to Build the DeepSeek of Europe Is On
European governments are pouring hundreds of millions of euros into AI programs to reduce reliance on U.S. tech giants, while labs across the continent chase a "DeepSeek of Europe" breakthrough. The push emphasizes open‑source model development, hoping collaborative research can...
7 Top Cybersecurity Projects for 2026
The 2026 cybersecurity roadmap highlights seven priority projects for CISOs, ranging from AI‑aware identity and access management to advanced email protection, autonomous code‑vulnerability discovery, and enterprise‑wide zero‑trust adoption. Leaders emphasize extending IAM controls to non‑human agents, leveraging small language models...
Wildix Positions AI as the Catalyst Moving the UC&C Channel Beyond Resale at 2026 Summit
At its 2026 virtual summit, Wildix emphasized AI as the catalyst moving its channel‑only model from simple resale toward advisory services that design, govern and optimize unified communications workflows. The company showcased the Wilma AI layer in live customer deployments...
Alorica Named Innovative Organization Winner in the 2026 BIG Innovation Awards
Alorica Inc. has been named an Innovative Organization winner in the 2026 BIG Innovation Awards, recognizing its ability to scale artificial intelligence across enterprise customer‑experience operations. The award highlights Alorica IQ, the company’s AI innovation engine, and evoAI, its conversational‑AI...
CompuChild Expands Its Suite of Artificial Intelligence and Machine Learning Classes for Children
CompuChild, a STEM‑certified after‑school franchise, announced the expansion of its artificial intelligence and machine learning curriculum for elementary and middle‑school students. The new suite includes four courses—AI and ML for Young Innovators, AI Game Creator, Video Game Design with ML,...
Seela Launches an All-in-One AI Creative Platform for Image Generation and Multi-Model AI Comparison
Seela has launched an all‑in‑one AI creative platform that combines text‑to‑image, image‑to‑image, and a multi‑model chatbot comparison workspace. The platform lets users submit a single prompt and receive side‑by‑side answers from LLMs such as ChatGPT, Claude, Grok and DeepSeek, while...
Genesco Announces Strategic Transformation of IT Operations
Genesco, the owner of Schuh, announced a strategic overhaul of its IT operating model to boost speed, scalability and AI‑enabled innovation. The transformation is designed to align technology capabilities with the retailer’s new business model and support long‑term growth. Genesco...
GA-ASI Makes Another Autonomous Aerial Intercept in Company-Funded Demo with MQ-20 Avenger(r)
General Atomics Aeronautical Systems (GA‑ASI) demonstrated its MQ‑20 Avenger executing a fully autonomous aerial intercept using government‑reference autonomy software. The jet employed an Anduril infrared search‑and‑track sensor to locate, track, and compute a simulated kill‑shot against a manned aggressor without...
Nous Research Releases NousCoder-14B: A Competitive Olympiad Programming Model Post-Trained on Qwen3-14B via Reinforcement Learning
Nous Research unveiled NousCoder-14B, a competitive programming model built on Qwen3-14B and fine‑tuned with execution‑based reinforcement learning. On the LiveCodeBench v6 benchmark, the model achieved a Pass@1 score of 67.87%, outpacing the Qwen3-14B baseline by 7.08 points. Training leveraged 24,000...
Hiring at India’s Big Four Outsourcers Stalls, as AI Seemingly Makes an Impact
India’s four largest IT outsourcers – HCL, Infosys, TCS and Wipro – have essentially halted hiring in the latest quarter, adding only 3,910 staff year‑to‑date, a stark contrast to their usual 10,000‑plus quarterly hires. Revenue growth remains modest, with HCL...
Avocados From Mexico Turns to AI to Advertise Around the Super Bowl Instead of a TV Buy
Avocados From Mexico is forgoing a traditional $8 million Super Bowl TV spot in favor of an AI‑driven Prediction Pit that delivers real‑time football forecasts and custom guacamole recipes. The experience features a realistic Rob Riggle avatar powered by a predictive AI...
How AI Agents Become Direct Customers For WooCommerce Stores
AI agents built on large language models are now completing full purchase flows on WooCommerce stores, bypassing traditional referral links. PayLayer’s free plugin introduces the x402 protocol, letting bots discover products, verify price and stock, pay, and receive a signed...
New OpenAI Leak Hints at Upcoming ChatGPT Features
OpenAI is quietly testing a major ChatGPT web update slated for rollout in the next few weeks. The preview, dubbed “Salute,” adds a task‑creation interface with file uploads and progress tracking. Additional changes include a model‑preference flag aimed at hospitality‑specific...
NRF 2026: AI Development Gets Practical
At NRF 2026, technology vendors emphasized moving AI from hype to practical applications in retail back‑office functions and store operations. They argue that AI’s potential—automated inventory management, demand forecasting, and labor scheduling—can only be realized if retailers first modernize their...
World Network Makes Italian Market Debut After Making Steady Progress in Past Year
World announced its Italian market debut on January 15, expanding its human verification network to a fifth EU country. The company now serves nearly 38 million users, with over 17 million verified identities across six continents. Recent upgrades include NVIDIA‑powered hardware, an...
Sequoia Seeks Part in Anthropic’s $25 Billion Funding Round
Sequoia Capital is poised to join a multibillion‑dollar financing round for Anthropic, the AI startup that recently unveiled its agentic "Cowork" feature for Claude. The round, led by Singapore’s sovereign fund GIC and Coatue, aims to raise at least $25 billion...
AI-Powered Screening for Low Bone Mass in X-Rays
A new artificial‑intelligence algorithm can identify low bone mass, a precursor to osteoporosis, directly from routine X‑ray images. The system was trained on over 200,000 annotated radiographs and achieved more than 90% sensitivity in clinical validation. By flagging at‑risk patients...
Stop Calling It 'The AI Bubble': It's Actually Multiple Bubbles, Each with a Different Expiration Date
The AI boom consists of three distinct layers—wrapper companies, foundation‑model providers, and infrastructure—each with its own risk profile and timeline. Wrapper startups that merely repackage APIs are expected to implode first, as large platforms absorb their functionality and margins evaporate....
Show HN: Lume 0.2 – Build and Run macOS VMs with Unattended Setup
Lume 0.2 is an open‑source VM runtime that runs macOS and Linux on Apple Silicon using Apple’s Virtualization Framework, delivering near‑native performance. It ships as a single binary offering both a CLI and an HTTP API for headless VM creation, automated...
NDSS 2025 – Compiled Models, Built-In Exploits
Researchers at NDSS 2025 unveiled a systematic study of bit‑flip attacks targeting deep‑learning executables compiled by modern DL compilers. Unlike prior work that focused on flipping weights within frameworks, the new approach exploits publicly known model structure embedded in the...
Vercel Releases Agent Skills: A Package Manager For AI Coding Agents With 10 Years of React and Next.js Optimisation Rules
Vercel has launched the open‑source agent‑skills package, a plug‑in style manager that turns curated React, Next.js, and web‑design best‑practice playbooks into reusable capabilities for AI coding agents. The initial release bundles three core skills—react‑best‑practices, web‑design‑guidelines, and vercel‑deploy‑claimable—each containing dozens of rule‑based...
Man Confused by AI-Generated Reports That He’s Dead
Denver Broncos beat writer Cody Roark discovered an AI‑generated Facebook post claiming he was dead and left a child. The post, from the now‑removed Wild Horse Warriors page, also fabricated false stories about Broncos players. Similar AI hallucinations have appeared...
4 in 5 Small Businesses Had Cyberscams Last Year, Almost Half Were AI Powered
A recent Identity Theft Resource Center survey shows that four out of five small businesses experienced a cyber‑scam or breach in the past year, with almost half of those attacks powered by artificial intelligence. The study found that 38% of...
South Koreans Now Spend More on AI Subscriptions than Netflix Each Month
South Korean consumers and businesses spent an estimated 80.3 billion won on AI subscriptions in December 2025, surpassing the average monthly Netflix revenue of 75 billion won in 2024. Credit‑card transactions for AI services surged from 52,000 in January 2024 to 1.666 million by year‑end,...
Snap's SnapGen++ Generates High-Resolution AI Images on iPhone in Under Two Seconds
Snap Inc. unveiled SnapGen++, a 0.4 billion‑parameter diffusion‑transformer that creates 1024 × 1024‑pixel images on an iPhone 16 Pro Max in just 1.8 seconds. The model replaces the earlier U‑Net‑based SnapGen and introduces a novel attention scheme that slashes per‑step latency from 2,000 ms to under 300 ms. Using...
‘Today’s AI Models Are Missing Critical Capabilities’: Google DeepMind’s Demis Hassabis
Google DeepMind CEO Demis Hassabis warned that today’s large language models still lack essential capabilities such as long‑term planning, continual learning, and a deep understanding of physics and causality. He argues that the next AI frontier lies in building “world...
Something Wild Happens to ChatGPT’s Responses When You’re Cruel To It
Researchers at the University of Pennsylvania tested 50 questions across five tonal variations for ChatGPT‑4o and found that increasingly rude prompts improved answer accuracy, reaching 84.8 % for very rude prompts versus 80.8 % for very polite and 75.8 % for ultra‑polite. The...
Mint Explainer: How AI Is a Threat to the Reputation of Top Influencers and What Can Be Done About It?
Top Indian influencers have recently taken legal action against AI‑generated deepfakes that misuse their likenesses, ranging from obscene videos to unauthorized commercial ads. While the Delhi High Court secured takedowns for some creators, only podcaster Raj Shamani has obtained a...
Google Upgrades AI Overviews with Gemini 3 Pro for Complex Queries
Google is integrating its newest Gemini 3 Pro model into AI Overviews, the concise answer boxes that appear beneath search results. The system now intelligently routes complex queries to Gemini 3 Pro while delegating simpler questions to faster, less expensive...
Even the Best AI Models Fail at Visual Tasks Toddlers Handle Easily
A new BabyVision benchmark reveals that leading multimodal AI models struggle with basic visual tasks that toddlers master before speaking. The top model, Gemini‑3‑Pro‑Preview, achieved only 49.7% accuracy, far below the 94.1% score of adult humans and the average three‑year‑old...
NVIDIA Releases PersonaPlex-7B-V1: A Real-Time Speech-to-Speech Model Designed for Natural and Full-Duplex Conversations
NVIDIA unveiled PersonaPlex-7B-v1, a 7‑billion‑parameter full‑duplex speech‑to‑speech model that merges automatic speech recognition, language understanding, and text‑to‑speech into a single transformer. The dual‑stream architecture processes user audio and agent output concurrently, enabling barge‑in, overlapping speech, and rapid turn‑taking. Hybrid voice...
Google Chrome Now Lets You Turn Off On-Device AI Model Powering Scam Detection
Google Chrome now lets users delete the on‑device AI model that powers the Enhanced Protection feature, which uses generative AI to detect scams, malicious downloads, and risky extensions. The toggle appears in Settings > System under “On‑device GenAI.” The capability is currently...
How to Build a Self-Evaluating Agentic AI System with LlamaIndex and OpenAI Using Retrieval, Tool Use, and Automated Quality Checks
The tutorial demonstrates how to construct a self‑evaluating, agentic AI system using LlamaIndex and OpenAI’s gpt‑4o‑mini model. It combines retrieval‑augmented generation, tool integration, and automated faithfulness and relevancy scoring to create a reliable RAG workflow. The ReActAgent orchestrates evidence retrieval,...
A Laser that Can Fire Light Pulses in One Billionth of a Second Is Set to Produce Structures 1000 Times...
Japanese researchers have demonstrated a femtosecond‑laser technique that patterns silicon and silica films with periodic nanoscale grooves, directly controlling phonon pathways. By matching groove spacing to phonon mean free paths, the method reduces thermal conductivity while delivering structures up to...
AI Tools Speed Development of Antibody Probes to See Activity Inside Living Cells
Researchers at Colorado State University used AI tools to convert conventional antibodies into intracellular “intrabodies” that light up specific histone modifications. By leveraging AlphaFold2 and ProteinMPNN, the team reduced design cycles to five candidates and achieved a 70% success rate,...
Elon Musk Proposed His Children Should Inherit Control of Human-Like AI, OpenAI Claims
Elon Musk reportedly told OpenAI officials that his children should inherit control of artificial general intelligence (AGI) systems. OpenAI alleges Musk’s push for a majority stake in the company was tied to this succession plan, echoing his earlier ambition to...
Tech Companies’ Access to UK Ministers Dwarfs that of Child Safety Groups
A Guardian investigation reveals that major US tech firms have met UK ministers far more frequently than child‑safety and copyright groups. Google recorded over 100 ministerial meetings, and the sector as a whole logged 639 meetings in the two years...