Search the archive

ALPHABET PLANS TO RAISE $80 BILLION FROM STOCK SALES TO FUND AI BUILDOUT

Alphabet plans to sell $80 billion in stock to fund investments into AI compute infrastructure due to unprecedented customer demand. The raise includes $10 billion from Berkshire Hathaway, $30 billion in an underwritten offering, and $40 billion from an at-the-market offering program for Class A and Class C shares expected to begin in the third quarter. Goldman Sachs, JPMorgan Chase, and Morgan Stanley are acting as joint book-running managers for the underwritten offering.

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 [21].

OPUS 4.8 JUST BROKE ARC-AGI-3

It tripled GPT-5.5's score.

MISTRAL SEARCH TOOLKIT FOR PRODUCTION AI PIPELINES

Mistral released Search Toolkit in public preview, an open-source framework that unifies data ingestion, retrieval, and evaluation within a shared interface.

CURSOR EXPANDS TEAMS USAGE LIMITS

Cursor announced higher Teams plan limits, a new Premium seat for heavy agent users, and additional spending controls for administrators.

Read the 2026 benchmark report from HUMAN

NEW SCREENSHOTS OF UPCOMING COPILOT SUPER APP

Microsoft looks likely to debut its unified Copilot app at Build 2026. Leaked screenshots show a GitHub Copilot tab, a Cowork tab, and a tab for Scout, an always-on AI agent. The company is trying to boost weak adoption by folding scattered tools into a single home. The integration of Teams hints that Scout may be able to run remotely.

MINIMAX M3

MiniMax M3 is an open weights model that achieves frontier-level performance on coding and agentic work. The model supports image and video input and can operate a desktop computer. It uses a new attention architecture that enables context scaling and can support ultra-long context windows of up to 1 million tokens. The model is available through MiniMax Code, the Token Plan, and MiniMax's API services.

COMPUTEX 2026 WILL BE NVIDIA'S BIGGEST EVENT OF THE YEAR. HERE'S WHAT TO EXPECT

Nvidia plans to unveil its N1X laptop chip at Computex 2026, featuring 20 ARM cores and an RTX 5070 equivalent GPU, promising improved VRAM allocation for AI applications. The company will also present its "Vera Rubin" AI platform for datacenters, aiming to strengthen its position in the AI market. Nvidia's focus on Physical and Agentic AI will highlight advancements in autonomous machines and robotics, while gaming announcements seem minimal.

CLAUDE OPUS 4.8: THE SYSTEM CARD

Claude Opus 4.8, released only six weeks after Opus 4.7, comes with a 244-page system card. While the updates are incremental, there is still a lot to talk about. Its capabilities are still well behind Mythos. This post looks at the system card, what's different between the two versions, and what this reveals about Mythos.

AGENTIC RL: TOKEN-IN, TOKEN-OUT DONE RIGHT

In reinforcement learning with LLMs, ensuring the model operates on the exact tokens sampled is crucial. Re-tokenizing can lead to drift and unreliable gradients. The solution involves never re-encoding decoded tokens and maintaining a buffer for sampled tokens to avoid drift and maintain accurate loss computation. This approach depends on a prefix-preserving chat template property, which most modern templates satisfy, ensuring reliable reinforcement learning loops without redundant re-rendering.

without the deploy cycle. Try AgentControl free

PI-DYNAMIC-WORKFLOWS

pi-dynamic-workflows is a Pi extension that adds a workflow tool. It allows assistants to write an economical JavaScript script that fans out work across many isolated subagents, then synthesizes the results. Subagents can read files, run shell commands, and call structured output compactly like a normal Pi turn. The tool is great for codebase audits, multi-perspective review, large refactors, and fan-out research.

INTRODUCING 1-BIT AND TERNARY BONSAI IMAGE 4B: IMAGE GENERATION FOR LOCAL DEVICES

Bonsai Image 4B is a family of compact image-generation models that can run high-quality diffusion inference on local hardware. The 1-bit variant is for applications where memory pressure, bandwidth, and deployment footprint are the primary restraints. The ternary variant has more representational flexibility, giving it improved visual quality and prompt fidelity while remaining extremely compact. The models can run directly on an iPhone.

GROK BUILD 0.1 ON API

xAI's grok-build-0.1 is now in public beta via the API, designed for agentic coding tasks like web development and debugging. The model processes over 100 tokens/second, costing $1 per million tokens in and $2 per million out. It integrates well with platforms like Grok Build, Cursor, and OpenClaw.

ECC

ECC is a comprehensive system for multi-harness agent workflows, featuring skills, instincts, memory optimization, and security scanning.

THE AI AGENT BOTTLENECK ISN'T MODEL PERFORMANCE — IT'S PERMISSIONS

Enterprise AI agents are struggling due to permissioning issues rather than model performance. Workday addresses this by using its system of record as the governance layer, integrating with Google's Gemini, and emphasizing agent accuracy. This setup ensures agents operate within defined user permissions, crucial for regulated sectors like HR and finance.

VERIFYING AGENTIC DEVELOPMENT AT SCALE

Cognition's Ido Pesok shares lessons from building autonomous end-to-end testing into Devin, noting that for the first time, more Devin sessions are now triggered asynchronously than interactively, making verified-before-merge results a hard requirement rather than a nicety. Devin's harness gained computer-use tools roughly six months ago, and the breakthrough came when engineers started running 10-20 Devins in parallel, each with its own dev server, something impossible on a single laptop.

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 [20].

EX-DEEPMIND RESEARCHERS RAISED $50M TO BUILD AI THAT FIGURES OUT WHICH SCIENTIFIC QUESTIONS ARE WORTH ASKING

Inherent, a London-based AI lab, is building a platform called Faraday aimed at figuring out which questions are worth asking.

OPENAI ROBOTICS IS HIRING

OpenAI Robotics is looking for full-stack hardware, ops, systems, and ML engineers to help it program and manufacture robots.

3 UPCOMING NOTEBOOKLM FEATURES WE ALL SHOULD BE WAITING FOR

NotebookLM's upcoming features include Personal Preferences, Connectors, and Canvas.

OPENAI OUTLINES PLAYBOOK FOR TRUSTWORTHY THIRD-PARTY AI MODEL EVALUATIONS

OpenAI published a comprehensive guide on May 28 for conducting trustworthy third-party evaluations of frontier AI models like GPT-5.5.

HOW TO AUTOMATE AI MODEL DOCUMENTATION WITH THE NVIDIA MCG TOOLKIT

NVIDIA's MCG Toolkit automates AI model documentation, creating comprehensive model cards in Model Card++ format rapidly.

Try Wispr Flow Free [6] | Get Flow

ANTHROPIC RAISED $65B IN SERIES H FUNDING

Anthropic announced a $65 billion Series H round at a $965 billion post-money valuation, citing strong enterprise adoption, $47 billion in run-rate revenue, and plans to expand compute capacity, research, and product development.

OPUS 4.8

Anthropic released Claude Opus 4.8 with benchmark improvements, adjustable effort controls, dynamic workflows in Claude Code, and a faster mode that became significantly cheaper.

HOW LONG IS ANTHROPIC'S LEASE WITH SPACEX? OPINIONS VARY

SpaceX earlier this month signed a major compute deal with Anthropic worth billions of dollars a month. However, Elon Musk recently downplayed the deal, saying that SpaceX had not committed to leasing its compute for years, even though it is possible that might happen. The agreement is actually a 180-day lease with a 90-day mutual cancellation thereafter. The short-term agreement was SpaceX's request as it may want the compute back at some point. Musk's statement directly contradicts SpaceX's S-1 filing, which presents the deal as a three-year agreement.

MICROSOFT TRIES TO GET BACK IN THE AI CODING GAME WITH NEW MODEL

Microsoft is developing a new AI model to strengthen its position in the AI coding arena. This effort highlights Microsoft's ongoing competition in AI development and its response to evolving industry demands. The initiative aims to enhance coding capabilities and support advancements in AI technology.

AGENT JUDGE: SOLVING LONG-CONTEXT EVALS FOR PRODUCTION AGENTS

Agent Judge improves evaluations for long-context, production agents by focusing on Search, Verification, and Adaptation. It tackles LLM judges' shortcomings by navigating long trajectories, verifying stateful actions against systems, and updating rubrics based on real feedback. Test results show Agent Judge, especially with refined rubrics, surpasses traditional LLM judges in accuracy and consistency, particularly in challenging scenarios.

HOW FAR BEHIND ARE OPEN MODELS?

Open models are generally not as capable as the best closed models. However, they aren't too far behind, with tests showing that they are only four to six months behind on public benchmarks. The gap was the smallest around the time of DeepSeek R1. It has since been growing.

INTRODUCING DYNAMIC WORKFLOWS

Jarred Sumner used dynamic workflows to rewrite Bun from Zig to Rust, achieving 99.8% test suite success with 750,000 lines of Rust in 11 days. Dynamic workflows involve Claude breaking tasks into subtasks, with agents running in parallel until results converge.

THE CURSOR DEVELOPER HABITS REPORT

Models now utilize more context to understand codebases, which reduces costs as input and cache-read tokens are cheaper than output tokens. This context-driven approach improves code calibration, increasing developer productivity and diff survival rates.

FOR OVER A DECADE, WE'VE ACCEPTED THAT END-TO-END BACKPROP IS THE ONLY WAY TO TRAIN DEEP NETWORKS

Holding the entire network in memory at once is why AI training is hitting a resource wall. Sakana Labs has found a new way to break the network into blocks and train them independently. The trick was to treat the network's forward pass like a diffusion model denoising a signal. This slashes the memory needed to train deep models.

MULTI-AGENT WORLD MODELS

NVIDIA γ-World is a generative world model that supports independently controllable, permutation-symmetric agents and delivers real-time rollouts with zero-shot generalization from two-player to four-player settings.

MINIMAX TEASES UPCOMING M3 MODEL WITH NEW SPARSE ATTENTION MECHANISM AND 15.6X LONG-CONTEXT RESPONSE SPEED BOOST

MiniMax has released a new in-depth technical report on the making of its popular M2 series of language models. The report sheds light on numerous engineering innovations and clever approaches. It teases a new sparse attention approach used in MiniMax's upcoming series of models that yields up to 15.6 times faster decode speed at long contexts. MiniMax's upcoming M3 models will make ultra-long-context AI agent deployment economically viable.

DATA ISN'T SCARCE. YOUR IMAGINATION IS

Asuka Zheng argues the "we're running out of training data" panic misses the actual shape of the data market, recounting her own SRE-replacement project that trained two world models until it stalled because end-to-end long-horizon incident trajectories from first anomaly to full resolution did not exist as a dataset.

SPACEX HAS ALMOST FINISHED WRITING V1.0 OF AN IN-HOUSE AI TRAINING STACK IN C

SpaceX's in-house AI training stack makes heavy use of pipeline parallelism by exact-mapping to 220k GB300s with 800G NICs, getting as close to bare metal as possible. The potential speed improvement is over an order of magnitude. SpaceX's next goal is to write the inference stack in C for simultaneous high-speed RL across a large block of GB300s.

A TOUR.

BYTEDANCE HAS HAD ENOUGH OF WAITING MONTHS FOR PROCESSORS, SO IT'S GOING TO MAKE THEM ITSELF

ByteDance has approached several external partners to work with to design a new chip to better support its AI infrastructure.

OPENAI PUBLISHED A FRONTIER GOVERNANCE FRAMEWORK

MISTRAL TO EXPLORE DESIGNING OWN CHIPS, CEO SAYS, AS IT RAMPS UP INFRASTRUCTURE BUILD

Mistral AI plans to design custom chips to control infrastructure and lower deployment costs as it expands its data center presence in Europe.

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 [28].

IBM'S "PROJECT LIGHTWELL"

Project Lightwell will establish a trusted enterprise clearinghouse to serve as a security coordination layer to help enterprises integrate secure patches directly into their existing software supply chains with enterprise-grade validation and lifecycle management.

AI IS CHANGING HOW CONSULTANTS GET PAID—AND MUCH MORE

Boston Consulting Group's revenue has grown 7% in the last year, and its headcount is expanding due to an infinite need from companies needing help to roll out AI.