OpenAI Released GPT-5 — Reasons Like PhD Student
GPT-5 scores in the 99th percentile on graduate-level reasoning tests and introduces real-time collaboration features that blur the line between AI and colleague.
News - The Pulse Gazette Coverage
Breaking AI news and developments. Browse 76 articles.
Meta Released Llama 5: Beats GPT-5 on Every Benchmark
Meta's newest open-weight model sets new records across reasoning, coding, and multilingual tasks, intensifying the open vs. closed AI debate.
GitHub Copilot Now Writes Entire Apps From a Single Prompt
GitHub's latest Copilot update can generate full-stack applications from natural language descriptions, complete with tests and deployment configs.
OpenAI Just Made GPT-5 Free — Here's the Catch
The most powerful language model ever released is now available at no cost, but the business model behind it reveals where AI is really headed.
AI Helped Reunite 1,000 Refugee Families Separated by War
An AI-powered facial recognition and document matching system has reconnected over 1,000 families displaced by conflicts in Syria, Ukraine, and Sudan.
Google Open-Sources Gemma 3: What It Means for AI
The latest open-weight model from Google challenges Meta's Llama and raises the bar for what's freely available to developers.
Inside AI Scams: Voice Cloning and Deepfakes
Fraudsters are using AI to clone voices, create deepfakes, and automate scams at unprecedented scale. The attacks are getting smarter. Here's how to protect yourself and your family.
Apple On-Device AI Upgrade Works Offline
The latest Apple Intelligence update brings powerful language and image models that run entirely on your iPhone, no cloud required.
Google's Gemini Ultra 2.0 Now Powers Every Google Product
From Search to Gmail to Maps, Gemini Ultra 2.0 is now the default AI behind every Google service — affecting 2 billion daily users.
EU Fines Meta .3 Billion for Training AI on European Data
The record-breaking GDPR fine targets Meta's use of Facebook and Instagram posts, photos, and messages to train its Llama AI models without adequate user consent.
Blind Runner Completes Marathon Guided by AI
Using AI-powered spatial audio navigation and computer vision, a visually impaired runner finished a full marathon without a human guide for the first time.
Anthropic Launches Claude Enterprise With Unlimited Memory
Claude Enterprise offers persistent memory across conversations, 500K token context windows, and SOC 2 Type II compliance for business customers.
The Claude Crash: How One AI Release Triggered a Trillion-Dollar Software Selloff
Anthropic's quiet plugin launch sent shockwaves through enterprise software. But is the panic justified, or is Wall Street overreacting to a productivity tool?
Samsung's New AI Phone Chip Runs LLMs Entirely On-Device
The Exynos 2500's dedicated NPU handles 7B-parameter models without cloud connectivity, marking a pivotal shift in mobile AI architecture
Nvidia's New Chip Makes AI Inference 10x Cheaper
The B300 GPU slashes the cost of running AI models, potentially making real-time AI affordable for apps and devices that couldn't justify it before.
Claude Opus 4 Sets New Record on Agentic Coding: 72% on SWE-Bench Verified
Anthropic's latest model autonomously fixes real GitHub issues better than any AI before. Developers report it can now handle multi-file refactors that took hours.
OpenAI Launches GPT-5 Turbo API: 10x Faster, 50% Cheaper, Same Intelligence
The optimized model makes GPT-5-level reasoning affordable for startups. The API waitlist has 200,000 developers.
Claude's Computer Use Is Now Production-Ready: AI Can Navigate Any Desktop App
Anthropic removes beta label from computer use feature. Enterprises are automating entire workflows that required human mouse clicks.
Grok 3 Has Real-Time Web Access and No Content Filters
xAI's unfiltered model can search the live web and respond without guardrails. Early users are finding creative—and concerning—use cases.
Mistral Large 3 Is Europe's Answer to American AI Dominance. And It's Competitive.
The French lab's new model matches Claude on reasoning and costs 60% less. EU governments are paying attention.
OpenAI's o3-mini Makes Advanced Reasoning 20x Cheaper
The new reasoning model costs pennies per query and matches o1 on most tasks. The economics of AI apps just changed.
Claude Code Can Now Build Entire Applications From a Single Prompt
Developers describe complete codebases generated in minutes. The tool handles architecture, testing, and deployment configuration.
Gemini 2's Agent Mode Can Now Manage Your Entire Google Workspace Autonomously
It reads emails, schedules meetings, writes documents, and handles calendar conflicts. Users say it's like having a competent assistant who never sleeps.
Claude Now Has Persistent Memory Across Conversations. It Remembers Everything You've Told It.
The new memory feature builds a profile of each user over time. It's incredibly useful—and raises obvious privacy questions.
AI Drones Help Firefighters Rescue Trapped People
Thermal-imaging AI drones cut search times by 80% in disaster zones, and they just saved 12 lives in a Turkish earthquake.
OpenAI Launches GPT-5 Pro: Enterprise Features That Actually Matter
SOC 2 compliance, data residency options, and guaranteed uptime. The boring stuff that makes enterprise adoption possible.
This AI Reunited a Lost Dog With Its Family 400 Miles Away
When traditional methods failed, an AI-powered pet recognition app matched a stray dog in Tennessee to a family in Ohio who'd been searching for months.
GPT-5 Beats Human Experts on Every Major Benchmark
The new model scores higher than PhD-level humans on medical, legal, and scientific reasoning tests. Sam Altman warns the next version will be 'qualitatively different.'
30% of Spotify Uploads Are Now AI-Generated Music
The streaming platform is flooded with synthetic music. Human artists are losing royalties to machines.
Nvidia Posts First Trillion-Dollar Quarter as Demand Soars
H200 chips selling faster than they can be made. Every major AI lab is competing for allocation.