AI News

AI news that matters. Updated daily.

Monday, April 20, 2026
Viewing Date
SunMonTueWedThuFriSat
13 stories 4 categories ~2m avg read

No stories match your filters.

Companies Apr 20

Amazon Puts Another $5B Into Anthropic, Secures $100B AWS Spending Pledge

$5 billion in, $100 billion out. That's the structure of Amazon's latest deal with Anthropic - a ratio that tells you exactly what Amazon is actually buying here.

Tools Apr 20

Google Expands Gemini in Chrome to Seven Asia-Pacific Countries

Google's Gemini assistant inside Chrome is now live in seven more countries: Australia, Indonesia, Japan, the Philippines, Singapore, South Korea, and Vietnam. The rollout was announced by the company on April 20, 2026.

Companies Notable Apr 20

Adobe Launches AI Agent Platform for Customer Experience Workflows

Adobe is entering the AI agent market for customer experience (CX), launching a platform that automates multi-step customer service workflows without human intervention at each step.

Research Notable Apr 20

The Sentence Pattern That Gives Away AI-Written Content

"It's not just an AI writing tell - it's practically a confession."

Policy Notable Apr 20

NSA Using Anthropic's Restricted Mythos Model Despite Pentagon Dispute

The NSA is reportedly using Anthropic's Mythos - a restricted AI model not publicly available - despite an ongoing dispute between Anthropic and the Pentagon over the company's government contracts.

Companies Notable Apr 20

Deezer: 44% of Songs Uploaded Daily Are AI-Generated

44%. Nearly half of every song uploaded to Deezer on any given day was made by an AI.

Research Notable Apr 20

What AI Chatbots Actually Do to Your Server When You Give Them a URL

What happens on your web server the moment someone pastes your URL into an AI chatbot? A developer ran that experiment, prompting ChatGPT, Claude, Perplexity, and Gemini with URLs and watching the Nginx access logs in real time. The full findings are on the Surfaced By blog.

Research Notable Apr 20

Research Suggests Daily AI Chatbot Use May Weaken Independent Thinking

What happens when you stop practicing the thinking you've handed off to a chatbot?

Policy Notable Apr 20

Atlassian Quietly Turns On AI Training Data Collection by Default

Atlassian has switched its AI training data collection to opt-out rather than opt-in across its suite of products, which includes Jira, Confluence, Trello, and Bitbucket. Unless an admin goes in and changes the settings, customer data is now being used to train Atlassian's AI models.

Companies Notable Apr 20

Claude Design Exposes a Structural Crack in Figma's Business Model

Only 33% of Figma's users in Q1 2025 were actually designers. Developers made up 30%, and everyone else - marketers, PMs, ops people building reports and decks - accounted for the remaining 37%. That number is now Figma's biggest liability.

Companies Notable Apr 20

Chinese Tech Workers Are Being Told to Train Their Own AI Replacements

Your boss wants you to build a version of yourself that doesn't need a salary. That's the situation facing a growing number of tech workers in China, where managers are reportedly instructing employees to train AI agents - software programs that can take actions and make decisions on someone's behalf - that can replicate their specific skills and work style.

Tools Apr 20

Claude Token Counter Gets Model-by-Model Comparison View

Simon Willison updated his Claude Token Counter to show token counts across different Claude models side by side. He published the details on his blog, and the update is small but practically useful for anyone calling Claude via API.

Companies Notable Apr 20

Hyatt Deploys ChatGPT Enterprise Across Its Global Hotel Workforce

Hyatt is putting ChatGPT Enterprise in front of employees across its global hotel portfolio, according to an announcement from OpenAI. The deployment covers productivity work and operations, with the company using both GPT-5.4 - the latest model in OpenAI's GPT-5 family - and Codex, OpenAI's coding-focused AI assistant.