TLDR 2026-04-15

Start free. $100 in credits →

ANTHROPIC ATTRACTS INVESTOR OFFERS AT AN $800 BILLION VALUATION

Anthropic has received several offers from investors for a new round of funding that could value the company at $800 billion or higher. The discussions are still early and the deal may fail to materialize. Anthropic has so far resisted such offers, so it is unclear if the company will accept investors' terms. It has also discussed a public listing as early as October.

LIKE ANTHROPIC, OPENAI WILL SHARE LATEST TECHNOLOGY ONLY WITH TRUSTED COMPANIES

OpenAI plans to share a new model called GPT‑5.4‑Cyber with hundreds of organizations before expanding release to other partners in the coming weeks. It plans to make the tool as widely available as possible while preventing misuse. The model is designed to identify security holes in software. OpenAI hopes to give defenders an edge over attackers by releasing the technology to a smaller group.

APPLE CHOOSES AMAZON SATELLITES FOR IPHONE, YEARS AFTER REJECTING STARLINK OFFER

Amazon has entered into an agreement with Apple to provide satellite service for its devices using the Amazon Leo network. Amazon recently signed a merger agreement to buy satellite operator Globalstar, which had already partnered with Apple for satellite messaging on the iPhone. Buying Globalstar will help Amazon enter the Direct-to-Device market, in which satellites provide connectivity to mobile phones. Apple is working on a framework for third-party apps to let developers add satellite connections to their apps.

MAX HODAK'S SCIENCE CORP. IS PREPARING TO PLACE ITS FIRST SENSOR IN A HUMAN BRAIN

Science Corporation plans to start human trials for its biohybrid brain-computer interface. It has hired a top neurobiologist to surgically place a sensor for a future interface into a patient's brain. The sensor will be implanted into the skull, but rest on top of the brain. The final device will be embedded with lab-grown neurons that can be stimulated with pulses of light and are designed to naturally integrate with the neurons in a patient's brain.

Register now

ANTHROPIC ADDS ROUTINES TO REDESIGNED CLAUDE CODE, HERE'S HOW IT WORKS

Anthropic's new repeatable routines feature for Claude Code runs tasks on Claude Code's web infrastructure, so users' Macs don't need to be online for each task. It is now available as a research preview. The Pro plan supports running 5 routines per day, the Max plan supports running 15 routines per day, and the Team/Enterprise plans support running 25 routines per day. Users can now run multiple Claude sessions side by side in a single window and manage them all from the sidebar.

MATT MULLENWEG SAYS “THE WHEELS HAVE FALLEN OFF” IN WIDE-RANGING WORDPRESS CRITIQUE

Matt Mullenweg has criticised the WordPress project, saying it spent years doing damage to itself and that it has a release culture that produces 'boring or mediocre crap'. He claims that the project isn't being beaten by competition but that it was in a state of self-inflicted decline. Some contributors believe that the project's slowdown was due to Mullenweg's public dispute with WP Engine in September 2024, but Mullenweg rejected the notion.

INSIDE GITHUB'S FAKE STAR ECONOMY

A study found 6 million fake stars across 18,617 repositories. AI/LLM repositories were the largest non-malicious category. There are at least a dozen websites selling stars for between $0.03 to $0.85 each. The FTC introduced a rule in 2024 banning fake social influence metrics with a penalty of $53,088 per violation. The SEC has already charged startup founders for inflating traction metrics during fundraising.

GOOGLE INTRODUCES “SKILLS” IN CHROME TO MAKE GEMINI PROMPTS INSTANTLY REUSABLE

Chrome has implemented 'Skills' in Chrome - reusable prompts that can be accessed while browsing with a single click. Skills make it easier to repeat tasks that were already possible with Gemini in Chrome. The feature makes it easy to save favorite prompts, making them easy to quickly access. The desktop version of Chrome will remember saved Skills across devices.

FULLY REMOTE)

TLDR's Applied AI team is tasked with making every process at TLDR legible to code, runnable by anyone, and composable into larger workflows. Join a small, fast moving team using the latest AI tools with an unlimited token budget. Learn more [15].

CYBERSECURITY LOOKS LIKE PROOF OF WORK NOW

A team evaluating Claude Mythos Preview's cyber capabilities found that they got better results the more tokens they spent.

PLAIN

Plain is a Python web framework for building apps designed for humans and agents.

SCALING MCP ADOPTION

This post discusses Cloudflare's best practices for securing MCP workflows and shares how the organization approaches deploying MCP.

ANTHROPIC SHIFTS ENTERPRISE BILLING TO PER-TOKEN PRICING. THE FLAT-FEE ERA IS OVER

Every major AI provider that runs agentic workloads is expected to move to usage-based billing within six months.

MECHANICAL SYMPATHY

Mechanical sympathy comes from an enormous context window learned over a human lifetime.

NAME-ONLY @CONTAINER QUERIES: A SOLUTION TO THE NAMING WARS

Name-only @container queries allow developers to use containers as a way to filter styles for specific use cases without the messiness of long classnames or the need for third-party tools.