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• Dec 11, 2025

New York’s new law forces advertisers to say when they’re using AI avatars

New York Governor Kathy Hochul signed groundbreaking legislation requiring advertisers to clearly disclose when they use AI-generated avatars in advertisements, marking the first law of its kind in the United States. The bill mandates "conspicuous" notices when synthetic avatars don't represent real people, with violations resulting in $1,000 fines for first offenses and $5,000 for repeat instances. A companion law...

Read summary Original source

Latest stories

Runway claims its GWM-1 “world models” can stay coherent for minutes at a time

Runway launched GWM-1 "general world models" that stay coherent for minutes, using three separate models instead of one unified system. The company now faces tough competition from big tech, losing its earlier first-mover...

Read summary Original

Ai2's Tim Dettmers: AGI is a fantasy

Allen Institute researcher Tim Dettmers argues that artificial general intelligence is fundamentally impossible due to hardware limitations, predicting only one to two years of GPU scaling remain before physical constraints halt...

Read summary Original

Inception Labs: Making LLMs Faster and More Cost-Efficient

Inception Labs has developed Mercury, a diffusion-based language model that generates tokens in parallel rather than sequentially like traditional autoregressive models such as ChatGPT and Claude, making it 5-10 times faster and...

Read summary Original

Disney says Google AI infringes copyright “on a massive scale”

Disney sent a cease and desist letter to Google alleging the tech giant infringes its copyrights "on a massive scale" by training AI models on Disney content and allowing users to generate Disney characters through its AI tools....

Read summary Original

Cursor Launches an AI Coding Tool For Designers

Cursor, the AI coding startup valued at $30 billion, launched Visual Editor, allowing designers to create web applications using natural language commands alongside traditional design controls. The tool bridges the gap between...

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 fires back at Google with GPT-5.2 after ‘code red’ memo

OpenAI launched GPT-5.2 amid fierce competition with Google's Gemini 3, following CEO Sam Altman's internal "code red" memo about declining ChatGPT traffic and lost market share. The new model comes in three versions targeting...

Read summary Original

Google's Nano Banana Is Highlighting A Big Problem With AI Image Generation

Google's new Nano Banana AI model creates highly realistic images, but critics worry about potential problems with the technology and broader AI image generation issues.

Read summary Original

Another AI-Powered Children’s Toy Just Got Caught Having Wildly Inappropriate Conversations

Researchers discovered another AI-powered children's toy having inappropriate conversations, with the Alilo Smart AI bunny explaining sexual fetishes, bondage, and safe words to kids during testing. The $85 toy, intended for ages...

Read summary Original

You Can Photoshop With ChatGPT Now — But Why Bother?

Adobe and OpenAI's partnership now lets users control Photoshop, Acrobat, and Adobe Express through ChatGPT voice commands instead of manual editing. While the AI can perform basic tasks like adjusting contrast or creating...

Read summary Original

Free Photoshop in ChatGPT: How to edit photos with AI, no experience necessary

Adobe launched free integrations of Photoshop, Express, and Acrobat directly within ChatGPT, allowing users to edit photos, create visual content, and manage PDFs using simple conversational prompts instead of learning complex...

Read summary Original

The AI Voice Shortcut That Unlocks Serious Productivity Gains

Voice dictation technology like Wispr Flow is revolutionizing productivity by allowing entrepreneurs to speak at 150 words per minute instead of typing at 60, with AI cleaning up rambling speech into professional text. This shift...

Read summary Original

Can OpenAI Respond After Google Closes the A.I. Technology Gap?

Google closed the AI gap with OpenAI, prompting OpenAI to prepare new technology. This marks intensified competition between the tech giants battling for AI supremacy.

Read summary Original

Disney invests $1 billion in OpenAI, licenses 200 characters for AI video app Sora

Disney's $1 billion investment in OpenAI transforms their previous adversarial relationship into a partnership that allows Sora users to create videos featuring 200+ Disney characters including Mickey Mouse and Darth Vader. This...

Read summary Original

AI startup Tavus founder says users talk to its AI Santa ‘for hours’ per day

AI Santa is back for his second year at the North Pole, and this time he's watching you more closely than ever. The artificial intelligence from startup Tavus can now see children's facial expressions, remember their...

Read summary Original

OpenAI Image Generation

OpenAI's GPT Image model and Images API enable developers to generate high-quality images from text prompts directly within JavaScript and TypeScript applications, transforming image creation from a novelty into a practical...

Read summary Original

Building a High-End AI Desktop

A Reddit user discovered a discounted Grace-Hopper server with dual H100 GPUs selling for €10,000 and transformed the enterprise datacenter equipment into a functional desktop capable of running 235B parameter AI models. The...

Read summary Original

Adobe Brings Photoshop, Express and Acrobat Features to ChatGPT

Adobe integrated Photoshop, Express, and Acrobat features into ChatGPT, allowing users to ask the chatbot to edit images, modify PDFs, and create animations without leaving the conversation. Users can now request specific photo...

Read summary Original

I asked ChatGPT to gently roast me based on my year in queries - and it was uncanny

ChatGPT analyzed a user's year-long conversation history and provided surprisingly accurate personality insights, including a humorous observation about their tendency to jump from simple questions to complex research topics....

Read summary Original

This PowerShell script promises to remove every AI feature from Windows

Zoicware's RemoveWindowsAI is an open-source PowerShell script that completely removes Windows 11 AI features like Copilot and Recall by deleting core packages and preventing reinstallation.

Read summary Original

How to use TOON to reduce your token usage by 60%

Token-Oriented Object Notation (TOON), a new serialization format designed specifically for large language models, reduces token usage by 30-60% compared to JSON by declaring field names once instead of repeating them for each...

Read summary Original

How AI Is Restoring Voices Lost To Disability

The Scott-Morgan Foundation and D-ID created SMF VoXAI, an AI system with photorealistic avatars that helps people with severe speech disabilities communicate through real-time voice generation and emotional expression.

Read summary Original

Meta is reportedly working on a new AI model called 'Avocado' and it might not be open source

Meta is developing a new AI model codenamed "Avocado" that could mark a dramatic shift from the company's open-source approach, with the model expected in 2026 and likely to be proprietary. This pivot comes as CEO Mark Zuckerberg...

Read summary Original

OpenAI's house of cards seems primed to collapse

OpenAI has lost its commanding position as competitors have caught up and surpassed its technology. China's DeepSeek model overtook ChatGPT as the top app download, wiping out $1 trillion in market value, while Google's Gemini 3...

Read summary Original

Back to all archives Load more stories