AI Models and the AI Industry
Frontier models, the chips and data centers built to train them, the jobs and markets they're reshaping, and the agents, ethics debates and great-power rivalry that surround all of it — the single fastest-moving topic this site covers. Verified specifications, official disclosures and editorial analysis are kept in clearly separated categories, never blended into one narrative.
Live Status
Which race matters more right now? The model race between GPT-5, Gemini, Kimi K3 and Qwen3.8-Max — or the US-China infrastructure race underneath it, where chips and export controls decide who can even compete. See the full comparison below.
AI Models and the AI Industry at a Glance
Frontier Models, Industry & Jobs, Infrastructure & Chips, Agents/Geopolitics/Society.
💸~$500B in AI-related debt and record capex — flagged as elevated risk, not a verified crash.
🏆GPT-5, Gemini, Kimi K3 and Qwen3.8-Max each lead different benchmarks.
🌐Export controls and semiconductor access increasingly decide who can compete at all.
👷We fact-check specific layoffs rather than citing headline job-loss totals.
🦾Humanoid robots and autonomous agents are the next frontier past chat-based models.
How to Read This Cluster
Four threads, kept deliberately separate so verified specifications, official disclosures and editorial analysis never blend into one narrative.
GPT-5, Kimi K3, Qwen3.8-Max, Apple's AI position and the evolution from chatbots to coding agents.
The AI boom, bubble-risk debate, SaaS disruption, Oracle's layoffs and the wider jobs-revolution record.
Nvidia, HBM memory, ASML lithography, Asia's manufacturing boom and the UAE's data-center build-out.
Agent history and ethics, the US-China AI cold war, finance agents, AI cyberwar and humanoid robots.
The Timelines
This is the living index to everything we publish on AI models and the AI industry — 28 timelines across four threads. Use the filters to narrow to what you need.
GPT-5 History & OpenAI Evolution
From the company founding and GPT-1 through GPT-4o and the reasoning o-series to GPT-5 and the GPT-5.6 family.
Kimi K3: Moonshot AI's 2.8T-Parameter Model
How Moonshot AI went from Kimi Chat to a 2.8-trillion-parameter open-weight frontier model with a 1M-token context.
Qwen3.8-Max: Alibaba's Open-Weight Flagship
Architecture, benchmarks, licensing status and deployment paths for Alibaba's Qwen3.8-Max.
Did Apple Lose the AI Race?
From Siri and the Neural Engine to Apple Intelligence, Private Cloud Compute and the rebuilt Siri at WWDC 2026.
The Generative AI Evolution: ChatGPT to Coding Agents
How generative AI moved from ChatGPT's 2022 debut through reasoning models to autonomous coding agents.
The AI Boom: Markets, Technology and the Global Economy
From the 1950 Turing Test through the 2017 Transformer to the 2025-2026 AI infrastructure race, with OECD, IMF and Stanford data.
Is the AI Boom Entering Its Reality Check?
Sequoia's $600B question, nearly $500B in AI-related debt, record hyperscaler capex and South Korea's leveraged chip sell-off.
The Death of the Software Subscription?
Gartner puts $234 billion of SaaS spending in the blast radius through 2030 — but calls mass application replacement unlikely.
Oracle Layoffs: AI, Cloud Restructuring & the Tech Workforce
Oracle's workforce fell by ~21,000 (13%) in FY2026 while CapEx surged 162% to $55.7B and cloud revenue grew 39%.
AI Jobs Revolution: Layoffs, Automation & Careers
A company-by-company AI-layoff fact-check, a dedicated India IT/GCC section, and a job-exposure framework, built on ILO, IMF and NASSCOM data.
AI Investment Boom: Where the Trillions Are Going
Corporate AI investment more than doubled to US$581.7 billion in 2025, and Big Tech plans US$720–745 billion of capex for 2026. Where the money goes.
Global AI Chip Race: Nvidia, HBM & the Supply Crunch
Nvidia GPUs, HBM memory, TSMC foundries, CoWoS packaging, ASML lithography and whether supply can keep up with AI demand.
ASML: How One Dutch Company Became Essential to AI Chips
ASML's timeline from its 1984 founding to EUV lithography, High-NA machines and the global semiconductor race.
Asia's AI Manufacturing Boom: Chips, HBM & Data Centres
Taiwan makes the chips, Korea supplies HBM, Japan ships the equipment, and China, India and Malaysia race to build their own supply chain.
UAE AI Data Center Timeline: The 5-Gigawatt Plan
How the UAE went from a 2017 AI strategy to a 5-gigawatt AI campus with G42, OpenAI and Microsoft.
NVIDIA History: From Gaming Graphics to AI Dominance
How a 1993 graphics start-up became the engine of AI: GeForce 256, CUDA, AlexNet, Tensor Cores, Hopper, Blackwell and Vera Rubin.
AMD's Comeback: From a $660 Million Loss to Nvidia's Challenger
From sub-$2 shares in 2015 to $34.6 billion revenue in 2025: Lisa Su, Zen, chiplets, Instinct GPUs and the OpenAI, Meta and Anthropic deals.
Meta's AI Chips: Every MTIA Generation
Meta's in-house AI chips from the first MTIA design in 2020 to MTIA 500, the scrapped Olympus training chip and the Nvidia and AMD orders alongside.
U.S. AI Data Centers and the Electric Grid
US data centers used 4.7% of national electricity in 2024. What the 2030 forecasts actually say, what is ghost demand, and who pays for the grid.
Finland AI Power: Nuclear, Cold Climate and Data Centers
Google's €13 billion Finland investment and 22-year Loviisa nuclear deal: what is confirmed, what is only announced, and the political fight over power.
AI Agents History: Autonomous AI & Multi-Agent Systems
From Alan Turing and early intelligent agents to LLM agents, AutoGPT, reasoning models and enterprise agentic AI.
AI Development & Ethics Timeline
The OECD AI Principles, the UNESCO Recommendation, the NIST AI RMF, ISO/IEC 42001 and the EU AI Act, tracked alongside the science.
US–China AI Cold War: Chips, Models & Data Centers
Nvidia and Huawei chips, semiconductor manufacturing, export controls and alliances like Pax Silica through 2030.
Agentic AI in Finance: Wall Street's Shift to AI Agents
Goldman Sachs' Claude-based agents, Anthropic's finance templates, and FINRA, ESMA, Bank of England and FSB oversight activity.
AI Cyberwar: How AI Is Transforming Hacking & Defense
Hackers use AI for faster attacks and supply-chain compromises, while defenders deploy AI agents for detection and response.
Saudi Arabia vs UAE: The Gulf AI Race
HUMAIN vs G42, Stargate UAE vs HUMAIN's 6.6 GW target, chip-access status and Falcon/Jais vs ALLaM, each labelled by status.
AI UPI Payments: How Agentic Payments Could Work in India
How India is preparing UPI for AI agents that pay on your behalf: UPI Circle, Reserve Pay, spending limits and the security questions.
Humanoid Robot Timeline: Optimus, Figure & Embodied AI
Tiangong Ultra's 100m sprint, Figure at BMW, Boston Dynamics Atlas, Tesla Optimus and the race for embodied AI.
US AI vs China AI: Which Race Matters More?
Two overlapping races, both live on this site. Closed frontier models versus open-weight challengers, and the chips that decide who can even compete.
US AI: OpenAI, Google, Anthropic
GPT-5 and the GPT-5.6 family lead on most benchmarks, built on Nvidia's chip advantage and hyperscaler capex now running into hundreds of billions of dollars a year.
Open the GPT-5 history timeline →China AI: Chips, Models, Export Controls
Qwen3.8-Max and Kimi K3 close the gap on open-weight benchmarks even as export controls restrict chip access — a rivalry playing out in silicon as much as in model cards.
Open the US-China AI cold war timeline →What We Are Watching
Whether AI-related debt and hyperscaler capex trigger a real correction or stay a contained risk, which lab actually holds the frontier-model lead as GPT-5.6, Gemini and Kimi K3 iterate against each other, whether chip export controls widen or narrow the US-China gap, and whether humanoid robots and autonomous agents move from demos to real deployed autonomy.