Generative AI news and analysis on Web Pulse: 13610 curated articles tracking the latest Generative AI developments, tools, and research, updated continuously from vetted sources.
ByteDance released Lance, an open-source multimodal AI model with 3 billion active parameters that natively handles image generation, video generation, editing, and visual reasoning within a single framework. The model a…
Large language models are now driving significant user growth for online services, with Buttondown attributing its recent spike in subscriptions to LLM referrals and Fat Gold reporting its first-ever subscription referra…
AdExchanger’s editorial team debated the shifting definition of “premium” content at Programmatic AI in Las Vegas, arguing that AI-generated slop and user engagement are blurring the line between high-quality and low-qua…
Amazon Web Services released a reference architecture for building an AI-powered recruitment assistant using Amazon Bedrock, designed to automate candidate evaluation, generate personalized interview questions, and provi…
NVIDIA founder and CEO Jensen Huang arrived in Taipei Saturday to kick off GTC Taipei at COMPUTEX, where he announced that the Vera Rubin supercomputer represents the largest product launch in Taiwan's history, involving…
Telecommunications companies globally are building sovereign AI factories using NVIDIA's Cloud Partner reference architecture to provide in-country AI infrastructure with data controls and performance guarantees. These t…
Arize AI published a guide on building LLM-as-a-Judge evaluators that function reliably in production, emphasizing that code-based checks should handle deterministic tasks like schema validation while LLM judges should o…
A 4-billion-parameter AI model released in early 2025 is now outperforming models seven times its size on standard reasoning benchmarks, with Google's Gemma 3 4B scoring 89.2% on GSM8K math reasoning and Microsoft's Phi-…
A controlled experiment using three Google accounts found that brands seeded into Gmail and Photos appeared 46 percentage points more often in AI Mode recommendations when Personal Intelligence was connected. Across 1,92…
Engineering leaders are designing AI-driven workflows without firsthand experience of how those tools actually function in a real codebase, creating a dangerous gap between strategy and execution. A CTO's plan to have ag…
OpenAI's GPT-next model disproved the 80-year-old Erdős planar unit distance problem in under 32 hours at a cost of less than $1,000, marking the first instance of a general-purpose AI solving a well-known open mathemati…
Google announced Tuesday at Google IO that it is unifying its AI coding tools under the Antigravity platform, consolidating products like Gemini CLI and Gemini Code Assist into a single agent-first development environmen…
Google announced a new family of Gemini 3.5 AI models, new Search and Gmail features, and updates to its Project Aura smart glasses during its I/O 2026 keynote. The announcements were the focus of the event, which was do…
Google has begun rolling out Gemini Spark, a personal AI agent designed to navigate users' digital lives and act on their behalf across Google's products and cloud infrastructure. The company tested the agent with a limi…
Modular has built a new data layer for LLM inference routing that solves the problem of querying cached blocks across hundreds of pods in microseconds. The company's architecture uses a specialized data structure with fa…
A developer describes two modes of working with AI coding assistants: 'synced mode' for building shared understanding and 'delegate mode' for offloading tasks once boundaries are solid. The author argues that effective d…
XAI released Grok Imagine Video 1.5, a video generation model capable of producing realistic footage with synchronized audio in a single pass. The company published a prompting guide based on extensive testing, emphasizi…
DeepSeek TUI, a terminal user interface for DeepSeek's models, has gained nearly 33,000 GitHub stars in two weeks. The tool can run against other providers including local models via Ollama or oMLX, and developers can in…
Amazon SageMaker AI now supports OpenAI-compatible API for real-time inference endpoints, allowing users to invoke models by changing only the endpoint URL without custom clients or code rewrites. The update enables bear…
NVIDIA has outlined nine techniques for customizing autonomous AI agents, ranging from simple prompt engineering to advanced reinforcement learning, to help businesses adapt general-purpose models for specialized tasks l…