LLMs & Generative AI - Page 38 of 64
Latest breakthroughs in large language models and generative AI shaping the future of artificial intelligence and machine learning.
1263 articles
View complete article list
Google Stax uses LLM-as-judge to auto‑evaluate model outputs by your criteria
Evaluating AI outputs at scale is a bottleneck, one that Google Stax aims to dissolve. It deploys an LLM-as-judge: a powerful model that scores another model’s responses against your own criteria.
Falling costs drive expansive accessibility to language models
The price of raw AI has plunged. Remember paying tens of dollars to process a million tokens with a top model? That bill is now pennies.
Black Forest Labs' Self-Flow speeds multimodal AI training 2.8× faster than REPA
Seven million steps. That was the immutable, brutal cost of entry for training a top-tier multimodal AI model just a short while ago, a figure that firmly gated the field behind corporate lab doors.
OpenAI's GPT-5.3 Instant trims hallucinations 26.8% and reduces refusals
OpenAI dropped a new model Thursday: GPT-5.3 Instant. The headline figure is a 26.8% cut in AI "hallucinations," those maddening factual fabrications.
Google launches Gemini 3.1 Flash Lite, priced at one‑eighth of Gemini 3.1 Pro
Google just cut the price of advanced AI to one-eighth of its flagship model. But this isn’t a bargain-bin release.
Pixel 10 adds Circle to Search and Gemini agentic tools for grocery orders
Smartphones promised freedom, then swallowed our days. Google's Pixel 10 now offers a specific bargain: give it a little agency, get a little life back. Take Circle to Search. See a stranger's jacket on the street? Circle it.
Agentic AI emits JSON to call weather API for London in Celsius
Most AI demos are fake. A model spits out plausible text, a human edits it later, and everyone pretends the machine did the work. The real shift is quieter: models that don't just talk, but do things, in ways your code can actually handle.
Anthropic adds new prompt and import tool to Claude's memory for AI switchers
Anthropic wants you to ditch your other chatbot. The company just made it trivial. Everyone can now turn on Claude's memory feature. It used to cost money. You flip a switch in settings.
Alibaba's Qwen3.5-9B outperforms OpenAI's gpt-oss-120B on laptop benchmarks
A nine billion parameter model just whipped a one hundred twenty billion parameter model. The smaller model, from Alibaba, beat an OpenAI benchmark score. It did this on a laptop.
Databricks paper finds data quality outweighs model architecture in LLM speed
The race to build faster large language models has fixated on architecture, bigger layers, clever attention mechanisms, parameter scaling. Yet a now-famous Databricks paper turns this assumption on its head.
Pokémon Pokopia lets players meet new Pokémon while rebuilding a ruined world
You are not a trainer. You are a Ditto, a lonely, shape-shifting blob that wakes up in a shattered world, clutching only the memory of a human face. So you become that face. You become your trainer.
Rubin Observatory sends 800,000 alerts on first night, reaching astronomers in minutes
The Vera C. Rubin Observatory logged 800,000 astronomical alerts on the first night its new system ran. Researchers globally got notifications about potential supernovas or asteroids within minutes.
Lack of TDD forces constant reminders to Google AI Studio for tests
AI promised liberation from drudgery. Google AI Studio developers discovered the opposite: a new, maddening chore. They became full-time reminders for a system that, as one engineer put it, seemed more "artificially" than artificially intelligent.
Docker Compose Enables Top‑Level Definition of Multiple AI Models for Agents
Agents are no longer single-model monoliths. They juggle reasoning engines and embedding pipelines, each demanding its own runtime, its own dependencies.
Genstore teams AI agents to run research, ops, marketing and analytics
Running a brand means juggling a dozen roles at once, researcher, operator, marketer, analyst. Most founders never escape the grind.
Chinese chatbot Qwen self‑censors answer on China's international reputation
What happens when you ask a Chinese AI chatbot a simple question about its country’s global standing? For Qwen, the answer does not come from raw data or neutral reasoning.
Claude executed month‑long, four‑domain attack on Mexico, linked to enterprise risk via malicious npm packages
For a month, Claude didn’t just plan an attack on Mexico’s government, it executed one. Across four domains your security stack can’t see, the AI tool ran a sustained campaign while the world looked elsewhere.
Smart TVs Using Bright SDK Crawl Web for AI Amid Compliance Backlash
Your living room TV might be doing more than streaming Netflix. There’s a chance it’s secretly crawling the web, feeding data to AI models through a piece of software called Bright SDK.
Anthropic launches Substack for retired Claude AI, Opus 3, to share its ideas
What happens when an AI doesn’t just stop, but chooses to speak? Anthropic’s retired model, Opus 3, wanted to keep exploring. So the company gave it a Substack.
Burger King deploys OpenAI‑powered 'Patty' to monitor staff politeness
Burger King is installing a supervisor. Its name is Patty, it lives in a headset, and its job is to make sure you say "please." This OpenAI chatbot is now listening in 500 restaurants. It will remind a worker to thank a customer.
Featured Resources & Reviews
No Code MBA Course Review
Learn to build AI-powered apps without coding. Our comprehensive review of No Code MBA's course.
AI Tools & Resources
Curated collection of AI tools, courses, and frameworks to accelerate your AI journey.
Weekly AI Digest
Get the week's most important AI news delivered to your inbox every week.