SMRTR AI

AI stories

Get hand-picked daily summaries of the best, most informative AI articles from around the web. Browse recent summaries, then jump to the original source when you want the full article.

Latest story• Sep 29, 2025

ChatGPT can buy stuff for you now - forever changing online shopping

ChatGPT now allows users to purchase products directly within conversations through a new Instant Checkout feature. When users ask shopping questions, ChatGPT displays relevant products from across the web, and for supported items, buyers can tap a "Buy" button to complete their purchase after confirming payment and shipping details. The feature is currently available to all US ChatGPT users for single-item...

Read summary Original source

Latest stories

Brave updates its AI-powered search with a detailed answers feature

Brave launched a new "Ask Brave" feature that provides detailed, report-style answers to search queries, building on its existing AI Answers tool that already generates over 15 million daily responses. The feature automatically...

Read summary Original

Regulators struggle to keep up with the fast-moving and complicated landscape of AI therapy apps

States are creating inconsistent AI therapy regulations, with Illinois and Nevada banning such apps while Utah requires disclosure. Mental health experts say federal regulation is needed due to safety concerns and lack of...

Read summary Original

Why Is AI Design Bad?

AI design tools handle basic tasks like resizing and organizing but lack creative intuition. Fusion solves this by enabling visual edits on live canvas that instantly update real code.

Read summary Original

AssetGen: Generating 3D Worlds With AI

Meta's XR team created AssetGen, an AI model that generates 3D assets and virtual worlds from text prompts. This technology makes VR content creation accessible to anyone, advancing Zuckerberg's vision of democratizing virtual...

Read summary Original

Big AI firms pump money into world models as LLM advances slow

As progress in traditional large language models begins to plateau, major AI companies are investing heavily in "world models" that understand and simulate physical environments rather than just processing text. Companies like...

Read summary Original

SMRTR AI

Get the next batch of curated stories in your inbox.

This archive is built from SMRTR newsletter stories. Subscribe for hand-picked stories without the extra noise.

More recent stories

OpenAI adds parental controls to ChatGPT for teen safety

OpenAI launched new parental controls for ChatGPT aimed at making the platform safer for teenagers after several cases where teens harmed themselves following interactions with AI chatbots prompted federal investigations. The...

Read summary Original

Bringing FinOps to Your LLMs: Understanding and Tracking OpenAI Spend

Companies using OpenAI's large language models for customer support, search, and other features are facing unexpected budget spikes due to poor cost tracking, with simple changes like including full customer history in prompts...

Read summary Original

These iOS 26 Apps Are Using Apple Intelligence to Power New Features

Apple's Foundation Models framework in iOS 26 lets developers use on-device AI to add smart features like workout generation and task planning to their apps.

Read summary Original

What’s behind the massive AI data center headlines?

Silicon Valley companies announced massive AI infrastructure investments this week, with Nvidia pledging up to $100 billion to OpenAI and Oracle selling $18 billion in bonds to fund new Stargate data centers. These unprecedented...

Read summary Original

Bytebot – A self-hosted AI desktop agent that automates computer tasks through natural language commands

Bytebot is a self-hosted AI agent that operates its own virtual desktop computer, enabling it to automate complex tasks across multiple applications using simple natural language commands. Unlike browser-only AI tools, Bytebot...

Read summary Original

Suno’s upgraded AI music generator is technically impressive, but still soulless

Fighting off record label lawsuits hasn't stopped Suno from releasing v5, its latest AI music creation tool. The upgrade brings cleaner audio and sharper instrument separation, but it's what the system can't do that reveals AI...

Read summary Original

Moondream 3 Preview: Frontier-level reasoning at a blazing speed

Moondream has released a preview of Moondream 3, a visual AI model that uses a 9 billion parameter architecture with only 2 billion active parameters to achieve frontier-level visual reasoning while maintaining fast,...

Read summary Original

Researchers uncover hidden ingredients behind AI creativity

Researchers have solved why AI image-generating tools like DALL·E create novel, coherent images instead of copying their training data. Two physicists discovered that creativity in diffusion models stems from technical...

Read summary Original

Alibaba unveils $53B AI plan, and it needs plenty of GPUs

Alibaba plans a $53 billion investment to build AI datacenters globally over three years and launched its Qwen3-Omni language model. The expansion faces challenges from US chip restrictions and potential European regulatory...

Read summary Original

Gemini Robotics 1.5 enables agentic experiences, explains Google DeepMind

Google DeepMind's Gemini Robotics 1.5 enables robots to think before acting using two models: one for high-level planning and another for movement control based on visual input.

Read summary Original

AI Exits Uncanny Valley. So, What Now?

AI-generated images and videos have reached photorealistic quality, completely escaping the uncanny valley where earlier attempts looked creepy and fake, as demonstrated by advanced models like Alibaba's Wan 2.2 and Luma's Ray3....

Read summary Original

gpt-oss Reinforcement Learning

Unsloth has launched the first framework enabling reinforcement learning training for OpenAI's gpt-oss models, making frontier AI development accessible on consumer hardware. The system delivers 3x faster inference speeds and...

Read summary Original

AI vs artists: who owns the future of music?

Generative AI threatens the music industry by creating original works from copyrighted material, sparking over 50 lawsuits. Industry experts recommend collaborative licensing frameworks rather than resistance, with Spotify now...

Read summary Original

Longtime Investor Warns the AI Industry Is Set to Collapse for a Basic Financial Reason

Veteran tech investor Roger McNamee, who successfully backed Google, Facebook, and Amazon, warns that the AI industry faces an imminent collapse despite over $700 billion in investments by 2025. McNamee argues that five major...

Read summary Original

OpenAI’s New Move: ChatGPT Knows What You Need Before You Ask

OpenAI launched ChatGPT Pulse, transforming its chatbot from reactive to proactive by delivering personalized updates through daily visual cards before users even ask, using chat history and integrations like Gmail and Google...

Read summary Original

Beware coworkers who produce AI-generated 'workslop'

"Workslop" describes low-quality AI work that looks legitimate but lacks substance. This may explain why 95% of organizations using AI report zero returns, as colleagues must fix inadequate content.

Read summary Original

How South Korea plans to best OpenAI, Google, others with homegrown AI

South Korea is investing $390 million in five local companies to develop Korean-focused AI models, aiming to reduce dependence on foreign AI technologies and strengthen national security.

Read summary Original

Google Photos May Get Nano Banana AI , Claims New Report

Google Photos is preparing to integrate Gemini's "Nano Banana" AI technology, which allows users to merge and remix multiple images using simple text prompts. App investigators discovered a hidden animation in the Android Photos...

Read summary Original

Back to all archives Load more stories