Search the archive

INTEL CEO TO OVERSEE AI EFFORTS AFTER EXECUTIVE DEPARTS FOR OPENAI

Intel's CTO Sachin Katti, who had led the chipmaker's AI efforts since January, has departed for OpenAI, where he will design and build compute infrastructure.

BUILD TIMES FOR GIGAWATT-SCALE DATA CENTERS CAN BE TWO YEARS OR LESS

Anthropic will be the first to use a 1-GW data center early next year.

LOVABLE NEARS 8 MILLION USERS IN FIRST YEAR

Lovable has surged to nearly 8 million users and sees 100,000 new projects built daily.

CHINA'S DEEPSEEK MAKES RARE PUBLIC COMMENT, CALLS FOR AI ‘WHISTLE-BLOWER' ON JOB LOSSES

A senior researcher warned that society is in a “honeymoon phase” with AI before widespread job automation and said AI companies need to alert the public about which jobs will soon become obsolete.

Read _The OpenAI Monetization Blueprint_ on the Metronome blog

EARLY LOOK AT IMAGES GENERATED BY NANO BANANA 2

Nano Banana 2 is expected to launch on November 11.  The model will have 2K native output and huge improvements across many areas. This article provides some examples of outputs from the new model along with comparisons with outputs from Nano Banana 1.

YOU CAN NOW GET MORE CODEX USAGE FROM YOUR PLAN AND CREDITS

GPT-5-Codex-Mini, a more compact and cost-efficient version of GPT-5-Codex that allows roughly 4x more usage at a slight capability tradeoff, is now available in OpenAI's CLI and IDE extension. OpenAI has increased rate limits by 50% for ChatGPT Plus, Business, and Edu users. There is now priority processing for ChatGPT Pro and Enterprise customers.

OPENAI'S $1 TRILLION INFRASTRUCTURE SPEND

OpenAI has committed to spending $1.15 trillion on hardware and cloud infrastructure between 2025 and 2035. The estimated annual spending and commitments convey an absolutely enormous level of potential and ambition. OpenAI will need to grow from around $10 billion in 2024 revenue to $577 billion by 2029. This article takes a look at the numbers.

THE STATE OF AI IN 2025: AGENTS, INNOVATION, AND TRANSFORMATION

Most organizations are still in the experimentation or piloting phase. High performers use AI to drive growth, innovation, and cost. Redesigning workflows is a key success factor. The potential impact of AI on employment is still debated - 43% of workers in a survey expected no change in the overall size of the workforce in their organizations in the coming year.

report

INTRODUCING NESTED LEARNING: A NEW ML PARADIGM FOR CONTINUAL LEARNING

Continual learning, the ability to acquire new knowledge without forgetting old information, remains a major obstacle to AGI since current models suffer from “catastrophic forgetting,” where learning new tasks diminishes proficiency on previous ones. A proof-of-concept architecture called Hope, a self-modifying model that optimizes its own memory, outperformed modern recurrent models and transformers on language modeling and excelled particularly at long-context needle-in-a-haystack tasks.

QUANTIZATION IS NOT A COMPROMISE — IT'S THE NEXT PARADIGM

K2-Thinking's release has led many developers to be curious about its native INT4 quantization format. INT4 accelerates RL training due to its low-latency profile. Quantization is no longer a tradeoff. Native low-bit quantization will become a standard paradigm for large model training with the evolution of param-scaling and test-time-scaling.

CLAUDE CODE INFRASTRUCTURE SHOWCASE

This repository contains a curated reference library of production-tested Claude Code infrastructure. It contains production-tested infrastructure for auto-activating skills via locks, specialized agents for complex tasks, and more. Using this infrastructure, skills will suggest themselves based on context, hooks will trigger skills at the right time, modular skills stay under context limits, dev docs preserve knowledge across resets, and agents will streamline complex tasks.

DEBT HAS ENTERED THE AI BOOM

Companies are pouring billions of dollars into cutting-edge computing facilities. The sheer scale of spending is causing investors to be concerned. Tech giants are turning to financing maneuvers that may add to the risk. Hyperscalers are leveraging a growing list of complex debt-financing options, including corporate set, securitization markets, private financing, and off-balance-sheet vehicles.

OPENAI: AI PROGRESS AND RECOMMENDATIONS

The gap between everyday use of AI and its core capabilities is massive. Chatbots are used to write emails or replace web searches, while the underlying models outperform human experts in math and coding competitions. OpenAI is advocating for light regulation now to cement its market position before the next leap in capability. Once systems become transformative enough to warrant a coordinated national response, there will be a shift to federal oversight that overrides state-level restrictions.

managed tokens, and more. Watch the Auth0 technical demo

UNDERWATER CABLES ARE A VITAL PIECE OF THE AI BUILDOUT AND INTERNET — INVESTMENT IS BOOMING

Investment into new subsea cable projects is expected to reach around $13 billion between 2025 and 2027.

TERMINAL-BENCH 2.0 LAUNCHES ALONGSIDE HARBOR, A NEW FRAMEWORK FOR TESTING AGENTS IN CONTAINERS

The releases aim to address long-standing pain points in testing and optimizing AI agents.

OPENAI ASKED TRUMP ADMINISTRATION TO EXPAND THE CHIPS ACT TAX CREDIT TO COVER DATA CENTERS

A letter from OpenAI's policy chief to the Trump administration proposes expanding the Chips Act beyond semiconductors to “de-risk” data-center buildouts and help close the “electron gap” with China, which added 429 GW of power capacity in 2024 compared to America's 51 GW.

HYBRID MODELS IN VLLM

Hybrid LLMs that combine attention with alternatives such as Mamba or linear attention are now first-class citizens in vLLM V1, enabling efficient long-context inference for models including Qwen3-Next and Granite 4.0.

Download the free whitepaper

SAM ALTMAN ON OPENAI'S REVENUE

Sam Altman revealed OpenAI's annual revenue run rate surpassed $20B, with future infrastructure commitments reaching $1.4 trillion. Plans include enterprise offerings, AI cloud services, and ventures into consumer devices and scientific discovery.

OPENAI READIES GPT-5.1 THINKING MODEL AHEAD OF GEMINI 3 PRO

GPT-5.1 Thinking has been spotted in a backend component responsible for powering advanced reasoning processes within ChatGPT, indicating a GPT-5.1 release is close. Rumors point to a lineup of three models (Mini, Thinking, and a possible Codex-focused upgrade), each designed to fit different user needs and compute constraints. OpenAI is likely aiming to synchronize the release with Google's upcoming Gemini 3 Pro launch.

$350B

Google plans to invest even more money into Anthropic. It is now in talks for a new funding round that could value Anthropic at $350 billion. The new deal could involve Google providing additional cloud computing services to Anthropic, a convertible note, or a priced funding round early next year. Google last month announced that it had inked a deal worth tens of billions for Anthropic to access Google's AI compute.

FORECASTS OF AI & ECONOMIC GROWTH

Most economists forecast 0.1-1.5% annual excess growth through 2035, while AI insiders predict 3-30%. Most economists, though, assume that today's ChatGPT is essentially the ceiling of AI progress and that visions of a virtual AI coworker won't really materialize, but how is it possible that cheap, PhD-level intelligence wouldn't boost productivity? The forecasts imply that either AGI will be far less general than claimed or that adoption will be dramatically slower than historical technology diffusion. AI's 2024 economic footprint appears to be around 0.5% of GDP.

5 THOUGHTS ON KIMI K2 THINKING

Kimi K2 Thinking, a reasoning MoE model developed by China's Moonshot AI team, shows just how many people are learning quickly to train excellent AI models. Early reports note a distinctive style and writing quality - the model is beating leading closed models on some benchmarks such as Humanity's Last Exam and BrowseComp. It is the closest open models have been to the closed frontier of performance ever.

Identity Challenge_

KIMI K2 THINKING

Chinese lab Moonshot AI released an open-source reasoning model that matches or exceeds GPT-5 and Claude Sonnet 4.5 on some benchmarks, including a 44.9% on Humanity's Last Exam and 71.3% on SWE-Bench Verified.

GEMINI FILE SEARCH TOOL

Google introduced File Search, a built-in RAG system for the Gemini API that simplifies retrieval tasks. It includes vector search, automatic citations, and broad file format support with free storage and on-demand embedding.

ACCUMULATING CONTEXT CHANGES THE BELIEFS OF LANGUAGE MODELS

Extended conversations can reshape how AI models answer the same question. After 10 rounds of debate about moral dilemmas, GPT-5 reversed its position 54.7% of the time when asked again. When models changed their stated beliefs, they also chose different tools to accomplish tasks.

LEAVING META AND PYTORCH

Soumith Chintala has left Meta after working at the company for 11 years. Chintala spent almost eight years leading PyTorch, taking it from nothing to more than 90% adoption in AI. He plans to move to something small and new. PyTorch has been left in capable hands, and Chintala is confident that its values will stay intact, but things may change in flavor because he will no longer impose his own taste from the top.

GOOGLE DEPLOYS NEW AXION CPUS AND SEVENTH-GEN IRONWOOD TPU — TRAINING AND INFERENCING PODS BEAT NVIDIA GB300 AND SHAPE 'AI HYPERCOMPUTER' MODEL

Google Cloud has introduced new AI-oriented instances powered by its own Axion CPUs and Ironwood TPUs. The new instances are aimed at both training and low-latency inference. Servers running these chips come equipped with Google's custom Titanium-branded controllers, which offload network, security, and I/O storage processing, enabling better management and higher performance. More details about Google's in-house designed processors are available in the article.

Try the AI solution that actually works.

ELON MUSK SAYS TESLA NEEDS TO BUILD ‘GIGANTIC CHIP FAB' TO MEET AI AND ROBOTICS NEEDS

Tesla doesn't yet manufacture its own microchips, but it has been designing custom chips for autonomous driving for several years.

INTRODUCING PARALLEL SEARCH: THE HIGHEST ACCURACY WEB SEARCH API ENGINEERED FOR AI

The Parallel Search API is a web search tool designed for AI agents engineered to deliver the most relevant, token-efficient web data at the lowest cost.

MICROSOFT FORMS SUPERINTELLIGENCE TEAM UNDER AI CHIEF SULEYMAN ‘TO SERVE HUMANITY'

In a thinly veiled dig at other AI labs, Mustafa Suleyman said the team will focus on practical technologies—like digital companions, disease diagnostics, and renewable energy—rather than “ill-defined and ethereal superintelligence."

ERIC ZELIKMAN, A TOP AI RESEARCHER WHO DEPARTED XAI IN SEPTEMBER, IS RAISING $1 BILLION AT A $4 BILLION VALUATION FOR HIS NEW STARTUP

Eric Zelikman, who recently left xAI, is raising $1 billion for his startup Humans& at a $4 billion valuation.

PERPLEXITY UPGRADED COMET ASSISTANT

Perplexity's Comet Assistant has been redesigned to handle more complex, multi-tab tasks with improved web environment interaction.

GOOGLE PLANS TO RELEASE GEMINI 3 PRO PREVIEW IN NOVEMBER

Google's Gemini 3 Pro preview model has appeared in VertexAI code, labeled '11-2025', suggesting a November launch. The timeline matches prior testing activity observed in AI Studio. Google has not issued any official statements about the launch.

OPENAI ISN'T YET WORKING TOWARD AN IPO, CFO SAYS

OpenAI's Chief Financial Officer, Sarah Friar, said the company is not planning to IPO in the short term. It will instead prioritize growth and R&D over profitability. OpenAI is hoping the government will support its offers to increase data center capacity by helping to guarantee the financing for the chips behind its deals. The depreciation rates of AI chips are uncertain, making it more expensive for companies to raise the debt needed to buy them. Friar claims OpenAI could reach profitability on very healthy gross margins if it weren't seeking to invest so aggressively.

GOOGLE GEMINI'S DEEP RESEARCH CAN LOOK INTO YOUR EMAILS, DRIVE, AND CHATS

Google Gemini's "Deep Research" now uses Gmail, Drive, and Chat data for AI-generated reports. The feature can create detailed research plans and reports, integrating personal workspace data with web searches. Currently available on desktop, Google plans to roll out mobile access soon.

SEMANTIC SEARCH FOR CURSOR'S CODING AGENT

Cursor's coding agent improved performance by incorporating custom semantic search over codebases, leading to higher accuracy, fewer user iterations, and better retention of code changes.

THOUGHTS BY A NON-ECONOMIST ON AI AND ECONOMICS