In a recent post on LessWrong, Andrew Critch imagines scenarios in which networks of seemingly well-behaved AI systems nonetheless results in humanity’s demise. In it, he mentions: … [both stories] follow a progression from less automation to more, and correspondingly from more human control to less … The comment is only made in passing, but it seems (a) interesting and (b) important: is this…
Keep Reading →
Latest posts
Does more automation mean less control?
July 02, 2021
Synaesthetic music visualisation with CycleGANs
February 04, 2021
About a decade ago, a friend and I were talking over dinner about our shared passion for electronic music. Our burning question was: how can we convince the world to love techno as much as we do? A common—and not always unfair—criticism is that electronic music can be repetitive, lacking an overall arc, or overly simplistic: especially when compared to something like a classical concerto. However…
Keep Reading →
Will we share a moral landscape with artificial intelligence?
January 07, 2021
In The Moral Landscape, Sam Harris describes a framework which allows for strictly rational conversations about morality. It is based on the idea that ethical questions are eventually decomposable into objective questions about suffering and flourishing: … in the moral sphere, it is safe to begin with the premise that it is good to avoid behaving in such a way as to produce the worst possible…
Keep Reading →
A future cancelled: existential risk deserves more of your attention – and our altruism
October 13, 2020
Bring to mind the person you care about most in the world. Now imagine sitting across from them, watching them play Russian roulette with a loaded revolver. If we found ourselves in that situation we’d jump out of our seat, rush over to them, grab the gun from their hand, and attempt to unload it. However, in The Precipice, Toby Ord estimates there is a 1-in-6 chance of humans wiping themselves…
Keep Reading →
September 17, 2020
It was a pleasure to speak with Joel last week!
Keep Reading →
Is training a neural network like forming a habit?
July 05, 2020
When training an artificial neural network, a simplified version of the classic workflow is: Set up an neural network with randomly-weighted neurons Feed an example from the training set into the network Calculate the difference between the actual and expected output Use backpropagation to update the weights of all the neurons in the network Go to #2 until you have processed the entire training…
Keep Reading →