An Engineer's Perspective on AI: Slow Thinking, Faster Fun — Jeff Turner
An evening with Jeff Turner — former computer science professor, now building at the intersection of causal AI and decentralised systems.
Most talks about AI either assume you already know how it works or assume you never will. This one is for the room in between: engineers who want the architecture, and curious non-experts who want to know why the thing on their phone is brilliant one minute and confidently wrong the next.
What we'll get into
How large language models actually work under the hood — and why that architecture produces both the surprising power and the stubborn limitations.
Fast thinking and slow thinking. Human brains move between quick intuition and deliberate reasoning. Current AI is stuck in a very different mode, and that gap explains a lot.
The four problems that shape every serious use of AI today. Brittleness, where small changes in input produce large unpredictable failures. Weak causal reasoning, where a model sounds insightful while missing cause and effect entirely. Hallucination, the confident generation of plausible falsehoods. And the jagged edge of capability — superhuman in one moment, startlingly incompetent the next.
What comes after the Turing test. Robot companions and AI partners are already forcing harder questions about intelligence, relationship, and what we actually want from machines.
Bring your questions. This is a room that tends to want the how-it-actually-works version, and rarely gets it.