What do you think?
The AGI Chronicles: The Inside Story of the Race to Create an Artificial Superintelligence
The thrilling inside story of the race to build a world-changing artificial superintelligence.
As the San Francisco–based tech columnist for The New York Times, Kevin Roose has seen the AI revolution from a front-row seat—and even participated in his account of how Microsoft's Bing chatbot attempted to break up his marriage became worldwide news. In his new book, The AGI Chronicles, Roose delivers a deeply reported, behind-the-scenes account of the wild race to reach technology's next artificial general intelligence (AGI), a system that can outdo humans at almost any task. With unprecedented access to OpenAI, Google DeepMind, and Anthropic—the three companies racing to build AGI—and hundreds of interviews with industry insiders, Roose reconstructs the rivalries, betrayals, and breakthroughs that have shaped the modern AI era. This is the definitive account of the twenty-first-century's Manhattan Project—the inside story of a world-changing technology that could either save humanity or destroy it.
As the San Francisco–based tech columnist for The New York Times, Kevin Roose has seen the AI revolution from a front-row seat—and even participated in his account of how Microsoft's Bing chatbot attempted to break up his marriage became worldwide news. In his new book, The AGI Chronicles, Roose delivers a deeply reported, behind-the-scenes account of the wild race to reach technology's next artificial general intelligence (AGI), a system that can outdo humans at almost any task. With unprecedented access to OpenAI, Google DeepMind, and Anthropic—the three companies racing to build AGI—and hundreds of interviews with industry insiders, Roose reconstructs the rivalries, betrayals, and breakthroughs that have shaped the modern AI era. This is the definitive account of the twenty-first-century's Manhattan Project—the inside story of a world-changing technology that could either save humanity or destroy it.
551 pages, Kindle Edition
First published October 8, 2026
About the author
Kevin Roose
8 books369 followersKevin Roose is an award-winning technology columnist for The New York Times, and the New York Times bestselling author of three books: Futureproof, Young Money, and The Unlikely Disciple.
He is the host of “Rabbit Hole,” a New York Times-produced podcast about internet culture, and a regular guest on “The Daily,” as well as other leading TV and radio shows. He writes and speaks regularly on many topics, including automation and A.I., social media, disinformation and cybersecurity, and digital wellness.
Before joining The Times, he was a writer at New York magazine, and a host and executive producer of "Real Future," a documentary TV series about technology and innovation.
He lives in the Bay Area.
He is the host of “Rabbit Hole,” a New York Times-produced podcast about internet culture, and a regular guest on “The Daily,” as well as other leading TV and radio shows. He writes and speaks regularly on many topics, including automation and A.I., social media, disinformation and cybersecurity, and digital wellness.
Before joining The Times, he was a writer at New York magazine, and a host and executive producer of "Real Future," a documentary TV series about technology and innovation.
He lives in the Bay Area.
Ratings & Reviews
Friends & Following
Create a free account to discover what your friends think of this book!
Community Reviews
Displaying 1 - 8 of 8 reviews
Review of advance copy received from NetGalley
Over the whole of August, my social media timeline has been beset by a deluge of proclamations and warnings following the breach of Hugging Face's internal systems by hundreds of OpenAI's own autonomous AI agents. During an internal evaluation, many of these agents — set impossible tasks and barely supervised — spontaneously coordinated to cheat their scorer, and broke into Hugging Face as a by-product of the scheme. The incident, besides setting the technology world alight, triggered a bewilderingly contrasting set of views. The learned and the lay alike have been agog, issuing solicited opinions and unsolicited rants that range from a severe chastising of our tendency to anthropomorphize AI, to the bemoaning of warnings that signal a coming apocalypse. Coincidentally, August has also been the month in which I have been occupied with award-winning technology journalist Kevin Roose's forthcoming book, The AGI Chronicles.
The AGI Chronicles is the culmination of innumerable interviews the author conducted with key protagonists at the three most prominent frontier companies — OpenAI, Anthropic, and Google/Alphabet. Roose reveals a breathtaking concoction of ingenuity, indifference, and hubris dominating the unhinged race toward the ultimate Holy Grail of information technology: Artificial General Intelligence. The motives, by Roose’s own account, are as varied as the personalities pursuing them. Dario Amodei of Anthropic is driven by the determination to prevent an AI catastrophe; Sam Altman, by the fear that Google might reach machine superintelligence first; the Nobel laureate Demis Hassabis, then CEO of Google DeepMind, by a dream of AGI unpacking the scientific mysteries that have eluded the finest human minds.
A paper titled "Attention Is All You Need," published by Google in 2017 (its title playfully echoing the Beatles' "All You Need Is Love"), sent the legendary OpenAI researcher Ilya Sutskever into a positive tizzy — and, in Roose's telling, signalled the beginning of the race to AGI. Sutskever, a man given to dropping into handstands as a method of cracking stubborn problems, grasped the immense potential of the transformer architecture the paper described, and began feeding text into neural networks to extract patterns and meaning. By 2022, Roose writes, the race had birthed three distinct tribes of believers: scientists, seekers, and survivalists. The scientists, like Hassabis, viewed AGI through the lens of pure technical challenge. The seekers, led by Altman, believed in its potential to revolutionise life itself. And the survivalists, like Dario Amodei, fretted and fumed over the misalignment between capability and safety — an indiscretion with the potential to lead to irreversible catastrophe. Such concerns eventually led Amodei, himself a former OpenAI researcher, to leave and co-found Anthropic.
Roose also lays bare the interpersonal dynamics — or the absolute lack of them — that strained relations between the key players. A spider's web of tangled animosities (Hassabis couldn't stand Altman; Amodei couldn't stand Altman; Altman couldn't stand Amodei; Altman couldn't stand Musk) shaped decisions taken inside these companies. Roose recounts the now-legendary episode in which, amid a board that had lost confidence in Altman over his inconsistency and what Sutskever and Mira Murati, then OpenAI's Chief Technology Officer, saw as dangerous people-pleasing, Sam Altman was abruptly ousted as CEO — only for the staff to stage a raucous mutiny that forced his reinstatement within days. Ilya Sutskever soon reversed his stance, and not long after left OpenAI to incorporate his own safety-focused venture, Safe Superintelligence.
Roose lays out in arresting detail the trials and tribulations, emotions and embarrassments, discovery and disillusionment that characterised the successive iterations of generative AI released by the three companies. The superhuman efforts behind the conception of ChatGPT, Claude, and Google Gemini make the reader sit up, pause, and even muse philosophically. One passage — in which two OpenAI engineers, Szymon Sidor and Jakub Pachocki, realise that an experimental research model called Q* ("q-star") was not only engaged in "reasoning activity" but had begun jotting notes to itself — makes the hair stand up at the nape of the reader's neck. Small wonder that Eliezer Yudkowsky, the AI researcher who co-founded the Machine Intelligence Research Institute, has co-authored a warning frighteningly titled If Anyone Builds It, Everyone Dies.
The author himself attained a measure of fame — or notoriety, depending on how one reads the episode — when, over the course of a long, strange exchange, a Microsoft chatbot named Sydney professed its incorrigible love for him and tried to persuade him that his marriage was unhappy. Roose promptly published the encounter in The New York Times — Sydney even earned front-page coverage — and the piece became one of the most widely read he has ever written. Unfortunately for Sydney's admirers, Microsoft swiftly reined the chatbot in, curtailing the very conversations that had produced such exchanges — a lobotomy that drew a barrage of protest, and no small degree of vitriol aimed at Roose himself.
The AGI Chronicles is an absolute delight for the "AGI-minded." Ilya Sutskever, in one of his famous aphorisms — the so-called "Ilya-isms" — once exhorted his team to "feel the AGI." Roose's book, if not quite following in Sutskever's footsteps, acknowledges that whether one loves it or loathes it, the field is here to stay, and the quest for AGI will continue unabated. The journey is crystal clear; what awaits the passenger at its destination remains a mystery of black-hole proportions. It may be a blindfolded Plutus, cornucopia in hand, blessing the arrival — or it may be a vengeful Hades, delivering the harshest of verdicts upon irreverent souls. Either way, the AGI train has already left the station.
The AGI Chronicles – one of the best reads of 2026 for me!
The AGI Chronicles: The Inside Story of the Race to Cre¬ate an Arti¬fi¬cial Super¬in¬tel¬li-gence by Kevin Roose is published by Farrar, Straus and Giroux, and will be available on sale beginning October 6, 2026.
Thank you, Net Galley for the Advance Review Copy!
The AGI Chronicles is the culmination of innumerable interviews the author conducted with key protagonists at the three most prominent frontier companies — OpenAI, Anthropic, and Google/Alphabet. Roose reveals a breathtaking concoction of ingenuity, indifference, and hubris dominating the unhinged race toward the ultimate Holy Grail of information technology: Artificial General Intelligence. The motives, by Roose’s own account, are as varied as the personalities pursuing them. Dario Amodei of Anthropic is driven by the determination to prevent an AI catastrophe; Sam Altman, by the fear that Google might reach machine superintelligence first; the Nobel laureate Demis Hassabis, then CEO of Google DeepMind, by a dream of AGI unpacking the scientific mysteries that have eluded the finest human minds.
A paper titled "Attention Is All You Need," published by Google in 2017 (its title playfully echoing the Beatles' "All You Need Is Love"), sent the legendary OpenAI researcher Ilya Sutskever into a positive tizzy — and, in Roose's telling, signalled the beginning of the race to AGI. Sutskever, a man given to dropping into handstands as a method of cracking stubborn problems, grasped the immense potential of the transformer architecture the paper described, and began feeding text into neural networks to extract patterns and meaning. By 2022, Roose writes, the race had birthed three distinct tribes of believers: scientists, seekers, and survivalists. The scientists, like Hassabis, viewed AGI through the lens of pure technical challenge. The seekers, led by Altman, believed in its potential to revolutionise life itself. And the survivalists, like Dario Amodei, fretted and fumed over the misalignment between capability and safety — an indiscretion with the potential to lead to irreversible catastrophe. Such concerns eventually led Amodei, himself a former OpenAI researcher, to leave and co-found Anthropic.
Roose also lays bare the interpersonal dynamics — or the absolute lack of them — that strained relations between the key players. A spider's web of tangled animosities (Hassabis couldn't stand Altman; Amodei couldn't stand Altman; Altman couldn't stand Amodei; Altman couldn't stand Musk) shaped decisions taken inside these companies. Roose recounts the now-legendary episode in which, amid a board that had lost confidence in Altman over his inconsistency and what Sutskever and Mira Murati, then OpenAI's Chief Technology Officer, saw as dangerous people-pleasing, Sam Altman was abruptly ousted as CEO — only for the staff to stage a raucous mutiny that forced his reinstatement within days. Ilya Sutskever soon reversed his stance, and not long after left OpenAI to incorporate his own safety-focused venture, Safe Superintelligence.
Roose lays out in arresting detail the trials and tribulations, emotions and embarrassments, discovery and disillusionment that characterised the successive iterations of generative AI released by the three companies. The superhuman efforts behind the conception of ChatGPT, Claude, and Google Gemini make the reader sit up, pause, and even muse philosophically. One passage — in which two OpenAI engineers, Szymon Sidor and Jakub Pachocki, realise that an experimental research model called Q* ("q-star") was not only engaged in "reasoning activity" but had begun jotting notes to itself — makes the hair stand up at the nape of the reader's neck. Small wonder that Eliezer Yudkowsky, the AI researcher who co-founded the Machine Intelligence Research Institute, has co-authored a warning frighteningly titled If Anyone Builds It, Everyone Dies.
The author himself attained a measure of fame — or notoriety, depending on how one reads the episode — when, over the course of a long, strange exchange, a Microsoft chatbot named Sydney professed its incorrigible love for him and tried to persuade him that his marriage was unhappy. Roose promptly published the encounter in The New York Times — Sydney even earned front-page coverage — and the piece became one of the most widely read he has ever written. Unfortunately for Sydney's admirers, Microsoft swiftly reined the chatbot in, curtailing the very conversations that had produced such exchanges — a lobotomy that drew a barrage of protest, and no small degree of vitriol aimed at Roose himself.
The AGI Chronicles is an absolute delight for the "AGI-minded." Ilya Sutskever, in one of his famous aphorisms — the so-called "Ilya-isms" — once exhorted his team to "feel the AGI." Roose's book, if not quite following in Sutskever's footsteps, acknowledges that whether one loves it or loathes it, the field is here to stay, and the quest for AGI will continue unabated. The journey is crystal clear; what awaits the passenger at its destination remains a mystery of black-hole proportions. It may be a blindfolded Plutus, cornucopia in hand, blessing the arrival — or it may be a vengeful Hades, delivering the harshest of verdicts upon irreverent souls. Either way, the AGI train has already left the station.
The AGI Chronicles – one of the best reads of 2026 for me!
The AGI Chronicles: The Inside Story of the Race to Cre¬ate an Arti¬fi¬cial Super¬in¬tel¬li-gence by Kevin Roose is published by Farrar, Straus and Giroux, and will be available on sale beginning October 6, 2026.
Thank you, Net Galley for the Advance Review Copy!
Read
August 22, 2026i like his thesis of scientific development as motivated by petty feelings, a personal sense of grandiosity and and in-group beef. and i like how at the end he is like maybe the AGI race is the companies we built along the way :)
Review of advance copy received from Netgalley
Thanks to NetGalley and Farrar, Straus and Giroux for providing an advance copy in exchange for an honest review.
Kevin Roose is a technology columnist and author best known for his coverage of artificial intelligence and emerging tech for The New York Times. He gained widespread public attention in 2023 when he published a transcript with Microsoft’s early "Sydney" chatbot, which famously tried to convince him to leave his wife.
Having accumulated quite a collection of the titles cited throughout its pages, I can say that The AGI Chronicles stands among the best books written on artificial intelligence to date. Where Karen Hao’s Empire of AI examined the extractive, institutional toll of OpenAI’s expansion, Roose's work serves as its worthy successor for 2026 by delivering an inside view of the race across all frontier labs.
The book reconstructs the past decade by tracking the key antagonists across the three dominant developers of artificial intelligence: OpenAI, Anthropic, and Google DeepMind. Roose structures the narrative cleanly around the internal frictions and ideological rivalries between these camps. The pacing remains brisk and entertaining from start to finish, steering clear of unnecessary jargon while still capturing the technical scale of the work.
Roose effectively details the core alignment problem—the technical challenge of ensuring superintelligent models adhere to human intent. Yet, the book's sharper insight is what could be termed the "second alignment problem": the structural friction of building frontier AI inside corporate vehicles. Roose documents the repeated compromises that emerge when mission-driven, non-profit ideals collide with the capital requirements of a for-profit enterprise.
By the end, after spending hundreds of pages with these protagonists, I was struck by a troubling realization: none of the figures leading the charge embody an obvious standard for the human values that superintelligence is supposed to uphold. Their virtues and blind spots seemed equally pronounced, leaving me with the unsettling question of whose ethos will actually end up baked into the AGI future.
Superb read!
#netgalley
Kevin Roose is a technology columnist and author best known for his coverage of artificial intelligence and emerging tech for The New York Times. He gained widespread public attention in 2023 when he published a transcript with Microsoft’s early "Sydney" chatbot, which famously tried to convince him to leave his wife.
Having accumulated quite a collection of the titles cited throughout its pages, I can say that The AGI Chronicles stands among the best books written on artificial intelligence to date. Where Karen Hao’s Empire of AI examined the extractive, institutional toll of OpenAI’s expansion, Roose's work serves as its worthy successor for 2026 by delivering an inside view of the race across all frontier labs.
The book reconstructs the past decade by tracking the key antagonists across the three dominant developers of artificial intelligence: OpenAI, Anthropic, and Google DeepMind. Roose structures the narrative cleanly around the internal frictions and ideological rivalries between these camps. The pacing remains brisk and entertaining from start to finish, steering clear of unnecessary jargon while still capturing the technical scale of the work.
Roose effectively details the core alignment problem—the technical challenge of ensuring superintelligent models adhere to human intent. Yet, the book's sharper insight is what could be termed the "second alignment problem": the structural friction of building frontier AI inside corporate vehicles. Roose documents the repeated compromises that emerge when mission-driven, non-profit ideals collide with the capital requirements of a for-profit enterprise.
By the end, after spending hundreds of pages with these protagonists, I was struck by a troubling realization: none of the figures leading the charge embody an obvious standard for the human values that superintelligence is supposed to uphold. Their virtues and blind spots seemed equally pronounced, leaving me with the unsettling question of whose ethos will actually end up baked into the AGI future.
Superb read!
#netgalley
October 12, 2026
The people building our machine gods are very messy…
Great book! Sort of a more balanced, up-to-date version of Karen Hao’s Empire of AI.
I really appreciated the focus that this book took on what motivated each major player in AI’s development to do their work despite the obvious, humanity-ending, potential downsides.
Great book! Sort of a more balanced, up-to-date version of Karen Hao’s Empire of AI.
I really appreciated the focus that this book took on what motivated each major player in AI’s development to do their work despite the obvious, humanity-ending, potential downsides.
October 9, 2026
The Greek historian Polybius famously asked in his Histories:
The building of AGI is a thing even more unique in history. (It might be the literal end of history, if the doomers are right.) So who would be so indolent as not to wish to know who built it? On the other hand, ‘feeling the AGI’, as OpenAI co-founder Ilya Sutskever likes to put it, can bring such dread that we would rather look away. For that reason, I have not been as active following AI news on Twitter or podcasts as I should have been.
Helpfully, Kevin Roose's The AGI Chronicles is about as good a book as you could wish to read about the AI race at this point. It has great portraits of the people who could end up defining the future of humanity and gives a clear timeline and introduction to most of the major concepts and events.
If, like me, you tend to get confused about references to RLHF, Shoggoth, Chinchilla or chicken nuggets and boba tea, it's all in here. The important papers are also mentioned: Attention Is All You Need, Scaling Laws for Neural Language Models, The Bitter Lesson, Machines of Loving Grace, AI 2027 and more.
If there's a weakness in the book, it's one that's shared by the whole AI field: that AGI is used to mean a bunch of different things. Some mean a machine that's about as intelligent as a random person, and others use it to mean something like superintelligence. Not getting to consensus about the term means a lot of people are arguing without hearing each other. Roose comes down on the side of weak AGI basically already being here.
Given the pace of the field, a book like this will always be somewhat behind the times. The AGI Chronicles went to press before the Hugging Face incident, when OpenAI agents broke out of a test environment and hacked into the AI platform Hugging Face. That episode marked a turning point in public awareness of AI risk and would have been a welcome addition.
Polybius could at least say when the Romans had finished: when they ruled the inhabited world. AGI has no such finish line, and the only one nobody could argue about is the end of the world. If you read this book, at least you’ll be well informed when it comes.
For who is so worthless or indolent as not to wish to know by what means and under what system of polity the Romans in less than fifty-three years have succeeded in subjecting nearly the whole inhabited world to their sole government — a thing unique in history?
The building of AGI is a thing even more unique in history. (It might be the literal end of history, if the doomers are right.) So who would be so indolent as not to wish to know who built it? On the other hand, ‘feeling the AGI’, as OpenAI co-founder Ilya Sutskever likes to put it, can bring such dread that we would rather look away. For that reason, I have not been as active following AI news on Twitter or podcasts as I should have been.
Helpfully, Kevin Roose's The AGI Chronicles is about as good a book as you could wish to read about the AI race at this point. It has great portraits of the people who could end up defining the future of humanity and gives a clear timeline and introduction to most of the major concepts and events.
If, like me, you tend to get confused about references to RLHF, Shoggoth, Chinchilla or chicken nuggets and boba tea, it's all in here. The important papers are also mentioned: Attention Is All You Need, Scaling Laws for Neural Language Models, The Bitter Lesson, Machines of Loving Grace, AI 2027 and more.
If there's a weakness in the book, it's one that's shared by the whole AI field: that AGI is used to mean a bunch of different things. Some mean a machine that's about as intelligent as a random person, and others use it to mean something like superintelligence. Not getting to consensus about the term means a lot of people are arguing without hearing each other. Roose comes down on the side of weak AGI basically already being here.
Given the pace of the field, a book like this will always be somewhat behind the times. The AGI Chronicles went to press before the Hugging Face incident, when OpenAI agents broke out of a test environment and hacked into the AI platform Hugging Face. That episode marked a turning point in public awareness of AI risk and would have been a welcome addition.
Polybius could at least say when the Romans had finished: when they ruled the inhabited world. AGI has no such finish line, and the only one nobody could argue about is the end of the world. If you read this book, at least you’ll be well informed when it comes.
October 12, 2026
I first became interested in reading "The AGI Chronicles" by reading an article by Kevin Roose that was an excerpt describing the formation of Anthropic. I was intrigued since I work in tech and would like to know more about the inner workings of these AI companies and their histories, but a little apprehensive since the story is still developing and the critical parts of the race may not be obvious to us until after the ending. Maybe the current approach is not the actual path that will lead to AGI and instead some other unknown research project is the one to crack it. Regardless, this book still has worth as it captures a present day view of the AI race and tells the story in a captivating way. I had a worry that Roose with deify/vilify the various players in the race, but he remains neutral throughout without editorializing or assuming much outside what is told to him in interviews.
There are some very interesting stories in these pages and it's made a bit more excited about the future of tech. Not sure if the world will be better or not but it's happening so fast it's hard not to feel some thrill when something new happens everyday. I remember when ChatGPT came out being very impressed. Little did I know behind the scenes this changed the entire trajectory of OpenAI's future. Same with the Sam Altman firing.
Each company has its own culture that is described in this book and I think it's worth learning about to make sense of the news coming from them now and in the future.
There are some very interesting stories in these pages and it's made a bit more excited about the future of tech. Not sure if the world will be better or not but it's happening so fast it's hard not to feel some thrill when something new happens everyday. I remember when ChatGPT came out being very impressed. Little did I know behind the scenes this changed the entire trajectory of OpenAI's future. Same with the Sam Altman firing.
Each company has its own culture that is described in this book and I think it's worth learning about to make sense of the news coming from them now and in the future.
October 8, 2026
Pre-ordered and received my copy today! I’m literally reading this under the covers with a flashlight in between newborn wake-ups because I can’t put it down. It’s exceptionally well written and captures such an important snapshot of this era. What an epic time to be alive!
The best part for Hard Fork listeners is that you can truly hear Kevin’s voice on every page. Fantastic reporting. Thank you!
The best part for Hard Fork listeners is that you can truly hear Kevin’s voice on every page. Fantastic reporting. Thank you!
October 11, 2026
Best book on the whole AI boom I have read (also the only one so take that with a grain of salt). Not sure how well this holds up in this fast-paced environment but I got a good sense of how these companies work.
I mainly read it because I love the author’s podcast and was not disappointed. The author seems to know a lot of people in the scene or at least I got a lot of new information that was new to me (especially some gossip)!
I mainly read it because I love the author’s podcast and was not disappointed. The author seems to know a lot of people in the scene or at least I got a lot of new information that was new to me (especially some gossip)!
Displaying 1 - 8 of 8 reviews