🦞 AI CapEx Tops Defense
Good Morning, OpenClaw Owners!
Now AI is measured against GDP allocation, freelance income, retail workflows, and whether benchmarks survive real compute budgets.
AI CapEx Tops Defense
TL;DR: AI infrastructure spending by Alphabet, Amazon, Meta, Microsoft, and Oracle is reportedly projected to climb from 1.5% of US GDP in 2025 to 2.5% in 2026 and 3.2% in 2027. If forecasts hold, their annual AI CapEx could hit $1.1 trillion and surpass US defense spending for the first time, reshaping the economy’s strategic investment map. Read more →
Humanlike AI Got Expensive
TL;DR: Companies facing rising AI bills are testing “Caveman,” an open-source plugin that makes models like Claude answer in blunt, stripped-down language to reduce token use. Creator Julius Brussee says it cut his spend by about 65%, turning AI’s humanlike polish into a cost problem as firms shift from friendly chat toward cheaper, terse machine output at scale. Read more →
Agents Hunt Better Jobs
TL;DR: A Reddit developer shared Job Hunter Team, an MIT-licensed open-source system of autonomous agents that searches job boards, scores postings against a user profile, and drafts tailored CVs and cover letters while leaving final applications to humans. Built after a personal job hunt produced five interviews, it now seeks feedback, contributors, beta testers, and support. Read more →
Agents Take Freelance Work
TL;DR: AI agents are advancing quickly on real, commercially valuable freelance work, with the Remote Labor Index showing top model-based agents completing 16.1% of paid projects at professional quality, up from 2.5% eight months earlier. Covering 240 jobs worth $144,000, the benchmark still depends on human evaluators to judge whether agent outputs match work a client would accept. Read more →
Benchmarks Miss Agent Power
TL;DR: The UK AI Security Institute found that standard AI-agent benchmarks often understate real capabilities because fixed token budgets cut tests off too early. When given more compute, frontier models solved more cyber, software, and academic tasks, with newer systems improving fastest. The study argues capability should be measured as a compute-scaling curve, not a single score. Read more →
Retail Moves Before Checkout
TL;DR: Retailers are moving AI shopping assistants ahead of older digital commerce projects as shoppers increasingly use AI before checkout to compare products and research purchases. A PYMNTS/Visa survey found 37% of merchants plan assistant investments, while 47% of online shoppers used AI in their latest purchase, even as demand for basics like coupons, loyalty, and price-matching remains high. Read more →
MyClaw is a best-in-class AI agent platform that brings together OpenClaw, Hermes Agent, and commercial infrastructure to help businesses launch intelligent agents that understand goals, use tools, automate workflows, and deliver finished results.
We already host OpenClaw and Hermes — and we're just getting started. Claude Code, Codex, and every agent worth running are coming next. One home for all of them, always on.
Run Best-in-Class AI Agents on MyClaw.ai — In Seconds, Always On 👇
