In partnership with
Introducing the first AI-native CRM
Connect your email, and you’ll instantly get a CRM with enriched customer insights and a platform that grows with your business.
With AI at the core, Attio lets you:
Prospect and route leads with research agents
Get real-time insights during customer calls
Build powerful automations for your complex workflows
Join industry leaders like Granola, Taskrabbit, Flatfile and more.
MORNING COFFEE
The Trillion-Dollar Ad Break
OpenAI announced on Friday that ChatGPT will begin testing advertisements in the United States. The figures were staggering—but it was the disconnect between ambition and monetisation that sent ripples through Silicon Valley.
The numbers: OpenAI has committed $1.4 trillion to AI infrastructure through 2030. The company expects to hit $20 billion in annualised revenue by end of 2025. It has 800 million weekly active users. And now, free and Go-tier users ($8/month) will see sponsored content at the bottom of their AI-generated answers.
Fidji Simo, OpenAI's CEO of Applications, emphasised that ads won't influence ChatGPT's answers
Three signals emerged that investors and builders are now parsing:
The Math —OpenAI needs revenue to justify $1.4 trillion in infrastructure spending. Current revenue, even at $20B, needs scale to meet the $60 billion annual Oracle contract alone. Advertising is a path towards driving that revenue
AGI Timeline —Superintelligence seems to be still a long way away even with all the advancements in models and capabilities
The Trust Paradox—Ad’s can create trust perception challenges and and it will be interesting to see how users embrace LLM ads
GROWTH HACK
The "LLM Citation" Engine
Turn your content into training data that AI models actually reference.
The Play: Structure your articles in clean, semantic Markdown that LLMs can easily parse and attribute. When AI models cite sources, they favor content with clear structure, explicit metadata, and machine-readable formatting over messy HTML or paywalled content.
Why This Works:
LLMs weight sources based on accessibility, structure, and authority signals embedded in the content itself. A well-formatted Markdown document with clear headers, semantic structure, and explicit attribution gets indexed and cited far more than equivalent content buried in JavaScript-heavy websites or PDF files.
The Implementation Stack:
Category | Tool | Notes |
|---|---|---|
Writing | Obsidian or VS Code | Native Markdown editing |
Publishing | GitHub Pages or Quartz | Free static hosting |
Metadata | YAML frontmatter | Schema.org compatible |
Syndication | RSS feed generator | Machine-readable distribution |
Verification | llms.txt standard | Explicit AI permissions |
Step 1: Structure Your Frontmatter
Every article needs machine-readable metadata:
---
title: "Your Article Title"
author: "Your Name"
date: 2026-01-19
description: "One-sentence summary"
tags: [ai, marketing, research]
schema: Article
canonical: https://yourdomain.com/article
---
Step 2: Use Semantic Headers
LLMs parse H2 and H3 tags to understand content hierarchy:
## Main Concept [Clear, searchable]
### Supporting Point [Specific, quotable]
Content that directly supports the header above.
Step 3: Include Explicit Citations
Make attribution easy for AI to replicate:
According to [Source Name](URL), "[exact quote]" (Date).
Step 4: Add an llms.txt File
The emerging standard for AI permissions:
# llms.txt
User-agent: *
Allow: /articles/
Preferred-citation: "Author Name, Site Name, 2026"
Step 5: Publish as Static Markdown
Host raw .md files alongside your HTML. GitHub Pages does this automatically. Tools like Quartz render beautiful sites while preserving original Markdown.
The Result:
Citation rate: 3–5x increase in AI-generated references
Discovery: Content surfaces in AI search responses
Authority: Builds compounding visibility as models retrain
Stop writing for Google. Start writing for Claude.
DAILY STAT
61% of Shoppers Prefer In-Store for Non-Essentials
The scale: Despite two decades of e-commerce growth, nearly two-thirds of consumers still choose physical stores when buying things they don't strictly need.
The human comparison: For every 10 people buying a new sweater, headphones, or home décor item, six of them are walking into a store—not clicking "add to cart."
The Shift:
The pandemic didn't permanently change behavior—it clarified preferences. E-commerce captured emergency buying (essentials, convenience). But when shoppers have time to browse, touch, and experience, they choose physical. Resonate CX's May 2025 research confirms what retail survivors already knew: digital is a channel, not a replacement.
The Economics:
E-commerce requires 15–20% of revenue for customer acquisition. Physical retail invests in real estate and staff but captures higher basket sizes and lower return rates. The blended model—research online, buy in store (ROPIS)—is winning.
What This Means for Builders:
• Omnichannel isn't a buzzword—it's table stakes. Your digital presence exists to drive physical traffic.
• "Showrooming" flipped. Customers now browse online but convert in person. Design for that journey.
• Experience beats convenience for discretionary spend. If you're selling non-essentials, invest in tactile, sensory touchpoints.
• Return rates for in-store purchases run 8–10% versus 20–30% online. The economics favor physical when customers can try before they buy.
The analog renaissance isn't nostalgia—it's optimization. In a world of infinite digital choice, physical presence becomes a signal of quality and trust.
TOOL TIP
Anara — The Research Paper Whisperer
URL: anara.com
What it does: Anara (formerly Unriddle) is an AI-powered research assistant that reads PDFs, videos, audio files, and even handwritten notes, then lets you interrogate them in natural language. Every answer includes clickable citations that jump to the exact source passage—solving the hallucination problem that plagues general-purpose AI tools in academic contexts.
Pricing:
Tier | Price | Limits | Use Case |
|---|---|---|---|
Free | $0/month | 15 AI questions/month, 10 uploads | Light coursework, trial |
Pro | $20/month | Unlimited questions, premium models | Active researchers |
Team | $30/seat/month | Shared workspaces, admin tools | Research groups |
Enterprise | Custom | SSO, dedicated support | Institutions |
Who it's for:
PhD students — Literature reviews in days, not weeks
Research analysts — Cross-reference multiple sources instantly
Medical professionals — Stay current on clinical literature
Legal teams — Parse case law and contracts with attribution
What makes it different:
Source highlighting — Every AI response links to exact passages, not just documents
Multimodal input — Upload lecture videos, voice memos, scanned handwriting
Collection chat — Query across multiple documents simultaneously
Graph view — Visualize connections between papers and concepts
Core capabilities:
PDF analysis: Upload and interrogate research papers
Video transcription: Index lecture content automatically
Citation generation: APA, MLA, Chicago formats built-in
Flashcard creation: Auto-generate study materials
Collaborative editing: Real-time manuscript co-authoring
Limitations:
Library quality depends on your uploads—garbage in, garbage out
10,000 page limit per file on Pro tier
No offline mode—requires internet connection
The bottom line: For serious researchers drowning in papers, Anara's $20/month Pro plan pays for itself in the first literature review. The citation-linked responses alone make it worth the switch from ChatGPT for academic work.
TICKER WATCH
T1 Energy (NYSE: TE) — $8.17
The Numbers That Matter:
Metric | Value |
|---|---|
Current Price | $8.17 |
52-Week Low | $0.92 |
Market Cap | $2.0B |
Analyst Target | $10–$15 |
Revenue Growth | 62%/year projected |
What They Do (Simple Version):
T1 makes solar panels in Texas. The U.S. government pays them cash for every panel they manufacture—that's the Section 45X tax credit. They just sold $160 million worth of these credits at 91 cents on the dollar. Free money for building stuff in America.
Why This Matters:
The stock was $0.92 a year ago. It's now $8.17. That's an 800%+ move—and the story isn't over.
Three things changed: (1) They bought a working factory, (2) they proved they can turn tax credits into cash, and (3) they're building a second factory that doubles their capacity.
The Upside Case:
Conservative: Analysts have a $10.17 average target = +24% from here
Bull case: Roth Capital raised target to $15 in December = +84% from here
Profitability: Expected in 2027 when both factories are running
Simple Math: $1,000 invested today could become $1,240 (conservative) or $1,840 (bull case) within 12 months.
The Math:
Every solar panel T1 makes generates ~$0.07/watt in tax credits. Their Dallas factory makes 5GW/year. That's $350M in annual credits they can sell for cash. Austin factory (under construction) adds another 5GW. Double the credits, double the cash.
The Risks (Be Honest):
Stock already ran 800%—could pull back
Second factory must get built on time
Government policy could change
They're not profitable yet
The Verdict:
This is a policy arbitrage play with real cash flows. The government is literally paying them to manufacture solar panels in America. If the Austin factory gets built and tax credits stay intact, there's another 50–80% upside. If policy changes or construction delays hit, you could give back gains fast.
Position: Speculative. Small position only (1–2% max). High risk, high reward.
Not financial advice. Do your own research.
WORKFLOW
The "Original Thinker" Curation Engine
Setup time: 45 minutes | Weekly value: 10+ hours saved filtering signal from noise
Automatically curate insights from human experts, filter out AI-generated content, distill into clean Markdown, and publish as both a newsletter and public documentation that LLMs can cite.
The Architecture:
Trigger: Daily at 6 AM
↓
Action 1: RSS Fetch — Pull from curated expert feeds
↓
Action 2: AI Slop Filter — Score content for human authorship signals
↓
Action 3: Insight Extraction — Distill key ideas with citations
↓
Action 4: Markdown Generation — Structure for human + machine readability
↓
Action 5: Dual Publish (if quality score > 80)
↓
Outcome: Newsletter + public docs updated
Step 1: n8n (Feed Aggregation)
Pull from RSS feeds of verified human writers:
{
"feeds": [
"https://stratechery.com/feed/",
"https://www.lennysnewsletter.com/feed",
"https://eugeneyan.com/feed.xml"
],
"max_items_per_feed": 5,
"since": "24h"
}
Signal: Bypasses algorithmic feeds—only sources you've vetted.
Step 2: Claude API (Slop Detection)
Score content for AI-generation signals:
const prompt = `Analyze this article for human authorship signals:
- Unique insights not in common training data
- Personal anecdotes or first-person experience
- Contrarian takes that require independent thinking
- Specific examples from recent events
Return: {score: 0-100, signals: [], red_flags: []}`;
Signal: Score below 60 likely indicates AI-generated or heavily derivative content.
Step 3: Anthropic API (Insight Extraction)
Distill to quotable key points:
{
"format": "markdown",
"include_source_url": true,
"max_insights": 3,
"require_attribution": true
}
Step 4: Dual Publishing
Push to both newsletter platform and public GitHub docs:
# Newsletter via ConvertKit API
curl -X POST https://api.convertkit.com/v3/broadcasts
# Public docs via GitHub
git add docs/insights/$(date +%Y-%m-%d).md
git commit -m "Daily insights $(date)"
git push origin main
Expansion Ideas:
Add Substack comment scraping for emerging voices
Build "original thinker" score leaderboard
Auto-generate podcast scripts from curated insights
Create RAG database for future LLM training on quality content
The future of curation isn't finding more content—it's filtering for what's actually worth reading.
THE BOTTOM LINE
OpenAI's advertising pivot reveals a fundamental truth: even $1.4 trillion in committed infrastructure spending doesn't guarantee the economics work. When the company that promised AGI starts selling ads, it's time to recalibrate expectations about who actually captures value in the AI revolution.
The through-line this week is trust and attribution. Anara wins because it solves AI's credibility problem—citations you can verify. T1 Energy wins because tax credits are a verifiable cash flow, not a speculative bet on future margins. The "LLM Citation" growth hack wins because it builds attribution into content itself. And 61% of shoppers choosing physical stores? That's trust expressed through presence.
The playbook: Build systems that verify. Create content that machines can cite. Invest in companies with provable economics, not narrative momentum. And remember that the offline world hasn't disappeared—it's simply waiting for people tired of algorithmic noise.
If you're building, structure your content for both humans and LLMs—the citation game is just beginning. If you're investing, look for tax credit arbitrage and proven monetization over ambitious infrastructure promises. If you're selling, the physical experience still matters more than you think.
Ship daily.
HackrLife Daily is read by growth marketers at Google, Adobe, LinkedIn, and creators building the future.