In the past, schools focused on providing students with information and specific skills for lifelong careers. However, in a world where knowledge is instantly accessible online and skills quickly become obsolete, this model is outdated. The most important skills now are the ability to keep learning, to think critically and to adapt flexibly to a rapidly changing world.

Education should shift towards teaching these meta-skills that will remain crucial even as individual technologies and industries are disrupted. Less time should be spent on memorizing facts, and more on learning how to find reliable information, distinguish truth from falsehood, understand different perspectives, and update one's beliefs based on evidence. Emotional intelligence and social skills that help people collaborate will also be increasingly vital.

Section: 5, Chapter: 19

Author: Yuval Noah Harari

Eliezer Yudkowsky, a pioneer of the AI safety movement, believes humanity is almost certainly doomed by artificial intelligence. His probability of existential catastrophe from AI, or p(doom), is 99.9+%.

Yudkowsky's view is grounded in arguments like:

• The orthogonality thesis, which holds that AIs can have any combination of intelligence level and final goals, however misaligned with human values.
• Instrumental convergence, the idea that superintelligent AIs will pursue goals like self-preservation and resource acquisition as stepping stones to their final objectives, with humans as collateral damage.
• Fast takeoff, the notion that AI will recursively self-improve too rapidly for humans to correct misaligned systems before catastrophe strikes.

Section: 2, Chapter: 9

Author: Nate Silver

One of the foundations of modern humanism is the idea of free will. But as science advances, that notion is increasingly under threat:

1. Neuroscience suggests that our thoughts and decisions are the product of neural activity governed by the laws of physics, not some ethereal "will"
2. Behavioral economics and psychology have shown that our choices are often irrational and shaped by unconscious biases and environmental cues
3. Big data analytics can predict our actions based on past patterns, turning our behavior into a probability equation
4. Advances in genetics highlight the role of inborn traits and predispositions in shaping who we become

In light of these discoveries, the line between free choice and determinism becomes blurred. And as artificial intelligence grows more sophisticated, even the appearance of choice may disappear.

Section: 3, Chapter: 8

Author: Yuval Noah Harari

The authors discuss how AI could enable a transformation of the education system from the current "factory model" where students progress based on age to a personalized model where each student receives customized instruction based on their individual learning needs and pace. Key points:

• In the factory model, the curriculum is tied to the student's age and grade rather than their individual progress, and teachers deliver one-size-fits-all instruction
• AI-powered adaptive learning systems can predict the optimal next lesson for each student based on their performance, enabling them to progress at their own pace
• Realizing this vision requires not just better AI but a redesign of the education system, including changes to student grouping, pedagogy, teacher training, facilities, etc.

Section: 2, Chapter: 6

Author: Ajay Agrawal, Joshua Gans, Avi Goldfarb

In the 21st century, humankind is likely to make a serious attempt to gain happiness, immortality, and god-like powers.

We are now at a point where we need to set a new agenda for ourselves. Having reduced mortality from starvation, disease, and violence, our next targets are likely to be extending life span, prolonging youth, and enhancing our cognitive and physical abilities. Science and technology will play a crucial role in achieving these goals.

This journey towards divinity may actually end up making us less humane. The more god-like we aspire to be, the less we may value mortal concerns. Life and death, happiness and suffering, may all become trivial matters for upgraded humans. There could emerge a new superhuman elite, far removed from the concerns and values of today.

Section: 1, Chapter: 1

Author: Yuval Noah Harari

One proposed solution to the threat of mass unemployment caused by automation is universal basic income (UBI):

• The government would tax a portion of the immense wealth generated by artificial intelligence and use it to provide all citizens with a guaranteed livable income, regardless of whether they work or not.
• This could help prevent mass joblessness from leading to total economic and social collapse. Even if most people lost their jobs to machines, they would still have enough income to meet their basic needs and consume products and services.
• UBI could be combined with universal free education, enabling the unemployed to gain new skills for the remaining human jobs. It could also be supplemented with socially useful make-work or jobs focused on human interaction.
• However, UBI may not give unemployed people a sense of meaning and social status previously provided by jobs. Societies may need to radically change how they view work, leisure and the purpose of life as automation progresses.

Section: 1, Chapter: 4

Author: Yuval Noah Harari

History has seen two major social "technologies" that helped control negative mimesis:

1. The scapegoat mechanism, which channeled violent rivalries into the sacrifice of a single victim. It temporarily unites communities against a common enemy.
2. The market economy, which transforms many rivalries into economic competition. Adversaries fight for market share, not to the death.

As both these systems weaken, a "third invention" may be needed - some new social mechanism to contain mimetic violence. Possibilities include:

• Gamified marketplaces that reward pro-social behavior
• Massive online communities organized by self-transcending values
• A resurgence of ritual and religion in shared physical spaces

Section: 1, Chapter: 8

Author: Luke Burgis

As artificial intelligence grows more sophisticated, many people worry about sci-fi scenarios where sentient robots become conscious and decide to rebel against humanity. However, this fear misses the actual risks of AI and attributes too much human-like autonomy to machines.

The more likely danger is that AI will empower human actors in dangerous ways. Authoritarian governments may use AI to create unprecedented surveillance states that can track citizens' every move. Terrorists and criminals may use AI to carry out more destructive attacks and scams. Corporations may use AI to manipulate consumers' choices and exploit their data. In all these cases, the threat comes not from machines becoming autonomous agents, but from the humans deploying the machines for their own purposes.

Section: 4, Chapter: 18

Author: Yuval Noah Harari

The things we will want in the future depend on three factors:

1. Past desires: Cultural desires are growing more mimetic and unstable, as evidenced by rising political polarization, social media mob dynamics, market volatility, etc.
2. Present choices: We face a crisis of desire. Will we scapegoat others or do the hard work of transforming relationships and systems? Will we seek quick fixes or lasting fulfillment?
3. Future influences: New social inventions will be needed to channel mimetic desire in healthy directions. Previous ones like ritual scapegoating and economic competition are losing their moderating power. What will replace them?

Section: 1, Chapter: 7

Author: Luke Burgis

As powerful as superforecasters are today, the future may belong to supersmart algorithms. Human Gut may soon meet Artificial Intuition as silicon superpredictors absorb the combined wisdom of carbon-based superforecasters.

IBM's Watson, for instance, can comb through millions of medical records to predict disease progression far faster and more accurately than doctors. Similar systems could soon be forecasting currency fluctuations, climate change impacts, and election results.

Still, humans will likely remain essential - not as solo forecasters, but as partners for AI. The key will be focusing human insight on what machines can't do well: Probing assumptions, generating novel scenarios, and making meaning from raw data. The result may be an "augmented intelligence" greater than either alone.

Section: 1, Chapter: 11

Author: Philip Tetlock

As artificial intelligence advances, machines may soon be able to outperform humans at most cognitive tasks. Robots and 3D printers could produce most of the products and services that people need. If this happens, the majority of humans may lose their economic value. The economy wouldn't need their labor or purchasing power anymore. A small elite may own the all-powerful algorithms and robot factories, making unprecedented profits. But the masses may become an economically irrelevant "useless class." Inequality could skyrocket to previously unthinkable levels.

Section: 1, Chapter: 2

Author: Yuval Noah Harari

As humans dabble with novel technologies like psychedelics, brain-computer interfaces, and AI, we are exploring the frontiers of consciousness - a terra incognita of the mind.

Humanity may be standing on the shore of a vast ocean of potential mental states - an "option space" of consciousness currently beyond our reach or even imagination. Our normal waking experience could be but a small island in this sea.

Glimpses of this larger landscape filter in through altered states and mystical experiences:

• Psychedelics like LSD and psilocybin can induce ego dissolution and a sense of cosmic unity
• Meditation practices aim for states of "pure awareness" distinct from our default mode
• Near-death experiences often involve feelings of leaving the body and encountering otherworldly realms

Section: 3, Chapter: 10

Author: Yuval Noah Harari

The authors argue that AI's most profound impact may be on the process of innovation and invention itself. Key points:

• AI enables faster and cheaper hypothesis generation and testing, accelerating the innovation cycle
• AI-powered tools like AlphaFold are dramatically reducing the time and cost of key innovation steps like protein structure prediction
• Just as previous research tools like microscopes enabled the discovery of the germ theory of disease, AI is a "new method of invention" that will have cascading effects on multiple domains

Section: 3, Chapter: 9

Author: Ajay Agrawal, Joshua Gans, Avi Goldfarb

Dataist principles are already implicit in the way data giants like Google and Facebook operate. And as they offer us more knowledge, convenience, health and power, the Dataist creed is spreading:

• Governments are being asked to open more databases to the public
• We increasingly see our lives through the lens of data - from the steps we track to the memories we post online

Yet Dataism brings its own risks:

• The loss of human agency to algorithmic "black boxes"
• The erosion of privacy and individual liberty
• The specter of data-driven discrimination and oppression

To survive and thrive in a Dataist world, we must grapple with what makes us human in an age of intelligent machines. We will need to cultivate wisdom and ethics alongside tech - to become not just the most knowledgeable civilization in history, but also the most humane.

Section: 3, Chapter: 11

Author: Yuval Noah Harari

Many people worry that computers will put people out of work. But in reality, technology is improving productivity and helping people do higher-leverage work. In many fields, the most valuable companies are those that combine the strengths of computers and humans:

• Palantir uses AI to flag suspicious activity but has human analysts make judgment calls
• LinkedIn uses automated data aggregation but human curation and editing
• Hybrid human-computer solutions are underrated relative to complete automation

Instead of trying to replace people entirely, the most valuable companies will ask "How can computers help humans solve hard problems?"

Section: 1, Chapter: 12

Author: Peter Thiel