Computerphile

Returning to the Markov Decision Process, this time with a solution. Nick Hawes of the ORI takes us through the algorithm, strap in for an epic episode!

Quicksort is a well known algorithm for sorting, Professor Graham Hutton shows how it works and then how to implement it in just five lines of code!

An AI model that changed the fortunes of silicon valley overnight. Deep Seek has been released open source, and requires far less hardware and investment. Mike Pound is based at the University of Nottingham.

Relatively speedy-to-access cache saves your computer having to trudge over to the RAM, but with multiple levels of cache memory, how does it all work?

The algorithm for differentiation relies on some pretty obscure mathematics, but it works! Mark Williams demonstrates Forward Mode Automatic Differentiation.

Discussing ideas of what happens after Generative AI plateaus, Dr Jakob Foerster is based at the University of Oxford.

Described as GenAIs greatest flaw, indirect prompt injection is a big problem, Mike Pound from University of Nottingham explains how it is like SQL Injection, except not...

Carbon is a brand new programming language still under development. It aims to dethrone C++ by being interoperable & compatible in both directions. Dr Guiffreida from the University of Nottingham gives us the lowdown on this ambitious new programming language.

GTC AI Conference is held each year by Nvidia and showcases AI and related technologies. This year Computerphile visited, here's a taste of what we saw. Full interviews coming soon, and yes, we did speak to him!

Nvidia CEO and co-founder Jensen Huang on various applications of GPUs and the rise of AI in all aspects of parallel processing.

As Large Language Models improve, the tokens they predict form ever more complicated and nuanced outcomes. Rob Miles and Ryan Greenblatt discuss "Alignment Faking" a paper Ryan's team created - ideas about which Rob made a series of videos on Computerphile in 2017.

Discussing how scammers take advantage of the way certain online systems work to leverage the delays. Dr Tim Muller is based at the University of Nottingham.

A seemingly simple problem that's "in general" incredibly difficult! CEO of Redwood Research Buck Shlegeris explains his favourite algorithmic fact!

What is Cuda and why do we need it? An Nvidia invention, Cuda is used in many aspects of parallel computing. We spoke to Stephen Jones, one of the architects of Cuda at the recent GTC conference.

Research suggests the rate at which AI is able to stay 'on task' is doubling every seven months. Is this the AI equivalent of Moore's Law? - Sydney Von Arx works on adversarial stress-testing at METR (Model Evaluation & Threat Research.)

Bashing out low-level code, it can be annoying to re-type the same commands over and over when you need to repeat a routine. Matt Godbolt explains how we can save frequently used code as a subroutine and reuse whenever its needed.

Mike Pound discusses hash collisions & the birthday paradox

The so-called 'Forbidden Technique' with Chana Messinger

Following the theme of AI research and safety, Aric Floyd talks about how some Large Language Models might follow the all too human trait of sandbagging - "lying" about their true capabilities.

Open source tools for creating videos? Lewis decides to help Mike take a break by creating 'MikeBot3000' - an AI generated Mike Pound...

Automating decision processes continued as Professort Nick Hawes of Oxford Robotics Institute explains how Monte Carlo Tree Search works.

Looking to the future, just how intelligent might the current crop of Large Language Models get? Daniel Kokotajlo joins us to discuss Ai2027.

Reinforcement Learning is how robots test the water in the real world.

Discussing World Foundation Models with TJ Galda from Nvidia.