Newsletter

Editorial

Coreset pruning cuts 80% of VLM tokens, keeps accuracy intact

The Frontier

Coreset pruning cuts 80% of VLM tokens, keeps accuracy intact

Vision-language models are drowning in their own pixels, and a Chinese Academy of Sciences team has a training-free fix: cut up to 80% of the visual tokens a multimodal model has to chew through — and lose nothing. The CAS Institute of Automation's Zidong Taichu team has released GMC (Grounded Message Coreset Pruning), a training-free method that cuts up to 80.2% of a vision-language model's visual tokens while retaining 97.78% of full-model capability — and in its most aggressive…

Blacksmith raises $45M to fix AI coding's testing bottleneck

The Arena

Blacksmith raises $45M to fix AI coding's testing bottleneck

AI is writing code faster than teams can check it — and investors are betting that fixing that gap is the next big business in the AI coding boom. Blacksmith, a startup that tests and validates software before it reaches production, has raised $45 million in a Series B led by Peak XV Partners, valuing the company at $550 million — nearly ten times the $60 million valuation it carried when it raised a $10 million Series A less than a year ago.

Lovable raises $400M at $13.3B, doubling valuation in eight months

The Arena

Lovable raises $400M at $13.3B, doubling valuation in eight months

Money moved on three fronts at once: Europe's hottest vibe-coding startup doubled its valuation, DeepSeek's long-rumored agent harness surfaced under a black-whale logo, and Mistral started charging a premium for EU data residency. Lovable, the Stockholm-based startup whose AI app builder helped turn "vibe coding" into a mainstream workflow, has raised $400 million at a $13.3 billion valuation — roughly double the $6.6 billion it commanded in December, per a Wall Street Journal report.

Tencent's AI-driven ads beat Q2 estimates, but profit misses

The Arena

Tencent's AI-driven ads beat Q2 estimates, but profit misses

Tencent beat second-quarter revenue estimates on accelerating game sales and AI-driven advertising, but the results laid bare the tension at the heart of its AI bet: core profit missed expectations, and capital spending jumped 65% quarter over quarter. Tencent posted revenue of 204.78 billion yuan ($30.36 billion) for the quarter ending in June, up 11% year on year and ahead of the 202.17 billion yuan analysts expected.

Deep Dive — Samsung's chip lab bet on Claude: 15x speed, zero margin for error

The Arena

Deep Dive — Samsung's chip lab bet on Claude: 15x speed, zero margin for error

Samsung Electronics has confirmed it put Anthropic's Claude to work inside its semiconductor design and verification flow — and the early numbers are startling. In one customized system-on-chip verification project, a task expected to take more than a month was finished in two days, internally assessed as roughly 15 times faster. In another, a second-year engineer with no prior experience of Claude Code completed in a day development work that normally requires a month of learning and building.

MLS-Bench: frontier models can't invent new ML methods

The Frontier

MLS-Bench: frontier models can't invent new ML methods

A benchmark from UC Berkeley, Princeton and Tsinghua researchers asks whether AI systems can invent the kind of generalizable methods that built modern machine learning — residual connections, attention, Adam — or merely recombine what humans already made. The early verdict is blunt: so far, they can't, even with the answer key in hand. MLS-Bench, built by 28 researchers across UC Berkeley, Princeton, Tsinghua and seven other universities, finds that frontier models can tune machine learning…

China approves mandatory autonomous-driving safety standard

The Guardrails

China approves mandatory autonomous-driving safety standard

China's market regulator just drew a regulatory line under two of AI's most safety-critical uses: driving cars and diagnosing patients. Beijing approved a mandatory national standard for autonomous driving system safety alongside new quality rules for AI medical devices, part of a 338-standard batch that signals how the country intends to govern AI where failures have physical consequences. China's market regulator approved a mandatory national standard for autonomous driving system safety…

AI 101 — What is chain of thought?

The Frontier

AI 101 — What is chain of thought?

Chain of thought is the step-by-step reasoning a language model generates before it gives a final answer — the "show your work" part of AI problem-solving. Instead of jumping straight from question to answer, the model writes out its intermediate steps, and that simple change measurably improves how well it handles math, logic, and anything with multiple stages. The idea surfaced in a January 2022 paper from Google Research: the authors found that asking a model to work through a problem aloud…

See all