Thread by @le_science4all

Today, at @EPFL_en @EPFL @EPFLevents, @harari_yuval, arguably today's most influential thinker, is giving a talk to an audience of 2700 (!).

Harari wrote the highly celebrated books Sapiens, Homo Deus, and 21 Lessons for the 21st Century. I highly recommend them!

He started with such a major claim: humans are hackable.

Using Biology, Computing and Data, science delivered the Ability to Hack Humans.

For the better, and the worse. If not for the worst.

Technology is upsetting our world. We urgently need to set it in the right direction.

Unfortunately, Harari claims, today's politics is almost solely about nostalgic fantasies. They discuss issues of the past.

He argues that we should instead focus on issues of the future. In particular on algorithms and data, who are bound to play an increasingly critical role.

He asserts that "those who control the flow of data will determine the trajectory of the future".

I strongly agree with this. I'd argue that the flow of data is the heartbeat of the modern world.

Now, ask yourself, who's doing this today? Who controls the flow of data?

By far, today's most central data flow management system are recommender systems. Not Google and Facebook. But their algorithms.

In particular, I'd claim that YouTube's AI is the one entity that the future of mankind currently most depends on...

According to @harari_yuval, the "worst news at the moment" is the technological arms race between companies and between countries, in particular the AI arms race.

The problem with this, he says, is that this race motivates us to discard safety and ethical concerns. #AISafety

@harari_yuval also argues that the power of controlling the flow of data should be much more decentralized.

Not sure about this though. Many scholars have argued that decentralization may further motivate the AI arms race, at the expense of vision and safety...

Three other scholars have joined Harari on stage. One of them has a beautiful blue short. This dude is a 2017 Nobel laureate @unil professor Jacques Dubochet. 😊

Harari seems to keep coming back to the (AI) arms race issue.

He says that, back in 1850s, both in UK and Germany, kids worked in mines. Both countries argued "if our kids don't work in mines, the other countries will outrun us".

No one even thought of sending kids to school.

Eventually both countries realized that sending kids school was actually beneficial to the economy. But it took a while.

This simple idea was out of reach, Harari argues, because of the arms race mindset. This blinded them from more effective solutions or more pressing problems.

Reacting on a neurotechnology of two @epfl and @unil researchers @harari_yuval encourages researchers to think beyond the single application of their work.

Researches often have side effects, typically in advancing the understanding of a field, which may profit malicious actors.

This resonates with a recent publication of Nick Bostrom on the "vulnerable world hypothesis". I highly recommend this read!

The idea is that the process of discovery is mostly irreversible. Each discovery is unlikely to have undesirable consequences.
lesswrong.com/posts/Tx6dGzYL…

However, it seems not-unlikely that at least one discovery will have potentially catastrophic side effects.

As a result, scientists seem to have a responsibility to prioritize research that will help defend against discoveries with potentially catastrophic side effects.

Harari concludes that his priority now is to redirect the public conversation towards the most pressing problem.

This requires, he argues, a lot of great communicators able to explain the upcoming challenges and opportunities.

Well... I am going to take this personally 😊

A huge thank you to @epfl and @Empowerment_F for organizing this great event!

And I feel extremely excited by the fact that these institutions want to promote a better communication of today's most pressing problems, the underlying science and the big ethical questions!