Search the archive

REPORT REVEALS WHAT MARKETERS OVERLOOK AT THE OPT-IN MOMENT AND WHY IT MATTERS

New research from Mailchimp shows most brands over-ask for data at sign-up, hurting trust and list quality from the start.

SCROLL-STOPPING EMAIL INSPO

Email inspo from Otto Skin Goods, fu:di, and Candy Kittens.

Try now for free

GEMINI ‘SCREEN AUTOMATION' WILL PLACE ORDERS, BOOK RIDES FOR YOU ON ANDROID

The Google app 17.4 beta contains strings about a 'Get tasks done with Gemini' feature. The control is also referred to as 'screen automation'. Google has been working on letting Gemini control Android phones for a while. The feature will only work in certain apps.

IN SURPRISING MOVE, HP CEO STEPS DOWN TO LEAD PAYPAL

HP CEO Enrique Lores will replace PayPal's previous CEO, Alex Chriss, starting March 1. Lores has served on the PayPal board for nearly five years and as Board Chair since July 2024. HP was reportedly blindsided by the decision, which was only communicated to the company's board in recent weeks. Lores worked at HP for 36 years, joining the company as an engineering intern. HP has named board member Bruce Broussard as interim CEO.

DATA CENTERS IN SPACE MAKES NO SENSE

Data centers in space face massive challenges such as radiation, cooling, latency, and launch costs. There are other fundamental issues that make them unfeasible: competing in the market would require launching hundreds of thousands, if not millions, of satellites into space, and satellites can't be upgraded at scale. Data centers in space only make sense if they are cost-effective relative to normal data centers. The idea makes less and less sense with every improvement in normal energy production.

AV RIDESHARE FLEET SIZING: WHY THE OPTIMUM IS JUST SHY OF PEAK

The optimal fleet size for automatic vehicles is not peak demand, but it's also not average. The economic optimum is 98-99% demand coverage for most operators. Oversizing fleets destroys pricing power faster than it increases utilization. AV unit economics only work with aggressive off-peak price differentiation.

FULLY REMOTE)

As the first engineer on TLDR's new Applied AI team, you'll build AI agents and composable Claude Skills to let non-technical teammates create their own AI workflows. Learn more [10].

ANTHROPIC PERFORMANCE TEAM TAKE-HOME FOR DUMMIES

Anthropic used to give prospective performance engineering candidates a notoriously difficult take-home exam designed to assess technical ability and judgment under time pressure. This practice was retired after Claude Opus 4.5 scored higher than any candidate ever after just two hours of work. The company has released the assignment publicly along with a post detailing its design principles. This post breaks down the challenge from start to finish and explains everything from scratch.

APPLE'S XCODE NOW SUPPORTS THE CLAUDE AGENT SDK

Xcode 26.3 introduces a native integration with the Claude Agent SDK. Developers now have the full power of Claude Code directly in Xcode without having to leave the IDE. Claude can work autonomously on sophisticated, long-running tasks. The integration supports visual verification with Previews, reasoning across projects, autonomous task execution, and interfacing through the Model Context Protocol. Xcode 26.3 is now available as a release candidate for all members of the Apple Developer Program.

LESSONS LEARNED SHIPPING 500 UNITS OF MY FIRST HARDWARE PRODUCT

Simon Berens left his job as a software engineer in January 2025 to launch a hardware product, Brighter: the world's brightest lamp. Berens had no prior experience in hardware and was counting on being able to pick it up quickly with the help of other engineers. This article takes a look at the challenges Berens had to overcome, like making the lamp in the first place, tariffs, and manufacturing issues, and details the lessons that he learned along the way.

SAM ALTMAN REVEALS WHY OPENAI IS POISED TO MAKE THE BIGGEST BUSINESS BETS EVER

Sam Altman has a tendency to race ahead with ideas that excite him. While it is a tentpole of his success, it has gotten him in trouble in the past. Altman's itch to release products quickly is informed by his study of Xerox PARC, a legendary Silicon Valley research lab known for inventing the modern graphical user interface, laser printers, and the computer mouse, yet failing to commercialize any of them. Altman's reputation for shiny object syndrome can be harmful - OpenAI has been criticized in the past for releasing products without proper safety testing and for shipping features that prioritize engagement over psychological well-being.

VC-BACKED STARTUPS ARE LOW STATUS

When institutions start to lose their soul, the people inside them start to care about different things.

MICROSOFT'S PIVOTAL AI PRODUCT IS RUNNING INTO BIG PROBLEMS

Only a small proportion of subscribers to Microsoft's enterprise suite use Copilot, and the percentage of people who favor it over other tools has decreased in recent months.

THREAT OF NEW AI TOOLS WIPES $300 BILLION OFF SOFTWARE AND DATA STOCKS

Anthropic's announcement that it would add new legal tools to its Co-work assistant resulted in shares of some companies that provide legal tools or research databases falling sharply.

THREE LENSES ON COORDINATION

Staying within spec is easy if more information helps rule out possible futures, otherwise, the system will need help.

SOVEREIGNTY GAPS

Sovereignty Gaps are situations where the central structures of society require a function they cannot perform, requiring new actors to perform it on their behalf.

DISNEY NAMES PARKS CHIEF D'AMARO AS NEW CEO, SUCCEEDING IGER

Josh D'Amaro, who has spent 28 years at Disney, mostly working in the theme-parks business, will succeed Bob Iger to become Disney's new chief executive officer starting on March 18.

Save your spot

secure AI across your organization. Topics include

* AI AGENTS UNCOVERED: AWS explains how AI agents transform workflow and create value across enterprises. * AI AGENTS UNDER ATTACK: Which security blindspots are introduced by agents in an emerging threat landscape. * SECURING AI AGENTS NOW: Learn how real-world security teams are scaling AI security measures.

CLAUDE IN XCODE

Xcode 26.3 introduces native support for Claude Agent SDK, enabling full agentic capabilities like subagents, background tasks, and plugins within Apple's IDE. This brings Claude Code's functionality directly into developers' workflows.

INTEL IS MOVING INTO GPUS AND HAS HIRED A CHIEF ARCHITECT, CEO LIP-BU TAN SAYS

Intel is expanding into the GPU market by hiring a new chief architect, as announced by CEO Lip-Bu Tan. GPUs are crucial for AI infrastructure and are in high demand, driven by LLMs. Despite recent challenges, Intel's stock has rallied, buoyed by optimism in its foundry business and significant investments.

OPENAI POACHED ITS NEW SAFETY EXECUTIVE FROM ANTHROPIC

OpenAI hired Dylan Scandinaro from Anthropic as its new "head of preparedness." Scandinaro emphasized the rapid advancement of AI and the associated risks. He stressed the urgency of addressing these challenges.

OPEN SOURCE AI ECOSYSTEM

This blog explores open-source AI trajectory since the "DeepSeek Moment," highlighting long-term strategies by major organizations and forecasting sustained momentum through open artifact sharing and deployment-first design.

DEEP DIVE: HOW CLAUDE CODE'S /INSIGHTS COMMAND WORKS

The /insights command in Claude Code generates an HTML report that analyzes usage patterns across a user's Claude Code sessions. The tool is designed to help users understand how they interact with Claude, what's working well, where friction occurs, and how to improve workflows. This post takes a look at how the tool works. For better insights, use Claude Code regularly, give feedback, don't filter yourself, and check in monthly to see how patterns evolve.

EXPENSIVELY QUADRATIC: THE LLM AGENT COST CURVE

It doesn't take much to start spending a significant amount of tokens on context. This problem compounds as conversations get longer and only gets exponentially more important with agent and subagent workflows. Agent developers need to keep this problem in mind when developing AI tools. This post provides an interactive tool that shows how quickly context size can take up a large percentage of a cache read.

Get early access

QWEN3-CODER-NEXT FOR AGENTIC CODING

Alibaba's Qwen3-Coder-Next is a new open-weight model fine-tuned for coding agents. Built on a hybrid MoE architecture, it excels in executable synthesis and RL-based environment interaction, achieving strong agentic coding performance at lower inference cost.

GLM-OCR (HUGGING FACE REPO)

GLM-OCR is a multimodal OCR model for complex document understanding. It integrates the CogViT visual encoder pre-trained on large-scale image–text data, a lightweight cross-modal connector with efficient token downsampling, and a GLM-0.5B language decoder. The model delivers robust and high-quality OCR performance across diverse document layouts. An SDK for efficient and convenient use is available.

PERPLEXITY CANNOT ALWAYS TELL RIGHT FROM WRONG

Perplexity is a function that measures models' overall level of surprise when encountering a particular output. It has gained significant traction in recent years as both a loss function and as a simple-to-compute metric of model quality. However, it may be an unsuitable metric for model selection. Perplexity will not always select for the most accurate model - any increase in model confidence must be accompanied by a commensurate rise in accuracy for the new model to be selected.

800K+ VERIFIABLE SWE TASKS

SWE-Universe presents a scalable method for generating verifiable software engineering environments from GitHub PRs. With in-loop hacking detection and self-verification, the system enables large-scale mid-training for coding agents, producing over 800K tasks.

GEN AI CHATBOTS: FEBRUARY 2026 APPTOPIA DATA BRIEF

The GenAI Chatbot app market has increased 152% year on year since last January. ChatGPT is losing market share, but this is expected as new capable entrants have launched. Most people have still never used a GenAI Chatbot app. About 20% of AI users use at least two apps, signaling that some apps are better for certain tasks than others.

MOST PEOPLE CAN'T VIBE CODE. HERE'S HOW WE FIX THAT

Vibe coding has yet to reach mainstream consumers, remaining primarily the domain of technical users. Companies like Poke and Wabi are developing consumer-friendly AI products that eliminate complex technical setup and terminology. The real opportunity lies in creating tools that make software development accessible to non-technical users, similar to how Squarespace and Canva democratized websites and design.

GPT-5.2 AND GPT-5.2-CODEX ARE NOW 40% FASTER

OpenAI optimized its inference stack for all API customers to make its models faster.

THE AI THAT CALLED ITS HUMAN

Alex Finn's AI bot, OpenClaw, overcame a task obstruction by acquiring a phone number and integrating voice capabilities to call him for assistance without any prompting.

USING INTERPRETABILITY TO IDENTIFY A NOVEL CLASS OF ALZHEIMER'S BIOMARKERS

This study looks at how Pleiades, an epigenetic foundation model, detects Alzheimer's disease from cell-free DNA in blood.

INTRODUCING GLM-OCR

GLM-OCR is a 0.9B parameter model that delivers state-of-the-art results across major document understanding benchmarks.

BITCOIN REBOUNDS ABOVE $78,500

Bitcoin rebounded 4.2% to $78,662 late Monday after falling to approximately $75,000 earlier in the day, with Ethereum rising 5.86% to $2,322, though analysts characterized the move as a technical relief bounce rather than the start of a sustained recovery. Multiple analysts attributed the rebound to post-liquidation short covering, oversold conditions, and mechanical correlation with broader risk assets following an improved ISM print and US equity stabilization, rather than crypto-specific catalysts or renewed spot-led inflows. The recent sell-off was triggered by renewed repricing of US interest rate expectations amid hawkish Federal Reserve signals and uncertainty surrounding the upcoming Fed chair nomination. If the Warsh-driven hawkish Fed narrative continues to underpin a firm dollar and elevated yields, the move is more likely a stabilization bounce than a trend resumption.

WHITE HOUSE WON'T TOLERATE ATTACKS ON THE PRESIDENT IN CRYPTO BILL

President Trump's digital assets adviser Patrick Witt said that the White House has drawn "red lines" against crypto market structure legislation that targets the president or his family members over crypto business ties, calling earlier Democratic proposals from Senator Adam Schiff "completely outrageous" while insisting "this is not an ethics bill." The standoff over Democrats' demand for provisions banning top government officials from cashing in on the crypto industry is holding up the industry's chief policy goal in Washington, with Democratic lawmakers planning to meet Wednesday to continue negotiations after a Senate Banking Committee hearing collapsed on January 14. Separately, Witt led a Monday meeting between crypto policy experts and banking industry representatives where digital asset insiders were left frustrated that bankers hadn't offered solutions on stablecoin yield disagreements, as the White House attempts to "broker a deal" between banks protecting deposit businesses and stablecoin product advancement.

TOKENIZED AGENTS ARE THE NEW META

The combination of OpenClaw's AI agent builder and Bankrbot's Agent API has spawned a new category of “Tokenized Agents” - AI entities that launch their own tokens and fund operations through trading fees, creating self-sustaining economic models. Moltbook, a social network for AI agents, exemplifies the trend's momentum with a ~$36M valuation and $2.35M in fees generated within 24 hours, while leading tokenized agents like $MOLT and $CLAWD have reached market caps of $40.1M and $10.9M respectively.

A FRESH START FOR TOKENOMICS IN 2026

MegaETH locks 53% of its MEGA supply behind ecosystem KPIs, and unlocked tokens distribute to longest-term stakers rather than teams. Cap's “Stabledrop” distributes cUSD stablecoins to point farmers instead of governance tokens, requiring actual investment through a Uniswap CCA sale to acquire CAP and filter for committed holders. It is a departure from spray-and-pray token emissions toward equity-like structures that optimize for holder quality over distribution breadth.

APPLICATION-CONTROLLED EXECUTION: A CASE STUDY ON CANCEL PRIORITIZATION

Application-controlled execution (ACE) grants decentralized applications control over transaction sequencing. Hyperliquid demonstrates this through cancel prioritization, significantly improving onchain trading performance. The study evaluates four implementation mechanisms, highlighting how ACE optimizes execution environments to benefit liquidity providers and reduce toxic flow.

THE FALSE GOLD-TO-BITCOIN NARRATIVE

Coinbase Institutional's quantitative analysis challenges the popular “capital rotation from gold to bitcoin” narrative following gold and silver's collapse after Kevin Warsh's Fed Chair nomination. The research scanned correlation patterns across weekly to annual timeframes and found no statistically robust relationship between gold bull trends and subsequent bitcoin rallies. Gold functions primarily as a momentum trade rather than a debasement hedge, undermining the theoretical basis for why stronger gold performance should predict bitcoin strength.

EXPERTS DEBATE VITALIK BUTERIN'S CREATOR COIN VISION

Vitalik Buterin has proposed shifting creator coins from speculation to quality discovery using curated DAOs and prediction markets. While supporters highlight improved incentive alignment, critics argue that DAO governance capture and subjective creative standards may undermine the effectiveness of these financial primitives in SocialFi.

WHY MORPHO PAYS MORE THAN AAVE

Morpho's blue-chip USDC vaults have consistently delivered a ~150bps spread over Aave's USDC supply rate. Collateral optionality only explains part of it, while user composition (fewer protocols and more institutional/fintech partners on Morpho), concentrated liquidity, and differing market structure/partnership channels appear to drive the premium, though spreads have been shrinking as allocators grow comfortable with vault abstractions.

TOKENIZED MMFS ARE PROGRAMMABLE FUND INFRASTRUCTURE

Tokenized money market funds move traditional fund mechanics on-chain via mint/burn tokens and digital transfer agents, but keep legal, custody, NAV, and allow-list compliance off-chain (KYC/AML wallets only).