Tech Trends 📰

‘Microsoft is the AI ringleader’: tech rivals flock to software giant’s stage

Microsoft is positioning itself as the leader in the AI industry by unveiling new tools and partnerships at its Build conference, including with OpenAI, Nvidia, and Elon Musk’s xAI.. Big tech figures like Sam Altman, Elon Musk, and Jensen Huang participated in the event, signaling Microsoft’s central role in AI development and deployment.. Microsoft showcased tools like its coding “agent” and emphasized integrating multiple AI models (including OpenAI, xAI, Anthropic, etc.) to give developers more choice.. Microsoft’s cloud platform Azure and productivity tools give it a strong competitive advantage over rivals like Amazon and Google in the enterprise AI market.. Despite Microsoft’s investment in OpenAI, it is pursuing its own commercial AI strategy, focusing on scalable, enterprise-ready AI applications rather than general AI ambitions..

Vercel debuts an AI model optimized for web development | TechCrunch

Vercel has launched a new AI model called “v0-1.0-md” focused on front-end and full-stack web development.. The model is accessible via API and supports text and image prompts; it’s currently in beta and requires a paid V0 plan.. It can handle large inputs (up to 128,000 tokens) and auto-fix common coding issues.. The model is compatible with tools using OpenAI’s API format and is evaluated using frameworks like Next.js.. Vercel’s launch follows a trend of companies releasing coding-focused AI models, although challenges like bugs and security issues still exist in AI-generated code..

Anthropic's new Claude 4 AI models can reason over many steps | TechCrunch

Anthropic launched Claude Opus 4 and Claude Sonnet 4, advanced AI models that handle complex reasoning, long tasks, and programming effectively.. Sonnet 4 is available to free and paid users; Opus 4 is exclusive to paid users and more powerful, priced at $15/$75 per million tokens via API.. Opus 4 beats leading models like Google Gemini 2.5 and OpenAI GPT-4.1 on coding benchmarks but doesn’t outperform all models in every area.. Both models can reason in steps, use tools like search engines, store useful info in memory, and show simplified reasoning summaries.. Anthropic is updating Claude Code with IDE integration and a new SDK, aiming to improve AI-assisted coding and developer workflows..

Google updates the Gemini app with real-time AI video, Deep Research, and more | TechCrunch

Gemini Live now supports real-time video and screen sharing for all iOS and Android users.. Gemini will integrate more deeply with apps like Google Maps, Calendar, and Tasks.. Two new AI plans launched—AI Pro ($20/month) and AI Ultra ($250/month) with advanced features.. Deep Research can now analyze user-uploaded PDFs and images to generate personalized reports..

GitHub’s new AI coding agent can fix bugs for you

GitHub launched an AI coding agent within Copilot that can fix bugs, add features, and improve documentation when assigned a task.. The agent sets up a virtual machine, clones the repo, analyzes the code, and logs its reasoning.. After completing its work, it tags the developer for review and can address feedback automatically.. It's available to Copilot Enterprise and Pro Plus users on GitHub’s site, mobile app, and CLI..

AI dev tools for Windows get a fresh coat of paint | TechCrunch

Microsoft launched Windows AI Foundry, a rebrand and expansion of Windows Copilot Runtime, to help developers build AI-powered apps more easily on Windows.. Windows AI Foundry offers a unified platform to fine-tune, optimize, and deploy AI models locally on Windows devices.. A key feature, Foundry Local, allows developers to run AI models directly on devices (Windows and macOS), using ONNX Runtime.. The platform auto-detects a device’s CPU, GPU, and NPU and fetches compatible AI tools and models, keeping them updated.. Developers can use a command-line interface (CLI) with commands like foundry model list/run and integrate Foundry Local via its SDK..

Google Launches New AI-Powered Scam Detection Features

Google is using its AI model, Gemini Nano, to detect and block scams on Chrome, Google Search, and Android.. The AI already helps protect users from tech support scams and will soon expand to cover more threats on Android. A new feature in Chrome for Android detects spam disguised as website notifications and lets users unsubscribe or view blocked content.. Scam detection is coming to Google Messages and Phone by Google, alerting users if a call or text seems suspicious.. Users can override AI warnings if they believe the alerts are errors, maintaining control over blocked content and communication..

WhatsApp introduces AI tool for profile and group pictures | The Express Tribune

WhatsApp has launched a feature for AI-generated profile and group pictures. Users can create images just by typing a text prompt. No need for real photos which is good for privacy. This feature is currently available to limited beta testers. Users get the option to generate an AI image while changing profile or group icon in settings.

Intel has limited customer commitments for latest chip manufacturing tech, CFO says

Intel's upcoming chip manufacturing tech has received only a small volume of confirmed customer commitments, according to CFO David Zinsner.. Big companies like Nvidia and Broadcom are testing Intel’s chip production, but many customers drop out after initial test phases.. Intel’s foundry business, which manufactures chips for external clients, aims to break even by 2027, needing billions in external revenue to do so.. Despite $4.7 billion in foundry revenue last quarter, most of it still comes from chips made for Intel’s own use.. New CEO Lip-Bu Tan is focusing on streamlining Intel’s operations, selling off non-core assets, but is not planning major strategic shifts..

Vibe-coding startup Windsurf launches in-house AI models | TechCrunch

Windsurf has launched its own AI models named SWE-1, SWE-1-lite, and SWE-1-mini, built for the full software engineering process, not just coding.. SWE-1 will be available only to paid users, while SWE-1-lite and SWE-1-mini will be accessible to both free and paid users.. SWE-1 performs comparably to GPT-4.1 and Claude 3.5 Sonnet but falls behind newer models like Claude 3.7.. Windsurf is shifting from relying on third-party AI models to building its own, expanding beyond just application development..

Google's Gemini chatbot can now more easily analyze GitHub projects | TechCrunch

Google Gemini can now connect to GitHub, but only for users on the $20/month Gemini Advanced plan.. Users can import code by pasting a GitHub URL, allowing Gemini to analyze, debug, and explain the code.. This update comes right after OpenAI launched a similar GitHub connector for ChatGPT deep research.. AI companies like Google and OpenAI are racing to add new features to stay competitive in the growing AI market..

Using AI tools like ChatGPT at work may harm your reputation, study finds | The Express Tribune

A Duke University study found that using AI tools like ChatGPT at work may make colleagues see you as lazy or less competent.. In experiments with over 4,400 people, AI users were viewed as less hireable and more replaceable.. Managers unfamiliar with AI were less likely to hire candidates who regularly used it.. The negative perception was reduced when AI use was clearly tied to the job’s needs.. Many employees hide their AI usage at work, earning them the nickname “secret cyborgs.”.