2023 Tech Predictions Scorecard: GPT-4, Threads, and Market Calls Reviewed
Analyzing my 2023 predictions: GPT-4 ✓, Meta Threads ✓, China rally ✓. Lessons on compressed market cycles and why direction beats timing in tech forecasting.

Yesterday I was deep in digital archaeology (sifting through dusty Looms and half-built frameworks) when a forgotten Notion table surfaced: "Zak's Public Predictions for 2023."
The universe has jokes. I'd just finished a meta-analysis of prior meta-analyses (recursion, always), and this card appeared like it wanted closure.
🧠 2023 Scorecard
- Meta's Twitter clone? → Threads launched July 5. Direction right, clock off.
- GPT-4 release? → March 14. Direct hit.
- China reopening & rally? → Borders opened Jan 8. HK/ADR basket +45% in Q1. Nailed it.
- Roaring 20s market run? → S&P +24%. Broadly on target.
⚙️ Lessons in a Compressed-Cycle World
1️⃣ Time is the leverage, not the lag
Pandemic → stimulus → AI boom → tariff backlash—all inside 36 months. Precision timing > raw foresight. Knowing where to lean is table-stakes; knowing when to press is alpha.
2️⃣ Prediction ≠ victory
Getting GPT-4 right was nice. But spotting it would kick off an 18-month GPU land grab—that was the unlock. Same way I pitched "AI-first music generation" in 2022 and got polite eye-rolls. Now? Term sheets.
3️⃣ Feedback loops are now AI-accelerated
In 2022, GPT-4 was vaporware. Today I handed the dusty Notion card to an LLM that scrapes, scores, and contextualizes bets in minutes. The intuition audit stack is here. Use it—or drift.
🎰 Working Framework
- Vector > spot-price. Go for directional truths, not headlines.
- Timestamp everything. Memory fades. Logs compound.
- Prediction-autopsies. Turn hindsight into upstream tuning.
- Translate insight to optionality. Own assets or skills that benefit from change.
- Stay constructively contrarian. Early looks wrong—until it pays.
Threads wasn't a miss. It was an early call.
In a world of compressed cycles and recursive feedback, direction matters more than perfect timing.
I'm parking the 2023 card in a public repo this time. Let's see what finds me in 2027. 📊
What predictions are you sitting on that deserve a timestamp? The future has a funny way of circling back.
What’s next
A few handpicked reads to continue the thread.
The Cold Swarm: Why Your AI-Personalized Pitch Is Dead on Arrival
11 min readAI-generated cold emails are flooding inboxes at unprecedented scale. Here is why the uncanny valley of artificial personalization is destroying professional communication—and the two survival strategies that still work.
The 10x PM Paradox: Why Organization Beats Genius Every Time
8 min readSilicon Valley worships the 10x engineer, but the real productivity multiplier is the PM who breaks work into atomic tasks. Here is why systematic organization beats raw intelligence—especially in the AI era.
Meta-Learning Eats Itself: When AI Tools Train on Their Own Usage
9 min readClaude Code learning from every "You're absolutely right!" moment could push dev tools from 1.2-5x productivity gains to genuine 10x++ multipliers. Recursive self-improvement through usage patterns changes everything.
About the Author

Engineer · systems gardener · philosopher-scientist · Between Curiosity, Code & Consciousness