Category: Systems
-
The Stargate Project: a $500 billion bet on AI infrastructure (2025)
In January 2025, OpenAI, SoftBank, Oracle and MGX announced Stargate, a plan to invest up to $500 billion in US AI data centers, just as DeepSeek was questioning the cost of compute.
-
DeepSeek-R1 and the efficiency shock (2025)
A Chinese startup released an open-weights reasoning model that rivaled the best at a fraction of the reported cost, and on January 27, 2025 it wiped a record sum off Nvidia's value.
-
Gemini 2.0 and the start of the agentic era (2024)
Google's Gemini 2.0, launched in December 2024, was pitched as a model for the agentic era: systems that do not just answer, but take multi-step actions on your behalf.
-
Gemini 1.0: Google's multimodal answer (2023)
Google rebranded its AI efforts under Gemini, a natively multimodal family meant to rival GPT-4, and learned how closely launch demos would now be scrutinised.
-
Llama 2: open(ish) weights change the game (2023)
Meta released a capable language model whose weights anyone could download and build on, pushing the debate over 'open' AI to the center of the industry.
-
GPT-4: the flagship that raised the bar (2023)
OpenAI's GPT-4 could read images as well as text and scored well on exams built for humans, cementing a fierce commercial race, and a secrecy that unsettled researchers.
-
ChatGPT: the week AI went mainstream (2022)
A free chat demo built on a fine-tuned language model reached a hundred million users in about two months and turned AI from a research topic into a public obsession.
-
AlphaGo vs. Lee Sedol: the machine that learned to play (2016)
DeepMind's AlphaGo beat one of the greatest Go players 4–1, winning not by brute-force search like Deep Blue but by learning intuition from data and self-play.
-
Deep Blue vs. Kasparov: the match that shook the world (1997)
IBM's chess computer defeated the world champion in six games, a media spectacle that proved machines could outplay humans, and sparked a debate that never quite ended.
-
Expert systems: when AI became a business (1980s)
Rule-based expert systems promised to bottle human expertise into software, they fuelled a billion-dollar boom, then collapsed into the second AI winter.
-
ELIZA: the chatbot that was never meant to be one (1966)
Joseph Weizenbaum's ELIZA used simple pattern-matching to simulate therapy, and accidentally proved that humans project understanding onto machines.