Скачать или смотреть Memory of the Machine: A Database History of AI (Ep. 540)

How does a machine remember?

From ELIZA’s canned responses in the 1960s to the hybrid vector‑and‑SQL stacks of 2026, this episode traces the entire evolution of AI memory — and shows how every generation of AI has been limited (or empowered) by the way it stores information.

In Memory of the Machine, we explore the hidden database history behind artificial intelligence:
• Why early chatbots had no memory at all
• How LISP and semantic networks tried to give AI a structured mind
• Why 1980s expert systems collapsed under their own IF–THEN rule bases
• How SQL and relational databases fueled the rise of big‑data AI
• Why vector databases became essential for meaning and similarity
• How modern AI uses RAG to combine facts + context
• And why the future of AI depends on building machines that can truly remember

This is Episode 540 of Database History — a synth‑wave documentary series about the engines that quietly run the world.

Chapters:
00:00 Intro
00:28 ELIZA and the Canned Response Era
01:40 LISP, Logic, and the First Knowledge Graphs
03:05 Expert Systems and the IF–THEN Boom
04:30 SQL, Metadata, and Deep Blue’s Big Data Moment
05:55 The Vector Revolution (2020–2024)
07:20 The 2026 Hybrid Stack (RAG + SQL + Vectors)
08:40 The Future of AI Memory
09:10 Outro

If you enjoy deep dives into the hidden infrastructure of technology, hit subscribe.

And tell me in the comments: What’s your favorite database rebrand in AI history?