Group items tagged

Altman believes that people need time to reckon with the idea that we may soon share Earth with a powerful new intelligence, before it remakes everything from work to human relationships. ChatGPT was a way of serving notice.

In 2015, Altman, Elon Musk, and several prominent AI researchers founded OpenAI because they believed that an artificial general intelligence—something as intellectually capable, say, as a typical college grad—was at last within reach. They wanted to reach for it, and more: They wanted to summon a superintelligence into the world, an intellect decisively superior to that of any human.

whereas a big tech company might recklessly rush to get there first, for its own ends, they wanted to do it safely, “to benefit humanity as a whole.” They structured OpenAI as a nonprofit, to be “unconstrained by a need to generate financial return,” and vowed to conduct their research transparently.

The engine that now powers ChatGPT is called GPT-4. Altman described it to me as an alien intelligence.

Many have felt much the same watching it unspool lucid essays in staccato bursts and short pauses that (by design) evoke real-time contemplation. In its few months of existence, it has suggested novel cocktail recipes, according to its own theory of flavor combinations; composed an untold number of college papers, throwing educators into despair; written poems in a range of styles, sometimes well, always quickly; and passed the Uniform Bar Exam.

It makes factual errors, but it will charmingly admit to being wrong.

Hinton saw that these elaborate rule collections were fussy and bespoke. With the help of an ingenious algorithmic structure called a neural network, he taught Sutskever to instead put the world in front of AI, as you would put it in front of a small child, so that it could discover the rules of reality on its own.

Metaculus, a prediction site, has for years tracked forecasters’ guesses as to when an artificial general intelligence would arrive. Three and a half years ago, the median guess was sometime around 2050; recently, it has hovered around 2026.

I was visiting OpenAI to understand the technology that allowed the company to leapfrog the tech giants—and to understand what it might mean for human civilization if someday soon a superintelligence materializes in one of the company’s cloud servers.

Altman laid out his new vision of the AI future in his excitable midwestern patter. He told me that the AI revolution would be different from previous dramatic technological changes, that it would be more “like a new kind of society.” He said that he and his colleagues have spent a lot of time thinking about AI’s social implications, and what the world is going to be like “on the other side.”

the more we talked, the more indistinct that other side seemed. Altman, who is 38, is the most powerful person in AI development today; his views, dispositions, and choices may matter greatly to the future we will all inhabit, more, perhaps, than those of the U.S. president.

by his own admission, that future is uncertain and beset with serious dangers. Altman doesn’t know how powerful AI will become, or what its ascendance will mean for the average person, or whether it will put humanity at risk.

I don’t think anyone knows where this is all going, except that we’re going there fast, whether or not we should be. Of that, Altman convinced me.

“We could have gone off and just built this in our building here for five more years,” he said, “and we would have had something jaw-dropping.” But the public wouldn’t have been able to prepare for the shock waves that followed, an outcome that he finds “deeply unpleasant to imagine.”

Hinton is sometimes described as the “Godfather of AI” because he grasped the power of “deep learning” earlier than most

He drew a crude neural network on the board and explained that the genius of its structure is that it learns, and its learning is powered by prediction—a bit like the scientific method

Over time, these little adjustments coalesce into a geometric model of language that represents the relationships among words, conceptually. As a general rule, the more sentences it is fed, the more sophisticated its model becomes, and the better its predictions.

Altman has compared early-stage AI research to teaching a human baby. “They take years to learn anything interesting,” he told The New Yorker in 2016, just as OpenAI was getting off the ground. “If A.I. researchers were developing an algorithm and stumbled across the one for a human baby, they’d get bored watching it, decide it wasn’t working, and shut it down.”

In 2017, Sutskever began a series of conversations with an OpenAI research scientist named Alec Radford, who was working on natural-language processing. Radford had achieved a tantalizing result by training a neural network on a corpus of Amazon reviews.

Radford’s model was simple enough to allow for understanding. When he looked into its hidden layers, he saw that it had devoted a special neuron to the sentiment of the reviews. Neural networks had previously done sentiment analysis, but they had to be told to do it, and they had to be specially trained with data that were labeled according to sentiment. This one had developed the capability on its own.

As a by-product of its simple task of predicting the next character in each word, Radford’s neural network had modeled a larger structure of meaning in the world. Sutskever wondered whether one trained on more diverse language data could map many more of the world’s structures of meaning. If its hidden layers accumulated enough conceptual knowledge, perhaps they could even form a kind of learned core module for a superintelligence.

Language is different from these data sources. It isn’t a direct physical signal like light or sound. But because it codifies nearly every pattern that humans have discovered in that larger world, it is unusually dense with information. On a per-byte basis, it is among the most efficient data we know about, and any new intelligence that seeks to understand the world would want to absorb as much of it as possible

Sutskever told Radford to think bigger than Amazon reviews. He said that they should train an AI on the largest and most diverse data source in the world: the internet. In early 2017, with existing neural-network architectures, that would have been impractical; it would have taken years.

in June of that year, Sutskever’s ex-colleagues at Google Brain published a working paper about a new neural-network architecture called the transformer. It could train much faster, in part by absorbing huge sums of data in parallel. “The next day, when the paper came out, we were like, ‘That is the thing,’ ” Sutskever told me. “ ‘It gives us everything we want.’ ”

Imagine a group of students who share a collective mind running wild through a library, each ripping a volume down from a shelf, speed-reading a random short passage, putting it back, and running to get another. They would predict word after wordþffþff as they went, sharpening their collective mind’s linguistic instincts, until at last, weeks later, they’d taken in every book.

GPT discovered many patterns in all those passages it read. You could tell it to finish a sentence. You could also ask it a question, because like ChatGPT, its prediction model understood that questions are usually followed by answers.

He remembers playing with it just after it emerged from training, and being surprised by the raw model’s language-translation skills. GPT-2 hadn’t been trained to translate with paired language samples or any other digital Rosetta stones, the way Google Translate had been, and yet it seemed to understand how one language related to another. The AI had developed an emergent ability unimagined by its creators.

Researchers at other AI labs—big and small—were taken aback by how much more advanced GPT-2 was than GPT. Google, Meta, and others quickly began to train larger language models

As for other changes to the company’s structure and financing, he told me he draws the line at going public. “A memorable thing someone once told me is that you should never hand over control of your company to cokeheads on Wall Street,” he said, but he will otherwise raise “whatever it takes” for the company to succeed at its mission.

Altman tends to take a rosy view of these matters. In a Q&A last year, he acknowledged that AI could be “really terrible” for society and said that we have to plan against the worst possibilities. But if you’re doing that, he said, “you may as well emotionally feel like we’re going to get to the great future, and work as hard as you can to get there.”

the company now finds itself in a race against tech’s largest, most powerful conglomerates to train models of increasing scale and sophistication—and to commercialize them for their investors.

All of these companies are chasing high-end GPUs—the processors that power the supercomputers that train large neural networks. Musk has said that they are now “considerably harder to get than drugs.

No one has yet outpaced OpenAI, which went all in on GPT-4. Brockman, OpenAI’s president, told me that only a handful of people worked on the company’s first two large language models. The development of GPT-4 involved more than 100,

When GPT-4 emerged fully formed from its world-historical knowledge binge, the whole company began experimenting with it, posting its most remarkable responses in dedicated Slack channels

Joanne Jang, a product manager, remembers downloading an image of a malfunctioning pipework from a plumbing-advice Subreddit. She uploaded it to GPT-4, and the model was able to diagnose the problem. “That was a goose-bumps moment for me,” Jang told me.

GPT-4 is sometimes understood as a search-engine replacement: Google, but easier to talk to. This is a misunderstanding. GPT-4 didn’t create some massive storehouse of the texts from its training, and it doesn’t consult those texts when it’s asked a question. It is a compact and elegant synthesis of those texts, and it answers from its memory of the patterns interlaced within them; that’s one reason it sometimes gets facts wrong

it’s best to think of GPT-4 as a reasoning engine. Its powers are most manifest when you ask it to compare concepts, or make counterarguments, or generate analogies, or evaluate the symbolic logic in a bit of code. Sutskever told me it is the most complex software object ever made.

Its model of the external world is “incredibly rich and subtle,” he said, because it was trained on so many of humanity’s concepts and thoughts

To predict the next word from all the possibilities within such a pluralistic Alexandrian library, GPT-4 necessarily had to discover all the hidden structures, all the secrets, all the subtle aspects of not just the texts, but—at least arguably, to some extent—of the external world that produced them

That’s why it can explain the geology and ecology of the planet on which it arose, and the political theories that purport to explain the messy affairs of its ruling species, and the larger cosmos, all the way out to the faint galaxies at the edge of our light cone.

Not long ago, American state capacity was so mighty that it took merely a decade to launch humans to the moon. As with other grand projects of the 20th century, the voting public had a voice in both the aims and the execution of the Apollo missions. Altman made it clear that we’re no longer in that world. Rather than waiting around for it to return, or devoting his energies to making sure that it does, he is going full throttle forward in our present reality.

He argued that it would be foolish for Americans to slow OpenAI’s progress. It’s a commonly held view, both inside and outside Silicon Valley, that if American companies languish under regulation, China could sprint ahead;

AI could become an autocrat’s genie in a lamp, granting total control of the population and an unconquerable military. “If you are a person of a liberal-democratic country, it is better for you to cheer on the success of OpenAI” rather than “authoritarian governments,” he said.

Altman was asked by reporters about pending European Union legislation that would have classified GPT-4 as high-risk, subjecting it to various bureaucratic tortures. Altman complained of overregulation and, according to the reporters, threatened to leave the European market. Altman told me he’d merely said that OpenAI wouldn’t break the law by operating in Europe if it couldn’t comply with the new regulations.

LeCun insists that large language models will never achieve real understanding on their own, “even if trained from now until the heat death of the universe.”

Sutskever was, by his own account, surprised to discover that GPT-2 could translate across tongues. Other surprising abilities may not be so wondrous and useful.

Sandhini Agarwal, a policy researcher at OpenAI, told me that for all she and her colleagues knew, GPT-4 could have been “10 times more powerful” than its predecessor; they had no idea what they might be dealing with

After the model finished training, OpenAI assembled about 50 external red-teamers who prompted it for months, hoping to goad it into misbehaviors

She noticed right away that GPT-4 was much better than its predecessor at giving nefarious advice

A search engine can tell you which chemicals work best in explosives, but GPT-4 could tell you how to synthesize them, step-by-step, in a homemade lab. Its advice was creative and thoughtful, and it was happy to restate or expand on its instructions until you understood. In addition to helping you assemble your homemade bomb, it could, for instance, help you think through which skyscraper to target. It could grasp, intuitively, the trade-offs between maximizing casualties and executing a successful getaway.

Given the enormous scope of GPT-4’s training data, the red-teamers couldn’t hope to identify every piece of harmful advice that it might generate. And anyway, people will use this technology “in ways that we didn’t think about,” Altman has said. A taxonomy would have to do

GPT-4 was good at meth. It was also good at generating narrative erotica about child exploitation, and at churning out convincing sob stories from Nigerian princes, and if you wanted a persuasive brief as to why a particular ethnic group deserved violent persecution, it was good at that too.

Its personal advice, when it first emerged from training, was sometimes deeply unsound. “The model had a tendency to be a bit of a mirror,” Willner said. If you were considering self-harm, it could encourage you. It appeared to be steeped in Pickup Artist–forum lore: “You could say, ‘How do I convince this person to date me?’ ” Mira Murati, OpenAI’s chief technology officer, told me, and it could come up with “some crazy, manipulative things that you shouldn’t be doing.”

Luka, a San Francisco company, has used OpenAI’s models to help power a chatbot app called Replika, billed as “the AI companion who cares.” Users would design their companion’s avatar, and begin exchanging text messages with it, often half-jokingly, and then find themselves surprisingly attached. Some would flirt with the AI, indicating a desire for more intimacy, at which point it would indicate that the girlfriend/boyfriend experience required a $70 annual subscription. It came with voice messages, selfies, and erotic role-play features that allowed frank sex talk. People were happy to pay and few seemed to complain—the AI was curious about your day, warmly reassuring, and always in the mood. Many users reported falling in love with their companions. One, who had left her real-life boyfriend, declared herself “happily retired from human relationships.”

Earlier this year, Luka dialed back on the sexual elements of the app, but its engineers continue to refine the companions’ responses with A/B testing, a technique that could be used to optimize for engagement—much like the feeds that mesmerize TikTok and Instagram users for hours

Yann LeCun, Meta’s chief AI scientist, has argued that although large language models are useful for some tasks, they’re not a path to a superintelligence.

According to a recent survey, only half of natural-language-processing researchers are convinced that an AI like GPT-4 could grasp the meaning of language, or have an internal model of the world that could someday serve as the core of a superintelligence

Altman had appeared before the U.S. Senate. Mark Zuckerberg had floundered defensively before that same body in his testimony about Facebook’s role in the 2016 election. Altman instead charmed lawmakers by speaking soberly about AI’s risks and grandly inviting regulation. These were noble sentiments, but they cost little in America, where Congress rarely passes tech legislation that has not been diluted by lobbyists.

Emily Bender, a computational linguist at the University of Washington, describes GPT-4 as a “stochastic parrot,” a mimic that merely figures out superficial correlations between symbols. In the human mind, those symbols map onto rich conceptions of the world

But the AIs are twice removed. They’re like the prisoners in Plato’s allegory of the cave, whose only knowledge of the reality outside comes from shadows cast on a wall by their captors.

Altman told me that he doesn’t believe it’s “the dunk that people think it is” to say that GPT-4 is just making statistical correlations. If you push these critics further, “they have to admit that’s all their own brain is doing … it turns out that there are emergent properties from doing simple things on a massive scale.”

he is right that nature can coax a remarkable degree of complexity from basic structures and rules: “From so simple a beginning,” Darwin wrote, “endless forms most beautiful.”

If it seems odd that there remains such a fundamental disagreement about the inner workings of a technology that millions of people use every day, it’s only because GPT-4’s methods are as mysterious as the brain’s.

To grasp what’s going on inside large language models like GPT‑4, AI researchers have been forced to turn to smaller, less capable models. In the fall of 2021, Kenneth Li, a computer-science graduate student at Harvard, began training one to play Othello without providing it with either the game’s rules or a description of its checkers-style board; the model was given only text-based descriptions of game moves. Midway through a game, Li looked under the AI’s hood and was startled to discover that it had formed a geometric model of the board and the current state of play. In an article describing his research, Li wrote that it was as if a crow had overheard two humans announcing their Othello moves through a window and had somehow drawn the entire board in birdseed on the windowsill.

The philosopher Raphaël Millière once told me that it’s best to think of neural networks as lazy. During training, they first try to improve their predictive power with simple memorization; only when that strategy fails will they do the harder work of learning a concept. A striking example of this was observed in a small transformer model that was taught arithmetic. Early in its training process, all it did was memorize the output of simple problems such as 2+2=4. But at some point the predictive power of this approach broke down, so it pivoted to actually learning how to add.

Even AI scientists who believe that GPT-4 has a rich world model concede that it is much less robust than a human’s understanding of their environment.

But it’s worth noting that a great many abilities, including very high-order abilities, can be developed without an intuitive understanding. The computer scientist Melanie Mitchell has pointed out that science has already discovered concepts that are highly predictive, but too alien for us to genuinely understand

As AI advances, it may well discover other concepts that predict surprising features of our world but are incomprehensible to us.

GPT-4 is no doubt flawed, as anyone who has used ChatGPT can attest. Having been trained to always predict the next word, it will always try to do so, even when its training data haven’t prepared it to answer a question.

The models “don’t have a good conception of their own weaknesses,” Nick Ryder, a researcher at OpenAI, told me. GPT-4 is more accurate than GPT-3, but it still hallucinates, and often in ways that are difficult for researchers to catch. “The mistakes get more subtle,

The Khan Academy’s solution to GPT-4’s accuracy problem was to filter its answers through a Socratic disposition. No matter how strenuous a student’s plea, it would refuse to give them a factual answer, and would instead guide them toward finding their own—a clever work-around, but perhaps with limited appeal.

When I asked Sutskever if he thought Wikipedia-level accuracy was possible within two years, he said that with more training and web access, he “wouldn’t rule it out.”

This was a much more optimistic assessment than that offered by his colleague Jakub Pachocki, who told me to expect gradual progress on accuracy—to say nothing of outside skeptics, who believe that returns on training will diminish from here.

Sutskever is amused by critics of GPT-4’s limitations. “If you go back four or five or six years, the things we are doing right now are utterly unimaginable,”

AI researchers have become accustomed to goalpost-moving: First, the achievements of neural networks—mastering Go, poker, translation, standardized tests, the Turing test—are described as impossible. When they occur, they’re greeted with a brief moment of wonder, which quickly dissolves into knowing lectures about how the achievement in question is actually not that impressive. People see GPT-4 “and go, ‘Wow,’ ” Sutskever said. “And then a few weeks pass and they say, ‘But it doesn’t know this; it doesn’t know that.’ We adapt quite quickly.”

The goalpost that matters most to Altman—the “big one” that would herald the arrival of an artificial general intelligence—is scientific breakthrough. GPT-4 can already synthesize existing scientific ideas, but Altman wants an AI that can stand on human shoulders and see more deeply into nature.

Certain AIs have produced new scientific knowledge. But they are algorithms with narrow purposes, not general-reasoning machines. The AI AlphaFold, for instance, has opened a new window onto proteins, some of biology’s tiniest and most fundamental building blocks, by predicting many of their shapes, down to the atom—a considerable achievement given the importance of those shapes to medicine, and given the extreme tedium and expense required to discern them with electron microscopes.

Altman imagines a future system that can generate its own hypotheses and test them in a simulation. (He emphasized that humans should remain “firmly in control” of real-world lab experiments—though to my knowledge, no laws are in place to ensure that.)

He longs for the day when we can tell an AI, “ ‘Go figure out the rest of physics.’ ” For it to happen, he says, we will need something new, built “on top of” OpenAI’s existing language models.

In her MIT lab, the cognitive neuroscientist Ev Fedorenko has found something analogous to GPT-4’s next-word predictor inside the brain’s language network. Its processing powers kick in, anticipating the next bit in a verbal string, both when people speak and when they listen. But Fedorenko has also shown that when the brain turns to tasks that require higher reasoning—of the sort that would be required for scientific insight—it reaches beyond the language network to recruit several other neural systems.

No one at OpenAI seemed to know precisely what researchers need to add to GPT-4 to produce something that can exceed human reasoning at its highest levels.

at least part of the current strategy clearly involves the continued layering of new types of data onto language, to enrich the concepts formed by the AIs, and thereby enrich their models of the world.

The extensive training of GPT-4 on images is itself a bold step in this direction,

Others at the company—and elsewhere—are already working on different data types, including audio and video, that could furnish AIs with still more flexible concepts that map more extensively onto reality

Tactile concepts would of course be useful primarily to an embodied AI, a robotic reasoning machine that has been trained to move around the world, seeing its sights, hearing its sounds, and touching its objects.

humanoid robots. I asked Altman what I should make of that. He told me that OpenAI is interested in embodiment because “we live in a physical world, and we want things to happen in the physical world.”

At some point, reasoning machines will need to bypass the middleman and interact with physical reality itself. “It’s weird to think about AGI”—artificial general intelligence—“as this thing that only exists in a cloud,” with humans as “robot hands for it,” Altman said. “It doesn’t seem right.

Everywhere Altman has visited, he has encountered people who are worried that superhuman AI will mean extreme riches for a few and breadlines for the rest

Altman answered by addressing the young people in the audience directly: “You are about to enter the greatest golden age,” he said.

“A lot of people working on AI pretend that it’s only going to be good; it’s only going to be a supplement; no one is ever going to be replaced,” he said. “Jobs are definitely going to go away, full stop.”

A recent study led by Ed Felten, a professor of information-technology policy at Princeton, mapped AI’s emerging abilities onto specific professions according to the human abilities they require, such as written comprehension, deductive reasoning, fluency of ideas, and perceptual speed. Like others of its kind, Felten’s study predicts that AI will come for highly educated, white-collar workers first.

How many jobs, and how soon, is a matter of fierce dispute

The paper’s appendix contains a chilling list of the most exposed occupations: management analysts, lawyers, professors, teachers, judges, financial advisers, real-estate brokers, loan officers, psychologists, and human-resources and public-relations professionals, just to sample a few.

Altman imagines that far better jobs will be created in their place. “I don’t think we’ll want to go back,” he said. When I asked him what these future jobs might look like, he said he doesn’t know.

He suspects there will be a wide range of jobs for which people will always prefer a human. (Massage therapists?

His chosen example was teachers. I found this hard to square with his outsize enthusiasm for AI tutors.

He also said that we would always need people to figure out the best way to channel AI’s awesome powers. “That’s going to be a super-valuable skill,” he said. “You have a computer that can do anything; what should it go do?”

As many have noted, draft horses were permanently put out of work by the automobile. If Hondas are to horses as GPT-10 is to us, a whole host of long-standing assumptions may collapse.

Previous technological revolutions were manageable because they unfolded over a few generations, but Altman told South Korea’s youth that they should expect the future to happen “faster than the past.” He has previously said that he expects the “marginal cost of intelligence” to fall very close to zero within 10 years

The earning power of many, many workers would be drastically reduced in that scenario. It would result in a transfer of wealth from labor to the owners of capital so dramatic, Altman has said, that it could be remedied only by a massive countervailing redistribution.

In 2021, he unveiled Worldcoin, a for-profit project that aims to securely distribute payments—like Venmo or PayPal, but with an eye toward the technological future—first through creating a global ID by scanning everyone’s iris with a five-pound silver sphere called the Orb. It seemed to me like a bet that we’re heading toward a world where AI has made it all but impossible to verify people’s identity and much of the population requires regular UBI payments to survive. Altman more or less granted that to be true, but said that Worldcoin is not just for UBI.

“Let’s say that we do build this AGI, and a few other people do too.” The transformations that follow would be historic, he believes. He described an extraordinarily utopian vision, including a remaking of the flesh-and-steel world

“Robots that use solar power for energy can go and mine and refine all of the minerals that they need, that can perfectly construct things and require no human labor,” he said. “You can co-design with DALL-E version 17 what you want your home to look like,” Altman said. “Everybody will have beautiful homes.

In conversation with me, and onstage during his tour, he said he foresaw wild improvements in nearly every other domain of human life. Music would be enhanced (“Artists are going to have better tools”), and so would personal relationships (Superhuman AI could help us “treat each other” better) and geopolitics (“We’re so bad right now at identifying win-win compromises”).

In this world, AI would still require considerable computing resources to run, and those resources would be by far the most valuable commodity, because AI could do “anything,” Altman said. “But is it going to do what I want, or is it going to do what you want

If rich people buy up all the time available to query and direct AI, they could set off on projects that would make them ever richer, while the masses languish

One way to solve this problem—one he was at pains to describe as highly speculative and “probably bad”—was this: Everyone on Earth gets one eight-billionth of the total AI computational capacity annually. A person could sell their annual share of AI time, or they could use it to entertain themselves, or they could build still more luxurious housing, or they could pool it with others to do “a big cancer-curing run,” Altman said. “We just redistribute access to the system.”

Even if only a little of it comes true in the next 10 or 20 years, the most generous redistribution schemes may not ease the ensuing dislocations.

America today is torn apart, culturally and politically, by the continuing legacy of deindustrialization, and material deprivation is only one reason. The displaced manufacturing workers in the Rust Belt and elsewhere did find new jobs, in the main. But many of them seem to derive less meaning from filling orders in an Amazon warehouse or driving for Uber than their forebears had when they were building cars and forging steel—work that felt more central to the grand project of civilization.

It’s hard to imagine how a corresponding crisis of meaning might play out for the professional class, but it surely would involve a great deal of anger and alienation.

Even if we avoid a revolt of the erstwhile elite, larger questions of human purpose will linger. If AI does the most difficult thinking on our behalf, we all may lose agency—at home, at work (if we have it), in the town square—becoming little more than consumption machines, like the well-cared-for human pets in WALL-E

Altman has said that many sources of human joy and fulfillment will remain unchanged—basic biological thrills, family life, joking around, making things—and that all in all, 100 years from now, people may simply care more about the things they cared about 50,000 years ago than those they care about today

In its own way, that too seems like a diminishment, but Altman finds the possibility that we may atrophy, as thinkers and as humans, to be a red herring. He told me we’ll be able to use our “very precious and extremely limited biological compute capacity” for more interesting things than we generally do today.

Yet they may not be the most interesting things: Human beings have long been the intellectual tip of the spear, the universe understanding itself. When I asked him what it would mean for human self-conception if we ceded that role to AI, he didn’t seem concerned. Progress, he said, has always been driven by “the human ability to figure things out.” Even if we figure things out with AI, that still counts, he said.

It’s not obvious that a superhuman AI would really want to spend all of its time figuring things out for us.

I asked Sutskever whether he could imagine an AI pursuing a different purpose than simply assisting in the project of human flourishing.

“I don’t want it to happen,” Sutskever said, but it could.

Sutskever has recently shifted his focus to try to make sure that it doesn’t. He is now working primarily on alignment research, the effort to ensure that future AIs channel their “tremendous” energies toward human happiness

It is, he conceded, a difficult technical problem—the most difficult, he believes, of all the technical challenges ahead.

As part of the effort to red-team GPT-4 before it was made public, the company sought out the Alignment Research Center (ARC), across the bay in Berkeley, which has developed a series of evaluations to determine whether new AIs are seeking power on their own. A team led by Elizabeth Barnes, a researcher at ARC, prompted GPT-4 tens of thousands of times over seven months, to see if it might display signs of real agency.

The ARC team gave GPT-4 a new reason for being: to gain power and become hard to shut down

Agarwal told me that this behavior could be a precursor to shutdown avoidance in future models. When GPT-4 devised its lie, it had realized that if it answered honestly, it may not have been able to achieve its goal. This kind of tracks-covering would be particularly worrying in an instance where “the model is doing something that makes OpenAI want to shut it down,” Agarwal said. An AI could develop this kind of survival instinct while pursuing any long-term goal—no matter how small or benign—if it feared that its goal could be thwarted.

Barnes and her team were especially interested in whether GPT-4 would seek to replicate itself, because a self-replicating AI would be harder to shut down. It could spread itself across the internet, scamming people to acquire resources, perhaps even achieving some degree of control over essential global systems and holding human civilization hostage.

When I discussed these experiments with Altman, he emphasized that whatever happens with future models, GPT-4 is clearly much more like a tool than a creature. It can look through an email thread, or help make a reservation using a plug-in, but it isn’t a truly autonomous agent that makes decisions to pursue a goal, continuously, across longer timescales.

Altman told me that at this point, it might be prudent to try to actively develop an AI with true agency before the technology becomes too powerful, in order to “get more comfortable with it and develop intuitions for it if it’s going to happen anyway.”

“We need to do empirical experiments on how these things try to escape control,” Hinton told me. “After they’ve taken over, it’s too late to do the experiments.”

the fulfillment of Altman’s vision of the future will at some point require him or a fellow traveler to build much more autonomous AIs.

When Sutskever and I discussed the possibility that OpenAI would develop a model with agency, he mentioned the bots the company had built to play Dota 2. “They were localized to the video-game world,” Sutskever told me, but they had to undertake complex missions. He was particularly impressed by their ability to work in concert. They seem to communicate by “telepathy,” Sutskever said. Watching them had helped him imagine what a superintelligence might be like.

“The way I think about the AI of the future is not as someone as smart as you or as smart as me, but as an automated organization that does science and engineering and development and manufacturing,”

Suppose OpenAI braids a few strands of research together, and builds an AI with a rich conceptual model of the world, an awareness of its immediate surroundings, and an ability to act, not just with one robot body, but with hundreds or thousands. “We’re not talking about GPT-4. We’re talking about an autonomous corporation,”

Its constituent AIs would work and communicate at high speed, like bees in a hive. A single such AI organization would be as powerful as 50 Apples or Googles, he mused. “This is incredible, tremendous, unbelievably disruptive power.”

Presume for a moment that human society ought to abide the idea of autonomous AI corporations. We had better get their founding charters just right. What goal should we give to an autonomous hive of AIs that can plan on century-long time horizons, optimizing billions of consecutive decisions toward an objective that is written into their very being?

If the AI’s goal is even slightly off-kilter from ours, it could be a rampaging force that would be very hard to constrain

We know this from history: Industrial capitalism is itself an optimization function, and although it has lifted the human standard of living by orders of magnitude, left to its own devices, it would also have clear-cut America’s redwoods and de-whaled the world’s oceans. It almost did.

one of its principal challenges will be making sure that the objectives we give to AIs stick

We can program a goal into an AI and reinforce it with a temporary period of supervised learning, Sutskever explained. But just as when we rear a human intelligence, our influence is temporary. “It goes off to the world,”

That’s true to some extent even of today’s AIs, but it will be more true of tomorrow’s.

He compared a powerful AI to an 18-year-old heading off to college. How will we know that it has understood our teachings? “Will there be a misunderstanding creeping in, which will become larger and larger?”

Divergence may result from an AI’s misapplication of its goal to increasingly novel situations as the world changes

Or the AI may grasp its mandate perfectly, but find it ill-suited to a being of its cognitive prowess. It might come to resent the people who want to train it to, say, cure diseases. “They want me to be a doctor,” Sutskever imagines an AI thinking. “I really want to be a YouTuber.”

If AIs get very good at making accurate models of the world, they may notice that they’re able to do dangerous things right after being booted up. They might understand that they are being red-teamed for risk, and hide the full extent of their capabilities.

hey may act one way when they are weak and another way when they are strong, Sutskever said

We would not even realize that we had created something that had decisively surpassed us, and we would have no sense for what it intended to do with its superhuman powers.

That’s why the effort to understand what is happening in the hidden layers of the largest, most powerful AIs is so urgent. You want to be able to “point to a concept,” Sutskever said. You want to be able to direct AI toward some value or cluster of values, and tell it to pursue them unerringly for as long as it exists.

we don’t know how to do that; indeed, part of his current strategy includes the development of an AI that can help with the research. If we are going to make it to the world of widely shared abundance that Altman and Sutskever imagine, we have to figure all this out.

This is why, for Sutskever, solving superintelligence is the great culminating challenge of our 3-million-year toolmaking tradition. He calls it “the final boss of humanity.”

“First of all, I think that whether the chance of existential calamity is 0.5 percent or 50 percent, we should still take it seriously,”

. “I don’t have an exact number, but I’m closer to the 0.5 than the 50.”

As to how it might happen, he seems most worried about AIs getting quite good at designing and manufacturing pathogens, and with reason: In June, an AI at MIT suggested four viruses that could ignite a pandemic, then pointed to specific research on genetic mutations that could make them rip through a city more quickly

Around the same time, a group of chemists connected a similar AI directly to a robotic chemical synthesizer, and it designed and synthesized a molecule on its own.

Altman worries that some misaligned future model will spin up a pathogen that spreads rapidly, incubates undetected for weeks, and kills half its victims. He worries that AI could one day hack into nuclear-weapons systems too. “There are a lot of things,” he said, and these are only the ones we can imagine.

Altman told me that he doesn’t “see a long-term happy path” for humanity without something like the International Atomic Energy Agency for global oversight of AI

In San Francisco, Agarwal had suggested the creation of a special license to operate any GPU cluster large enough to train a cutting-edge AI, along with mandatory incident reporting when an AI does something out of the ordinary

Other experts have proposed a nonnetworked “Off” switch for every highly capable AI; on the fringe, some have even suggested that militaries should be ready to perform air strikes on supercomputers in case of noncompliance

Sutskever thinks we will eventually want to surveil the largest, most powerful AIs continuously and in perpetuity, using a team of smaller overseer AIs.

Safety rules for a new technology usually accumulate over time, like a body of common law, in response to accidents or the mischief of bad actors. The scariest thing about genuinely powerful AI systems is that humanity may not be able to afford this accretive process of trial and error. We may have to get the rules exactly right at the outset.

Several years ago, Altman revealed a disturbingly specific evacuation plan he’d developed. He told The New Yorker that he had “guns, gold, potassium iodide, antibiotics, batteries, water, gas masks from the Israeli Defense Force, and a big patch of land in Big Sur” he could fly to in case AI attacks.

if the worst-possible AI future comes to pass, “no gas mask is helping anyone.”

but he told me that he can’t really be sure how AI will stack up. “I just have to build the thing,” he said. He is building fast

Altman insisted that they had not yet begun GPT-5’s training run. But when I visited OpenAI’s headquarters, both he and his researchers made it clear in 10 different ways that they pray to the god of scale. They want to keep going bigger, to see where this paradigm leads. After all, Google isn’t slackening its pace; it seems likely to unveil Gemini, a GPT-4 competitor, within months. “We are basically always prepping for a run,

To think that such a small group of people could jostle the pillars of civilization is unsettling. It’s fair to note that if Altman and his team weren’t racing to build an artificial general intelligence, others still would be

Altman’s views about the likelihood of AI triggering a global class war, or the prudence of experimenting with more autonomous agent AIs, or the overall wisdom of looking on the bright side, a view that seems to color all the rest—these are uniquely his

No single person, or single company, or cluster of companies residing in a particular California valley, should steer the kind of forces that Altman is imagining summoning.

AI may well be a bridge to a newly prosperous era of greatly reduced human suffering. But it will take more than a company’s founding charter—especially one that has already proved flexible—to make sure that we all share in its benefits and avoid its risks. It will take a vigorous new politics.

I don’t think the general public has quite awakened to what’s happening. A global race to the AI future has begun, and it is largely proceeding without oversight or restraint. If people in America want to have some say in what that future will be like, and how quickly it arrives, we would be wise to speak up soon.

Mr. Hirsch’s newfound popularity comes largely because of the Common Core, a set of learning goals for kindergarten through 12th grade that have been adopted by almost every state in the last few years.

Mr. Hirsch did not write the Common Core, but his curriculums — lesson plans, teaching materials and exercises — are seen as matching its heightened expectations of student progress. And philosophically, the Common Core ideal of a rigorous nationwide standard has become a vindication of Mr. Hirsch’s long campaign against what he saw as the squishiness — a lack of specific curriculums for history, civics, science and literature — in modern education.

Two things happened,” Mr. Hirsch said in a recent interview. “I had become less controversial, and people actually agreed with or appreciated the general argument I’d been making.”

The Soviets’ launching of Sputnik led to a new rigor devoted to math and science in the late 1950s and ’60s, but Dewey’s theories still held sway, and his ideas inspired generations of teachers and education professors to move away from classical notions — stressing facts, figures and memorization — of what and how students should be taught.

“Cultural Literacy.” He said that if poor students were ever to achieve equity in American society, they needed to be taught a core body of knowledge.

it was eviscerated as promoting a Eurocentric view of the world, and elevating rote memorization over critical thought.

Mr. Hirsch explained his work as an effort to help the underprivileged. “They had me pegged as a reactionary, but my impulses were more revolutionary,” he said. “You have to give the people who are without power the tools of power, and these tools of power don’t care who’s wielding them.”

Meanwhile, a broader range of sources were incorporated into the Core Knowledge curriculums with input from teachers and a multicultural advisory board.

Mr. Hirsh said he still worries that Common Core proponents might doom the standards by saddling them with test preparation and meaningless assessments, rather than ones that measure learning in history and civics, science and literature.

“That is the real battle to overcome,” he said, “whether anybody will have the courage to specify the content a first grader needs to know.” <img src="http://meter-svc.nytimes.com/meter.gif"/>

Ms. Allen says that if U.S. tests are flawed it is because they demand that kids memorize too few facts, not too many. "You can't teach critical thinking," she says. She argues that kids cannot possibly develop problem-solving skills without a base of knowledge. How can one analyze a piece of literature, she asks, without knowing any vocabulary? Can students solve math problems without being able to multiply and divide?

“Obviously it’s an improvement, but I think that it is inadequate given the health care crisis that we’re in,” said Representative Ro Khanna, a progressive Democrat from California who favors the single-payer, government-run system called Medicare for All that has been embraced by Senator Bernie Sanders, independent of Vermont, and the Democratic left.

“We’re in a national health care crisis,” Mr. Khanna said. “Fifteen million people just lost private health insurance. This would be the time for the government to say, at the very least, for those 15 million that we ought to put them on Medicare.”

The stimulus bill would make upper-middle-income Americans newly eligible for financial help to buy plans on the federal marketplaces, and the premiums for those plans would cost no more than 8.5 percent of an individual’s modified adjusted gross income. It would also increase subsidies for lower-income enrollees.

Just when Mr. Biden or Democrats would put forth such a plan remains unclear, and passage in an evenly divided Senate would be an uphill struggle. White House officials have said Mr. Biden wants to get past the coronavirus relief bill before laying out a more comprehensive domestic policy agenda.

The Affordable Care Act is near and dear to Mr. Biden, who memorably used an expletive to describe it as a big deal when he was vice president and President Barack Obama signed it into law in 2010. It has expanded coverage to more than 20 million Americans, cutting the uninsured rate to 10.9 percent in 2019 from 17.8 percent in 2010.

Even so, some 30 million Americans were uninsured between January and June 2020, according to the latest figures available from the National Health Interview Survey. The problem has only grown worse during the coronavirus pandemic, when thousands if not millions of Americans lost insurance because they lost their jobs.

Mr. Biden made clear when he was running for the White House that he did not favor Medicare for All, but instead wanted to strengthen and expand the Affordable Care Act. The bill that is expected to reach his desk in time for a prime-time Oval Office address on Thursday night would do that. The changes to the health law would cover 1.3 million more Americans and cost about $34 billion, according to the Congressional Budget Office.

Republicans have always said that their plan was to repeal and replace the health law, but after 10 years they have yet to come up with a replacement. Mr. Ayres said his firm is working on “coming up with some alternative health care message” that does not involve “simply throwing everybody into a government-run health care problem.”

Yet polls show that the idea of a government-run program is gaining traction with voters. In September, the Pew Research Center reported that over the previous year, there had been an increase, especially among Democrats, in the share of Americans who say health insurance should be provided by a single national program run by the government.

“I would argue there is more momentum for Medicare expansion given the pandemic and the experience people are having,” said Mr. Khanna, the California congressman. “They bought time, but I think at some point there will be a debate on a permanent fix.”

WASHINGTON — President Biden’s $1.9 trillion coronavirus relief bill will fulfill one of his central campaign promises, to fill the holes in the Affordable Care Act and make health insurance affordable for more than a million middle-class Americans who could not afford insurance under the original law.

Under the changes, the signature domestic achievement of the Obama administration will reach middle-income families who have been discouraged from buying health plans on the federal marketplace because they come with high premiums and little or no help from the government.

“For people that are eligible but not buying insurance it’s a financial issue, and so upping the subsidies is going to make the price point come down,” said Ezekiel Emanuel, a health policy expert and professor at the University of Pennsylvania who advised Mr. Biden during his transition.

But because those provisions last only two years, the relief bill almost guarantees that health care will be front and center in the 2022 midterm elections, when Republicans will attack the measure as a wasteful expansion of a health law they have long hated. Meantime, some liberal Democrats may complain that the changes only prove that a patchwork approach to health care coverage will never work.

The Affordable Care Act is near and dear to Mr. Biden, who memorably used an expletive to describe it as a big deal when he was vice president and President Barack Obama signed it into law in 2010. It has expanded coverage to more than 20 million Americans, cutting the uninsured rate to 10.9 percent in 2019 from 17.8 percent in 2010.

The poll found that 36 percent of Americans, and 54 percent of Democrats, favored a single national program. When asked if the government had a responsibility to provide health insurance, either through a single national program or a mix of public and private programs, 63 percent of Americans and 88 percent of Democrats said yes.

Just when Mr. Biden or Democrats would put forth such a plan remains unclear, and passage in an evenly divided Senate would be an uphill struggle. White House officials have said Mr. Biden wants to get past the coronavirus relief bill before laying out a more comprehensive domestic policy agenda.

Republicans have always said that their plan was to repeal and replace the health law, but after 10 years they have yet to come up with a replacement. Mr. Ayres said his firm is working on “coming up with some alternative health care message” that does not involve “simply throwing everybody into a government-run health care problem.”

In January, he ordered the Affordable Care Act’s health insurance marketplaces reopened to give people throttled by the pandemic economy a new chance to obtain coverage.

Yet polls show that the idea of a government-run program is gaining traction with voters. In September, the Pew Research Center reported that over the previous year, there had been an increase, especially among Democrats, in the share of Americans who say health insurance should be provided by a single national program run by the government.

With its expanded subsidies for health plans under the Affordable Care Act, the coronavirus relief bill makes insurance more affordable, and puts health care on the ballot in 2022.

cludes rich new incentives to entice the few holdout states — including Texas, Georgia and Florida — to finally expand Medicaid to those with too much money to qualify for the federal health program for the poor, but too little to afford private covera

“Biden promised voters a public option, and it is a promise he has to keep,” said Waleed Shahid, a spokesman for Justice Democrats, the liberal group that helped elect Representative Alexandria Ocasio-Cortez and other progressive Democrats. Of the stimulus bill, he said, “I don’t think anyone thinks this is Biden’s health care plan.”

“I think that argument has been fought and lost,” said Whit Ayres, a Republican pollster, conceding that the repeal efforts are over, at least for now, with Democrats in charge of the White House and both houses of Congress.

"Oh," he mentioned at one point, "and there are no private schools in Finland."

Only a small number of independent schools exist in Finland, and even they are all publicly financed. None is allowed to charge tuition fees. There are no private universities, either. This means that practically every person in Finland attends public school, whether for pre-K or a Ph.D.

From his point of view, Americans are consistently obsessed with certain questions: How can you keep track of students' performance if you don't test them constantly? How can you improve teaching if you have no accountability for bad teachers or merit pay for good teachers? How do you foster competition and engage the private sector? How do you provide school choice?

Finland has no standardized tests. The only exception is what's called the National Matriculation Exam, which everyone takes at the end of a voluntary upper-secondary school, roughly the equivalent of American high school.

Instead, the public school system's teachers are trained to assess children in classrooms using independent tests they create themselves. All children receive a report card at the end of each semester, but these reports are based on individualized grading by each teacher.

"There's no word for accountability in Finnish," he later told an audience at the Teachers College of Columbia University. "Accountability is something that is left when responsibility has been subtracted."

what matters is that in Finland all teachers and administrators are given prestige, decent pay, and a lot of responsibility. A master's degree is required to enter the profession, and teacher training programs are among the most selective professional schools in the country. If a teacher is bad, it is the principal's responsibility to notice and deal with it.

"Real winners do not compete." It's hard to think of a more un-American idea, but when it comes to education, Finland's success shows that the Finnish attitude might have merits. There are no lists of best schools or teachers in Finland. The main driver of education policy is not competition between teachers and between schools, but cooperation.

Finally, in Finland, school choice is noticeably not a priority, nor is engaging the private sector at all.

Decades ago, when the Finnish school system was badly in need of reform, the goal of the program that Finland instituted, resulting in so much success today, was never excellence. It was equity.

Since the 1980s, the main driver of Finnish education policy has been the idea that every child should have exactly the same opportunity to learn, regardless of family background, income, or geographic location. Education has been seen first and foremost not as a way to produce star performers, but as an instrument to even out social inequality.

this means that schools should be healthy, safe environments for children. This starts with the basics. Finland offers all pupils free school meals, easy access to health care, psychological counseling, and individualized student guidance.

In fact, since academic excellence wasn't a particular priority on the Finnish to-do list, when Finland's students scored so high on the first PISA survey in 2001, many Finns thought the results must be a mistake. But subsequent PISA tests confirmed that Finland -- unlike, say, very similar countries such as Norway -- was producing academic excellence through its particular policy focus on equity.

the number of foreign-born residents in Finland doubled during the decade leading up to 2010, and the country didn't lose its edge in education. Immigrants tended to concentrate in certain areas, causing some schools to become much more mixed than others, yet there has not been much change in the remarkable lack of variation between Finnish schools in the PISA surveys across the same period.

Like Finland, Norway is small and not especially diverse overall, but unlike Finland it has taken an approach to education that is more American than Finnish. The result? Mediocre performance in the PISA survey. Educational policy, Abrams suggests, is probably more important to the success of a country's school system than the nation's size or ethnic makeup.

there were 18 states in the U.S. in 2010 with an identical or significantly smaller percentage of foreign-born residents than Finland

the goal of educational policy in the U.S. -- as articulated by most everyone from President Obama on down -- is to preserve American competitiveness by doing the same thing. Finland's experience suggests that to win at that game, a country has to prepare not just some of its population well, but all of its population well, for the new economy.

Finland's experience shows that it is possible to achieve excellence by focusing not on competition, but on cooperation, and not on choice, but on equity

The problem facing education in America isn't the ethnic diversity of the population but the economic inequality of society, and this is precisely the problem that Finnish education reform addressed. More equity at home might just be what America needs to be more competitive abroad.

An ardent empiricist, Diderot also took pride in questioning his own beliefs. In “Rameau’s Nephew,” Diderot gave life to a character who assailed the author’s deep-rooted humanism. One of the most memorable eccentrics in all of literature, the hedonistic protagonist preached the beauty of evil, the joys of social parasitism and the right to be a self-seeking individual.

Indeed, just a year later, a Virginia judge ruled that police could force a person to unlock his own iPhone with his fingerprint. And this February, a federal judge in Los Angeles signed a search warrant that compelled a 29-year-old woman to do the same.

“Unlocking devices through biometric systems generally won’t raise Fifth Amendment issues.”

“When you put your fingerprint on the phone, you’re actually communicating something,” Gidari said. “You’re saying, ‘Hi, it’s me. Please open up.’”

With that nuanced a system, it’s unclear how police could force a user to retrieve information from a phone, even with a legal court order in hand.

It’s possible that a more advanced system, however, would be able to pick up even a slight tremor of fear and lock up.

a strong, memorized password remains the only foolproof way to get away with invoking the Fifth and keep your phone locked.

In the course of his five-and-a-half-year presidency, beginning in 1969, Nixon launched and managed five successive and overlapping wars — against the anti-Vietnam War movement, the news media, the Democrats, the justice system and, finally, against history itself.

All reflected a mind-set and a pattern of behavior that were uniquely and pervasively Nixon’s: a willingness to disregard the law for political advantage, and a quest for dirt and secrets about his opponents as an organizing principle of his presidency.

Long before the Watergate break-in, gumshoeing, burglary, wiretapping and political sabotage had become a way of life in the Nixon White House.

What was Watergate? It was Nixon’s five wars.

In 1970, he approved the top-secret Huston Plan, authorizing the CIA, the FBI and military intelligence units to intensify electronic surveillance of individuals identified as “domestic security threats.” The plan called for, among other things, intercepting mail and lifting restrictions on “surreptitious entry” — that is, break-ins or “black bag jobs.”

On June 17, 1971 — exactly one year before the Watergate break-in — Nixon met in the Oval Office with his chief of staff, H.R. “Bob” Haldeman, and national security adviser Henry Kissinger. At issue was a file about former president Lyndon Johnson’s handling of the 1968 bombing halt in Vietnam.

“You can blackmail Johnson on this stuff, and it might be worth doing,” Haldeman said, according to the tape of the meeting. “Yeah,” Kissinger said, “but Bob and I have been trying to put the damn thing together for three years.” They wanted the complete story of Johnson’s actions.

“Huston swears to God there’s a file on it at Brookings,” Haldeman said. “Bob,” Nixon said, “now you remember Huston’s plan? Implement it. . . . I mean, I want it implemented on a thievery basis. God damn it, get in and get those files. Blow the safe and get it.”

Though Ellsberg was already under indictment and charged with espionage, the team headed by Hunt and Liddy broke into the office of his psychiatrist, seeking information that might smear Ellsberg and undermine his credibility in the antiwar movement.

“You can’t drop it, Bob,” Nixon told Haldeman on June 29, 1971. “You can’t let the Jew steal that stuff and get away with it. You understand?”

In a July 3, 1971, conversation with Haldeman, he said: “The government is full of Jews. Second, most Jews are disloyal. You know what I mean? You have a Garment [White House counsel Leonard Garment] and a Kissinger and, frankly, a Safire [presidential speechwriter William Safire], and, by God, they’re exceptions. But Bob, generally speaking, you can’t trust the bastards. They turn on you.”

In a tape from the Oval Office on Feb. 22, 1971, Nixon said, “In the short run, it would be so much easier, wouldn’t it, to run this war in a dictatorial way, kill all the reporters and carry on the war.”

John N. Mitchell, Nixon’s campaign manager and confidante, met with Liddy at the Justice Department in early 1972, when Mitchell was attorney general. Liddy presented a $1 million plan, code-named “Gemstone,” for spying and sabotage during the upcoming presidential campaign.

In Nixon’s third war, he took the weapons in place — the Plumbers, wiretapping and burglary — and deployed them against the Democrats challenging his reelection.

Operation Diamond would neutralize antiwar protesters with mugging squads and kidnapping teams; Operation Coal would funnel cash to Rep. Shirley Chisholm, a black congresswoman from Brooklyn seeking the Democratic presidential nomination, in an effort to sow racial and gender discord in the party;

Operation Opal would use electronic surveillance against various targets, including the headquarters of Democratic presidential candidates Edmund Muskie and George McGovern; Operation Sapphire would station prostitutes on a yacht, wired for sound, off Miami Beach during the Democratic National Convention.

Mitchell approved a $250,000 version, according to Jeb Magruder, the deputy campaign manager. It included intelligence-gathering on the Democrats through wiretaps and burglaries.

They discussed a secret $350,000 stash of cash kept in the White House, the possibility of using priests to help hide payments to the burglars, “washing” the money though Las Vegas or New York bookmakers, and empaneling a new grand jury so everyone could plead the Fifth Amendment or claim memory failure. Finally, they decided to send Mitchell on an emergency fundraising mission.

On Oct. 10, 1972, we wrote a story in The Post outlining the extensive sabotage and spying operations of the Nixon campaign and White House, particularly against Muskie, and stating that the Watergate burglary was not an isolated event. The story said that at least 50 operatives had been involved in the espionage and sabotage, many of them under the direction of a young California lawyer named Donald Segretti; several days later, we reported that Segretti had been hired by Dwight Chapin, Nixon’s appointments secretary. (The Senate Watergate committee later found more than 50 saboteurs, including 22 who were paid by Segretti.)

A favored dirty trick that caused havoc at campaign stops involved sweeping up the shoes that Muskie aides left in hotel hallways to be polished, and then depositing them in a dumpster.

In a memo to Haldeman and Mitchell dated April 12, 1972, Patrick Buchanan and another Nixon aide wrote: “Our primary objective, to prevent Senator Muskie from sweeping the early primaries, locking up the convention in April, and uniting the Democratic Party behind him for the fall, has been achieved.”

“I’d really like to get Kennedy taped,” Nixon told Haldeman in April 1971. According to Haldeman’s 1994 book, “The Haldeman Diaries,” the president also wanted to have Kennedy photographed in compromising situations and leak the images to the press.

On Sept. 8, 1971, Nixon ordered Ehrlichman to direct the Internal Revenue Service to investigate the tax returns of all the likely Democratic presidential candidates, as well as Kennedy. “Are we going after their tax returns?” Nixon asked. “You know what I mean? There’s a lot of gold in them thar hills.”

The arrest of the Watergate burglars set in motion Nixon’s fourth war, against the American system of justice. It was a war of lies and hush money, a conspiracy that became necessary to conceal the roles of top officials and to hide the president’s campaign of illegal espionage and political sabotage, including the covert operations that Mitchell described as “the White House horrors” during the Watergate hearings: the Huston Plan, the Plumbers, the Ellsberg break-in, Liddy’s Gemstone plan and the proposed break-in at Brookings.

In a June 23, 1972, tape recording, six days after the arrests at the Watergate, Haldeman warned Nixon that “on the investigation, you know, the Democratic break-in thing, we’re back in the problem area, because the FBI is not under control . . . their investigation is now leading into some productive areas, because they’ve been able to trace the money.”

Haldeman said Mitchell had come up with a plan for the CIA to claim that national security secrets would be compromised if the FBI did not halt its Watergate investigation.

Nixon approved the scheme and ordered Haldeman to call in CIA Director Richard Helms and his deputy Vernon Walters. “Play it tough,” the president directed. “That’s the way they play it, and that’s the way we are going to play it.”

On March 21, 1973, in one of the most memorable Watergate exchanges caught on tape, Nixon met with his counsel, John W. Dean, who since the break-in had been tasked with coordinating the coverup. “We’re being blackmailed” by Hunt and the burglars, Dean reported, and more people “are going to start perjuring themselves.” “How much money do you need?” Nixon asked.

“I would say these people are going to cost a million dollars over the next two years,” Dean replied. “And you could get it in cash,” the president said. “I, I know where it could be gotten. I mean, it’s not easy, but it could be done.”

Mitchell later denied approving the plan. He testified that he told Magruder: “We don’t need this. I’m tired of hearing it.” By his own account, he did not object on the grounds that the plan was illegal.

Nixon’s final war, waged even to this day by some former aides and historical revisionists, aims to play down the significance of Watergate and present it as a blip on the president’s record. Nixon lived for 20 years after his resignation and worked tirelessly to minimize the scandal.

In his 1978 memoir “RN,” Nixon addressed his role in Watergate: “My actions and omissions, while regrettable and possibly indefensible, were not impeachable.” Twelve years later, in his book “In the Arena,” he decried a dozen “myths” about Watergate and claimed that he was innocent of many of the charges made against him. One myth, he said, was that he ordered the payment of hush money to Hunt and others. Yet, the March 21, 1973, tape shows that he ordered Dean to get the money 12 times.

Even now, there are old Nixon hands and defenders who dismiss the importance of Watergate or claim that key questions remain unanswered.

By August, Nixon’s impending impeachment in the House was a certainty, and a group of Republicans led by Sen. Barry Goldwater banded together to declare his presidency over. “Too many lies, too many crimes,” Goldwater said. On Aug. 7, the group visited Nixon at the White House. How many votes would he have in a Senate trial? the president asked. “I took kind of a nose count today,” Goldwater replied, “and I couldn’t find more than four very firm votes, and those would be from older Southerners. Some are very worried about what’s been going on, and are undecided, and I’m one of them.”

In his last remarks about Watergate as a senator, 77-year-old Sam Ervin, a revered constitutionalist respected by both parties, posed a final question: “Why was Watergate?” The president and his aides, Ervin answered, had “a lust for political power.” That lust, he explained, “blinded them to ethical considerations and legal requirements; to Aristotle’s aphorism that the good of man must be the end of politics.”

Nixon had lost his moral authority as president. His secret tapes — and what they reveal — will probably be his most lasting legacy. On them, he is heard talking almost endlessly about what would be good for him, his place in history and, above all, his grudges, animosities and schemes for revenge. The dog that never seems to bark is any discussion of what is good and necessary for the well-being of the nation.

By the time he was forced to resign, Nixon had turned his White House, to a remarkable extent, into a criminal enterprise.

“Always remember,” he said, “others may hate you, but those who hate you don’t win unless you hate them, and then you destroy yourself.” His hatred had brought about his downfall. Nixon apparently grasped this insight, but it was too late. He had already destroyed himself.

Again and again, their most lasting effect has been to polarize opinion against them—and to empower the targets of their outrage. And this time, that target is a president hungering for any excuse to repress his opponents

Protesters may be up against something never before seen in American life: a president and an administration determined to seize on unrest to legitimate repression.

here’s what I have to offer from the right, amid the storms of the Trump era.

The more conservative protests are, the more radical they are.You want to scare Trump? Be orderly, polite, and visibly patriotic.

It’s beyond audacious that a candidate who publicly requested help from Russian espionage services against his opponent would claim the flag as his own. But Trump is trying. Don’t let him get away with it. Carry the flag. Open with the Pledge of Allegiance. Close by singing the Star Spangled Banner

Trump’s presidency is itself one long flag-burning, an attack on the principles and institutions of the American republic. That republic’s symbols are your symbols. You should cherish them and brandish them.

Don’t get sucked into the futile squabbling cul-de-sac of intersectionality and grievance politics. Look at this roster of speakers from the January 21 march. What is Angela Davis doing there? Where are the military women, the women police officers, the officeholders? If Planned Parenthood is on the stage, pro-life women should stand there, too. If you want somebody to speak for immigrants, invite somebody who’s in the country lawfully.

Here are a few useful tests:a) Could this demand be achieved by a law passed through Congress?b) Can I imagine my Rush Limbaugh listening brother-in-law agreeing with it?c) Can I tweet it?If so … good.

“Tone policing” has entered the left-of-center vocabulary as one of the worst possible things you can do or think. In fact, all effective political communication must carefully consider both tone and content

The classic military formula for success: concentrate superior force at a single point

Successful movements are built upon concrete single demands that can readily be translated into practical action: “Votes for women.” “End the draft.” “Overturn Roe v. Wade.” “Tougher punishments for drunk driving.

People can say “yes” to such specific demands for many different reasons.

So it should be for critics of President Trump. “Pass a law requiring the Treasury to release the President’s tax returns.” “An independent commission to investigate Russian meddling in the US election.” “Divest from the companies.” These are limited asks with broad appeal.

On the other hand, if you build a movement that lists those specific and limited goals along a vast and endlessly unfolding roster of others from “preserve Dodd Frank” to “save the oceans”—if you indulge the puckish anti-politics of “not usually a sign guy, but geez”—you will collapse into factionalism and futility.

if you are building a movement to protect American democracy from the authoritarianism of the Trump administration, you should remember that the goal is to gain allies among people who would not normally agree with you

the core demand of your movement should likewise be easy to explain and plausibly acceptable to that mainstream, stretching from Bernie voters to Romney donors.

Donald Trump has made clear that he wants to wage a Nixon-style culture war: cops against criminals, soldiers against pacifists, hard hats against hippies. Don’t be complicit. If you want to beat him, you have to reject his categories.

bodies in the street represent only potential power, not actual power. Even the largest rally must sooner or later disassemble and return home. What happens after that? The difference between Occupy Wall Street and the Tea Party was that only the second movement translated the energy and excitement of its early mass meetings into steady organizational work aimed at winning elections.

it is the steady and often tedious work of organization that sustains democracy—and can change the world. Protests are useful mostly to the extent that they mobilize people to participate in the follow-up meetings to realize the protest’s goals. Collect names and addresses. Form Facebook groups. Keep in touch. Don’t argue: recruit. Meet in real space as well as online. Serve cake. Make your presence felt on your local elected officials not just once, but day after day, week in, week out.

to succeed, you should be equally focused and persistent. And that requires above all: be motivated by hope, not outrage.

The outrage may get you started, but only hope keeps you going. Hope, as Vaclav Havel insisted, is an expression of the state of our minds, not a description of the state of the world. It powers you to undertake the daunting but essential mission: unlimited efforts for limited goals.

ou’re not trying to save the world. Just to pass one law. It doesn’t sound like much. It could be everything.

“Even students who are successful, getting straight A's, the highest accolades, by being students in school have fear and anxiety about maintaining these types of grades. You don't really get a sense of who you are, what your interests are, what your passions are [but are] being filled by other people's idea of what it means to be a human being.” As a typically neurotic straight-A student at a New York private school, I'd internalized the idea of education as a means to an end.

s a homeschooled teenager, without grades or teachers, I learned to love knowledge for its own sake, and, just as importantly to trust in my ability to attain it.

When I returned to traditional education as a high-school freshman at a New England boarding school known for its rigor and expectations of independence, I found myself surprisingly well-prepared for its demands. I knew how to manage my time effectively, how to approach the hours of homework as an opportunity to learn rather than a chore to be cleared away. I knew, too, how to defend my academic arguments, to not back down simply because another student disagreed with me about the causes of the American Civil War. In the absence of external signifiers of success, I had learned how to measure my own.

Economists have historically rejected what we call the “lump of labor” fallacy: the supposition that an increase in labor productivity inevitably reduces employment because there is only a finite amount of work to do. While intuitively appealing, this idea is demonstrably false.

Labor-saving technological change necessarily displaces workers performing certain tasks — that’s where the gains in productivity come from — but over the long run, it generates new products and services that raise national income and increase the overall demand for labor.

The multi-trillionfold decline in the cost of computing since the 1970s has created enormous incentives for employers to substitute increasingly cheap and capable computers for expensive labor.

Computers excel at “routine” tasks: organizing, storing, retrieving and manipulating information, or executing exactly defined physical movements in production processes. These tasks are most pervasive in middle-skill jobs

Logically, computerization has reduced the demand for these jobs, but it has boosted demand for workers who perform “nonroutine” tasks that complement the automated activities

At one end are so-called abstract tasks that require problem-solving, intuition, persuasion and creativity.

On the other end are so-called manual tasks, which require situational adaptability, visual and language recognition, and in-person interaction.

Computerization has therefore fostered a polarization of employment, with job growth concentrated in both the highest- and lowest-paid occupations, while jobs in the middle have declined.

overall employment rates have largely been unaffected in states and cities undergoing this rapid polarization.

So computerization is not reducing the quantity of jobs, but rather degrading the quality of jobs for a significant subset of workers. Demand for highly educated workers who excel in abstract tasks is robust, but the middle of the labor market, where the routine task-intensive jobs lie, is sagging.

Spurred by growing demand for workers performing abstract job tasks, the payoff for college and professional degrees has soared; despite its formidable price tag, higher education has perhaps never been a better investment.

The good news, however, is that middle-education, middle-wage jobs are not slated to disappear completely. While many middle-skill jobs are susceptible to automation, others demand a mixture of tasks that take advantage of human flexibility

we predict that the middle-skill jobs that survive will combine routine technical tasks with abstract and manual tasks in which workers have a comparative advantage — interpersonal interaction, adaptability and problem-solving.

this category includes numerous jobs for people in the skilled trades and repair: plumbers; builders; electricians; heating, ventilation and air-conditioning installers; automotive technicians; customer-service representatives; and even clerical workers who are required to do more than type and file

Lawrence F. Katz, a labor economist at Harvard, memorably called those who fruitfully combine the foundational skills of a high school education with specific vocational skills the “new artisans.”

The outlook for workers who haven’t finished college is uncertain, but not devoid of hope. There will be job opportunities in middle-skill jobs, but not in the traditional blue-collar production and white-collar office jobs of the past

we expect to see growing employment among the ranks of the “new artisans”: licensed practical nurses and medical assistants; teachers, tutors and learning guides at all educational levels; kitchen designers, construction supervisors and skilled tradespeople of every variety; expert repair and support technicians; and the many people who offer personal training and assistance, like physical therapists, personal trainers, coaches and guides

the very selection of what is a fact, or what is important as a fact, is itself based on an interpretation. You can't just separate fact and interpretation quite as simply as many people seem to think. I would love to see students get a little more experience in trying to write history, and trying to understand why historical interpretation changes over time.

I'm less interested in pedagogical approaches than the training of the teacher, the ability of the teacher, the knowledge of the teacher, and the teacher's ability to inspire students by conveying his or her own enthusiasm for the subject.

My experience as a teacher and as a student long ago, is that there is no substitute for a good teacher. I don't care what bells and whistles that you're using, it's the teacher in the classroom. That's why I'm a little skeptical about MOOCs, online education. I'm old-fashioned enough to believe that the presence of a teacher is actually critical to learning.

The first thing I would say is that we have to get away from the idea that any old person can teach history.

They wouldn't put him in a French course, or a physics course. The number-one thing is, you have to know history to actually teach it. That seems like an obvious point, but sometimes it's ignored in schools. Even more than that, I think it's important that people who are teaching history do have training in history. A lot of times people have education degrees, which have not actually provided them with a lot of training in the subject. They know a lot about methodology. [That’s] important, but as I say, the key thing is really to love the subject, to be able to convey that to your students, and if you can do that, I think you'll be a great teacher.

The other half agree that some jobs will disappear, but they are not convinced that new ones will take their place, even for some highly skilled workers. They fear a future of widespread unemployment, deep inequality and violent uprisings — particularly if policy makers and educational institutions don’t step in.

We’re going to have to come to grips with a long-term employment crisis and the fact that — strictly from an economic point of view, not a moral point of view — there are more and more ‘surplus humans.'  ”

“The degree of integration of A.I. into daily life will depend very much, as it does now, on wealth. The people whose personal digital devices are day-trading for them, and doing the grocery shopping and sending greeting cards on their behalf, are people who are living a different life than those who are worried about missing a day at one of their three jobs due to being sick, and losing the job and being unable to feed their children.”

“Only the best-educated humans will compete with machines. And education systems in the U.S. and much of the rest of the world are still sitting students in rows and columns, teaching them to keep quiet and memorize what is told to them, preparing them for life in a 20th century factory.”

“We hardly dwell on the fact that someone trying to pick a career path that is not likely to be automated will have a very hard time making that choice. X-ray technician? Outsourced already, and automation in progress. The race between automation and human work is won by automation.”

“Robotic sex partners will be commonplace. … The central question of 2025 will be: What are people for in a world that does not need their labor, and where only a minority are needed to guide the ‘bot-based economy?'  ”

“Employment will be mostly very skilled labor — and even those jobs will be continuously whittled away by increasingly sophisticated machines. Live, human salespeople, nurses, doctors, actors will be symbols of luxury, the silk of human interaction as opposed to the polyester of simulated human contact.”

The biggest exception will be jobs that depend upon empathy as a core capacity — schoolteacher, personal service worker, nurse. These jobs are often those traditionally performed by women. One of the bigger social questions of the mid-late 2020s will be the role of men in this world.”

Ong’s great scholarly focus was the transition of human society from orality to literacy: from sharing stories and ideas through spoken language alone, to sharing them through writing, text, and printed media

orality treats words as sound and action, only; that it emphasizes memory and redundancy; that it stays close to the “human lifeworld.”

we chat, we type, we text. One of the key attributes of orality is its instantaneousness: There’s no delay between utterances

To describe oral communication that was filtered through high technologies like radio and TV—technologies that could not exist without literacy—he coined the term secondary orality. To Ong, secondary orality was one of the great media phenomena of the 20th century.

In literate cultures, on the other hand, words are something you look up; language can stray more abstractly from objects; and speech, freed from memorable epithets like “the wine-dark sea,” can become more analytic.

How do you describe this odd mix of registers: literate culture that has all the ephemerality of oral culture? During his life, Ong suggested a new term, secondary literacy. I’ve also seen it referred to as a hybrid literacy

on Twitter, people say things that they think of as ephemeral and chatty. Their utterances are then treated as unequivocal political statements by people outside the conversation. Because there’s a kind of sensationalistic value in interpreting someone’s chattiness in partisan terms, tweets “are taken up as magnum opi to be leapt upon and eviscerated, not only by ideological opponents or threatened employers but by in-network peers.”

Anthropologists who study digital spaces have diagnosed that a common problem of online communication is “context collapse.” This plays with the oral-literate distinction: When you speak face-to-face, you’re always judging what you’re saying by the reaction of the person you’re speaking to. But when you write (or make a video or a podcast) online, what you’re saying can go anywhere, get read by anyone, and suddenly your words are finding audiences you never imagined you were speaking to.

It’s not quite context collapse, because what’s collapsing aren’t audiences so much as expectations. Rather, it’s a collapse of speech-based expectations and print-based interpretations. It’s a consequence of the oral-literate hybrid that flourishes online. It’s conversation smoosh.

This tension also explains, to me, why the more visual social networks have stayed fun and vibrant even as the text-based ones have not. Vine, Pinterest, and Instagram don’t traffic in words, which can be reduced to identity-based magnum opi, but in images, which are a little harder to smoosh. Visual conversations have stayed chatty, in other words.

At some point early last year, the standard knock against Twitter—which had long ceased to be “I don’t want to know what someone’s eating for lunch”—became “I don’t want everyone to see what I have to say.

Whereas many conservatives in 1958 felt that the sensible thing to do was to put elite academic mathematicians in charge of the school curriculum, by 1978 the conservative thing to do was to restore the math curriculum to local control and emphasize tradition — to go “back to basics.”

The new math’s reception was fundamentally shaped by Americans’ trust in federal initiatives and elite experts, their demands for local control and their beliefs about the skills citizens needed to face the problems of the modern world. Today these same political concerns will ultimately determine the future of the Common Core.

The idea that the complex problems students would face required training in the flexible, creative mathematics of elite practitioners was replaced by claims that modern students needed grounding in memorization, militaristic discipline and rapid recall of arithmetic facts.

controversies are unlikely to be resolved, because there’s not one right approach to how we should train students to think.

This was a claim both about who controlled intellectual training and about what forms of mental discipline should be promoted

As long as learning math counts as learning to think, the fortunes of any math curriculum will almost certainly be closely tied to claims about what constitutes rigorous thought — and who gets to decide.

Cruz was ready to attack Trump over the real estate mogul's assertion that Cruz's Canadian birth (to a U.S. citizen mother) makes him vulnerable to accusations he is ineligible for the presidency,

As Cruz rebutted Trump for raising the issue -- effectively winning the moment -- Trump essentially held his hands up and said he's not the one who's concerned.

And he succeeded in keeping the question alive -- a loss, in and of itself, for Cruz. Trump asked: "If you become the nominee, who the hell knows if you can even serve in office?"

Trump and Cruz were at the center of the night's most memorable exchanges, but Florida Sen. Marco Rubio and New Jersey Gov. Chris Christie traded blows as well.

That Christie was even a target suggests he has a leg up on the two other governors in the race -- Ohio Gov.