Part 7 · 1 chapters · ~8 min

Mobile, Cloud and AI

The smartphone (iPhone 2007, Android 2008), app stores, mobile in Africa and M-Pesa, cloud computing from AWS (2006), open-source infrastructure (Linux, Kubernetes), the deep learning revolution (AlexNet 2012), transformers (2017), large language models and ChatGPT (2022), and what the history suggests about what comes next.

8

The present, as history

Every layer of a modern product has a history in this course: the transistor and the stored program in the chip, Unix and C in the operating system, TCP/IP and HTTP in the network, the relational model in the database, the smartphone in the client, the cloud in the server, and the transformer in the AI features. Knowing where they came from makes it easier to see which constraints are permanent (physics, the halting problem) and which are accidents of history (the 8-bit byte, QWERTY, the shape of URLs).

Brian Merchant, The One Device (2017); Ashish Vaswani and others, "Attention Is All You Need" (2017); Alex Krizhevsky, Ilya Sutskever and Geoffrey Hinton, "ImageNet Classification with Deep Convolutional Neural Networks" (2012); GSMA Mobile Economy reports.

MOBILE, CLOUD AND AI
the last two decades
the smartphoneiPhone (2007), Android (2008): thecomputer most people own.Africa goes mobileMobile subscriptions leapfroglandlines; M-Pesa (2007).the cloudAmazon S3 and EC2 (2006): rentcomputers by the hour.deep learningAlexNet (2012) wins ImageNet withGPUs.transformers"Attention Is All You Need"(2017).LLMsGPT-3 (2020), ChatGPT (Nov 2022),and the AI tools of today.
swipe the figure sideways, or tap expand for full screen
1/4
mobile
Smartphones made computing personal and constant; in Africa the phone became the main and often only computer, carrying banking (M-Pesa, mobile apps) and the internet.
the phone is the computerespecially in Africa