December 15, 2020

13: “Humans might one day need to beg AIs for our sentient rights” – AI expert Roman Yampolskiy – Sentientist Conversations

38 minutes

Full show notes & links here.

Roman is a Professor of Computer Science & Engineering at the University of Louisville. He is known for his work on behavioral biometrics, the security of cyberworlds & artificial intelligence safety. He founded the field of intellectology – the analysis of the forms & limits of intelligence. He is director of the Cyber Security Laboratory in the department of Computer Engineering &Computer Science at the Speed School of Engineering. Roman has written over 100 publications, including many books spanning these fields.

In these Sentientist Conversations, we talk about the two most important questions: “what’s real?” & “what matters?”

To catch the cameo from Luna the puppy ("seems conscious") watch the video of our conversation here. Don't forget to subscribe to our channel while you're there.

We discuss:

Growing up in the Soviet Union. Not much religion around.

Not meeting anyone religious until coming to the US as an adult

Not finding religious arguments interesting or compelling

Fascination with intersection of big ideas, philosophy, science

Questions can come from religion, but standards of evidence come from science

Comfort with others holding supernatural beliefs – helps us remain open-minded

Is god analogous with someone running a world simulation?

We need to get better at evaluating evidence. Should be separate from theories/hypotheses

The need for scientific humility

Deep fakes

Freedom as an ethical foundation, subject to not hurting others or restricting the freedom of others

Can we develop a pop-up AI that guides our ethics?

Consciousness / sentience warrants protection

Can Artificial Intelligences achieve consciousness or even super-consciousness?

Humans might need to beg future AIs for our rights, as we grant rights to animals

The hypocrisy of thinking animals should have rights, but enjoying eating them (theory vs. practice & cognitive dissonance)

Why so many AI researchers are ready to acknowledge AI sentience but forget or disregard non-human animal sentience

Consciousness, sentience, qualia, the “Hard Problem”, David Chalmers

Assessing sentience

The role of observers in quantum physics. Could there be some non-material element of consciousness?

Meeting Luna the puppy “Seems conscious!”

Will future AIs warrant protection/rights

Ending animal farming to set a good example to our future AI overlords

AIs will prefer “sentientism” to “humanism”

Substrate independence

Sentience/consciousness as a spectrum, simple to super (beyond human)

Ethical challenges with non-sentient AI

If human agents can’t agree (value alignment problem), can we even move towards a shared environment that will make us all happy? Maybe everyone could have their own individual virtual world!

Even a positively negotiated shared environment wouldn’t be as good for each of us as a perfect individual environment – just don’t switch it off

Can animal farming go away in a few years via clean-meat etc?

Veganism and moral resolutions to cognitive dissonance, vs. tech alternatives removing blockers

The dangers of disenfranchising humans if we grant rights too broadly (e.g. to trillions of bacteria or sentient AIs)

Equal vs. degrees of moral consideration

“Most of us will be as ethical as our choices”.

...more

View all episodes

By Jamie Woodhouse

2424 ratings