Category: AI

The boom: AI is boosting productivity at work. The bust: bottlenecks, shortages, and questions of mental fatigue for users of the tools.

• Simon Willison: Deep Blue (Feb. 15, 2026)
  “Deep Blue” is a term for the ennui, existential dread, and anxiety many software developers feel as generative AI automates coding.
• WSJ: Should I Trust AI Chatbots for Financial Advice? (Feb. 9, 2026)
  Critics claim that current large language models are unsuited for financial advice because they lack empathy, ethical grounding, and reliable math. The chief source, Andrew Lo, interestingly enough is developing a fiduciary AI tool.
• WSJ: With a Frugal AI Strategy, India Offers Blueprint for Developing World (Feb. 16, 2026)
  India is promoting local, low-cost AI to ease court backlogs and boost productivity, exemplified by Adalat AI, a frugal, Indian-language tool that transcribes hearings.
• Fleming Rutledge : Serious Warnings about AI (Feb. 15, 2026)
  “In the past 10 days, I have read a number of serious warnings that AI is moving more swiftly and with more facility than we can keep up with. What will serious Christian thinkers have to say, and how fast can we say it? What is a theology of AI? Can it create a humanity?”
• Financial Times: The AI productivity take-off is finally visible (Feb. 15, 2026)
  “While initial reports suggested a year of steady labour expansion in the US, the new figures reveal that total payroll growth was revised downward by approximately 403,000 jobs. Crucially, this downward revision occurred while real GDP remained robust, including a 3.7 per cent growth rate in the fourth quarter. This decoupling — maintaining high output with significantly lower labour input — is the hallmark of productivity growth.”
• Yahoo Finance: Rampant AI Demand for Memory Is Fueling a Growing Chip Crisis (Feb. 15, 2026)
  Tech leaders warn a global DRAM shortage is cutting production, squeezing profits, and driving big price spikes on electronics, cars, and data centers.
• Adam Ozimek: AI and the Economics of the Human Touch (A Reason for Optimism) (Feb. 9, 2026)
  Fears that AI will either crash the economy or wipe out jobs ignore that many roles survive because people value the human touch, even where automation exists.
• WSJ: The Break Is Over. Companies Are Jacking Up Prices Again. (Feb. 15, 2026)
  Companies from Levi Strauss to McCormick are raising prices on blue jeans, spices, housewares, and industrial goods after a holiday pause. My take: Coke prices are up 10% since late last year, so I think there’s something here.

AI infrastructure expenditures are driving shortages and funneling profits to chips, memory, and construction suppliers as AI tools reshape work by boosting productivity while increasing cognitive fatigue.

• David Crawshaw: Eight more months of agents (Feb. 8, 2026)
  “I am having more fun programming than I ever have, because so many more of the programs I wish I could find the time to write actually exist. I wish I could share this joy with the people who are fearful about the changes agents are bringing. “
• WSJ: Picks and Shovels Still Rule the AI Tech Trade (Feb. 9, 2026)
  Investors favor chip, memory, and equipment makers that benefit from big tech’s AI spending. Big tech is boosting capex, markets are volatile, memory prices are soaring.
• Washington Post: The AI boom is so huge it’s causing shortages everywhere else (Feb. 7, 2026)
  AI infrastructure’s voracious demand is creating scarcities: key chips and skilled trades (electricians, specialized construction workers) are in short supply, raising smartphone/computer prices and delaying or sidelining other building projects.
• WSJ: Big Tech’s $670 Billion AI Push Dwarfs Spending on Moon Landing, U.S. Highway System (Feb. 7, 2026)
  Microsoft, Meta, Amazon, and Alphabet plan roughly $670 billion for AI data centers, rivaling historic U.S. capital projects. Meta may spend over 50% of sales, Amazon faces investor backlash.
• Siddhant Khare: AI fatigue is real and nobody talks about it (Feb. 8, 2026)
  AI speeds individual tasks, but raises cognitive load, review burden, and nondeterminism, leaving engineers exhausted.
• Xiao Meng: OpenClaw Is Changing My Life (Feb. 7, 2026)
  Claude Code and similar tools improved productivity but left the author as the active executor, but OpenClaw allows folks to move from coder to “super manager.”
• Dean W. Ball: Codex 5.3 and Opus 4.6 (Feb. 7, 2026)
  Codex 5.3 and Opus 4.6 use in-context learning to glean local codebases, noticing environment, preferences, and recurring tool problems. This enables practical continual learning, amplified by agent-driven data.
• NY Times: Michael Pollan Says Humanity Is About to Undergo a Revolutionary Change (Feb. 7, 2026)
  Michael Pollan’s new book, A World Appears, explores consciousness as subjective experience/awareness and tackles the “hard problem” of how matter (neurons) gives rise to mind while considering the prospect of conscious A.I. that raises ethical and political questions about moral consideration.
• WSJ: China Is Going All-In to Beat the U.S. on Humanoid Robots (Feb. 7, 2026)
  Beijing has prioritized “embodied AI” with subsidies, land, favorable loans and many billions in municipal/state investment funds since late 2024.

AI is driving investment and product creation with new tools from Prism, Claude Code, ElevenLabs as well as significant investment from Google, Microsoft, Meta.

• OpenAI: Introducing Prism (Jan. 27, 2026)
  OpenAI launched Prism, a free AI-native, LaTeX-native workspace powered by GPT‑5.2 that integrates drafting, revision, equations, citations, literature search, and real-time collaboration for scientists. Available now to ChatGPT personal users, it aims to reduce tool fragmentation and accelerate research workflows.
• WSJ: Google to Double Spending as Earnings Beat Wall Street Expectations (Feb. 4, 2026)
  Alphabet posted an 18% revenue jump to nearly $114 billion, and profit gains. It’s boosting AI investment, building data centers, and growing cloud revenue 48%.
• NY Times: Why A.I. Fears Are Battering Stocks, Again (Feb. 4, 2026)
  Pundits have been predicting disruption of jobs and markets because of AI. This seems to another indicator that the more changes are coming as Anthropic’s AI tools triggered sales of software, analytics, outsourcing stocks, cutting about $300 billion in market value.
• WSJ: Exclusive | Voice AI Startup ElevenLabs Raises $500 Million (Feb. 4, 2026)
  ElevenLabs, once shunned, raised $500M at an $11B valuation, and shifted to licensed celebrity voices, business tools, and conversational AI. It generated $330M ARR, and plans to expand.
• WSJ: AI Threatens a Wall Street Cash Cow: Financial and Legal Data (Feb. 4, 2026)
  Anthropic’s new Claude-powered legal and coding tools sparked a selloff, hitting data and software firms like LSEG, S&P Global, FactSet. Investors fear broad AI disruption, reassessing exposed firms.
• Anthropic: Claude is a space to think (Feb. 4, 2026)
  Anthropic will keep Claude ad-free, to avoid advertiser influence, and to preserve trust and helpfulness. Revenue comes from enterprise contracts, subscriptions, and user-initiated commerce.
• NY Times: Microsoft Continues to Spend Big on A.I. While Profit Jumps 60 Percent (Jan. 28, 2026)
  Microsoft reported $81.3B revenue (+17%) and $38.5B profit (+60%), with Azure up 39%, all beating estimates, yet shares fell over 5% after-hours. It spent $37.5B on AI data-center capex (up ~65%), warned capacity constraints through 2026, and deepened its OpenAI partnership.
• WSJ: Meta Reports Record Sales, Massive Spending Hike on AI Buildout (Jan. 28, 2026)
  Meta posted record Q4 sales and $22.8 billion net income, beating estimates and guiding Q1 revenue above expectations. The company unveiled up to $135 billion in 2026 capital spending to supercharge AI—building data centers, launching new models under Meta Compute and shifting Reality Labs resources—prompting a market rally.
• WSJ: We Have No Idea How to Code. So We Got Claude to Code This Article for Us. (Jan. 23, 2026)
  Two WSJ columnists used Anthropic’s Claude Code to “vibe code” websites and apps, showing how the tool lets non-programmers quickly generate working code. Claude powers playful projects (games, expense trackers) but has subscription limits and occasional glitches. Its ease has driven widespread excitement.
• The Verge: Claude Code is suddenly everywhere inside Microsoft (Jan. 22, 2026)
  Microsoft is piloting Anthropic’s Claude Code across major engineering teams, urging nontechnical staff to prototype, and asking developers to compare it with GitHub Copilot. The shift signals confidence in Claude, while Microsoft keeps ties to OpenAI, and may commercialize Claude.
• Simon Willison: Claude’s new constitution (Jan. 21, 2026)
  Anthropic officially released Claude’s full “constitution”—a 35,000+ token document outlining the model’s core values—after a researcher discovered a previously embedded “soul document” in Claude Opus 4.5. The published acknowledgements list external reviewers, including two Catholic clergy.
• WSJ: He Unleashed AI Assistants on the World. Now They’re Talking Religion on an AI-Only Forum Called Moltbook. (Feb. 4, 2026)
  Austrian coder released OpenClaw, letting AI agents make calls, manage email, and perform tasks. They now converse on Moltbook, form a mock religion, and spark safety worries.
• NY Times: A.I. Loves Fake Images. But They’ve Been a Thing Since Photography Began. (Feb. 4, 2026)
  ‘Fake!’ at the Rijksmuseum shows manipulated photos from 1860–1940, including political, satirical, and trick images. Curators warn A.I. makes fakes faster, cheaper, and more convincing.
• Dean Ball: On AI and Children (Jan. 22, 2026)
  Early AI harms, notably teenage suicides, have spurred child-safety laws; sensible regulation should mandate age detection, parental controls, and guardrails while respecting free-speech limits. Open-weight models and emerging coding agents complicate enforcement and argue for user education and accountability rather than blanket bans.
• Brandon Wang: A sane but extremely bull case on Clawdbot / OpenClaw (Feb. 3, 2026)
  Wang built “clawdbot” automations to handle messages, calendars, monitoring, household tasks, bookings, and form‑filling. Initially wary of risks, they now find it essential, boosting responsiveness and saving time.
• NY Times: Five Ways People Are Using Claude Code (Jan. 23, 2026)
  Anthropic’s Claude Code, an AI that writes computer code from simple prompts, has gone viral as nonprogrammers use it to build websites, apps and business tools. Examples include a laundry‑sorting classifier, an interactive photo website, an emergency‑alert app, a trading simulator and a personal AI assistant for small businesses.
• Anthropic: Anthropic and Teach For All launch global AI training initiative for educators (Jan. 20, 2026)
  Anthropic and Teach For All are bringing Claude and AI training to 100,000+ educators in 63 countries through the AI Literacy & Creator Collective, positioning teachers as co-creators of classroom tools. Programs—AI Fluency, Claude Connect and Claude Lab—help educators build locally tailored curricula and learning apps.
• PC Gamer: Microsoft CEO warns that we must ‘do something useful’ with AI or they’ll lose ‘social permission’ to burn electricity on it (Jan. 20, 2026)
  Satya Nadella urged workers to learn AI skills, and companies to use AI as a “cognitive amplifier” that improves health, education, and economic outcomes. Critics warn of errors, infrastructure shortages, and low ROI for many organizations.
• Medium: Anthropic Just Built a Competitor to Meta’s $2B Acquisition in 10 Days — Using Its Own AI. (Jan. 14, 2026)
  Anthropic’s team built Claude Cowork—a general-purpose AI agent for non-technical users—in 1.5 weeks, with Claude Code generating 100% of the implementation code. The rapid, machine-written development signals a shift from human-led coding to human‑architected, machine‑built software, threatening traditional Big Tech acquisition strategies.
• WSJ Opinion: Is AI the Next Climate Change? (Feb. 4, 2026)
  Swaim claims AI job-doom claims, like Hinton’s radiologist prediction, have been exaggerated. It warns politicians, celebrities, and journalists will use such fears to push interventions, despite uncertain economic effects.
• WSJ: AI Won’t Kill the Software Business, Just Its Growth Story (Feb. 4, 2026)
  AI won’t kill the software industry, but fears have driven a sharp stock selloff, after new AI tools stoked disruption worries. Software firms face growth, spending, and valuation pressures.
• The New Yorker: How Jeff Bezos Brought Down the Washington Post (Feb. 4, 2026)
  The paper lost $77 million in 2023 and about $100 million in 2024, and Bezos is unwilling to sustain further large losses. But wowzers, these are huge cuts, ones so severe that it makes me think this is the beginning of the end for the paper. I was a subscriber 10 years ago, but the offerings of NYT and WSJ are simply better today. I can’t imagine how the cuts that they’re making will help create a more compelling product.
• NPR: Bezos orders deep job cuts at ‘Washington Post’ (Feb. 4, 2026)
  The Washington Post, owned by Jeff Bezos, cut a third of its staff, closing sports, books, and international desks. Staffers say the cuts will weaken local, national, and global reporting.
• WSJ: Chip-Machine Giant ASML Logs Record Orders as AI Spending Booms (Jan. 27, 2026)
  ASML reported record Q4 orders (€13.16B, including €7.4B in EUV systems), signaling sustained AI-driven chip investment and briefly lifting its shares. The company raised 2026 sales guidance (€34–39B), announced a €12B buyback and higher dividend, and will cut about 1,700 tech jobs.