Superminds: The Surprising Power of People and Computers Thinking Together

by Thomas W. Malone

From the founding director of the MIT Center for Collective Intelligence comes a fascinating look at the remarkable capacity for intelligence exhibited by groups of people and computers working together.

If you're like most people, you probably believe that humans are the most intelligent animals on our planet. But there's another kind of entity that can be far smarter: groups of people. In this groundbreaking book, Thomas Malone, the founding director of the MIT Center for Collective Intelligence, shows how groups of people working together in superminds -- like hierarchies, markets, democracies, and communities -- have been responsible for almost all human achievements in business, government, science, and beyond. And these collectively intelligent human groups are about to get much smarter.

Using dozens of striking examples and case studies, Malone shows how computers can help create more intelligent superminds simply by connecting humans to one another in a variety of rich, new ways. And although it will probably happen more gradually than many people expect, artificially intelligent computers will amplify the power of these superminds by doing increasingly complex kinds of thinking. Together, these changes will have far-reaching implications for everything from the way we buy groceries and plan business strategies to how we respond to climate change, and even for democracy itself. By understanding how these collectively intelligent groups work, we can learn how to harness their genius to achieve our human goals.

Drawing on cutting-edge science and insights from a remarkable range of disciplines, Superminds articulates a bold -- and utterly fascinating -- picture of the future that will change the ways you work and live, both with other people and with computers.

Genres: Nonfiction, Artificial Intelligence, Science, Technology, Business, Philosophy, Psychology

384 pages, Hardcover

Published May 15, 2018

Reviews

[Adnan Khaleel · Rating 1 out of 5]

This book probably made sense 3 decades ago, but nothing new is really discussed here. The idea of super minds besides be a abstract construct, lacks any tangible practical use. I’ve wasted my time on this book, I’d advise you not do the same.

[Terralynn Forsyth · Rating 2 out of 5]

Pretty disappointing - felt like the explanations of "superminds" and collective intelligence were well laid out, but the arguments for why thinking in this way for the "future of work" were unclear. It was a pretty general and unoriginal overview of an highly discussed topic. Some interesting concepts, such as "integrated information theory", for understanding, measuring, and assessing consciousness are introduced, but practicality is missing from this book.

[Kyle · Rating 3 out of 5]

Reasonably well written, it is both too complicated to introduce someone to the topic and too broad to move the subject forward.

The author has a particular view of intelligence and emergent thinking, which is fine, but it did not seem based on research. It seemed largely a stealth advertisement for his think tank.

[Jim Crocker · Rating 5 out of 5]

This one should get you thinking. No it's not about some kind of computer mind-meld. It's a serious discussion about the "intelligence" of groups of people. Here's a quote:

"Groups have scientifically measurable properties. Psychologists can measure intelligence of groups. Some groups are smarter than others and we can understand why."

All I can say is that I "knew" it all along. I'll be reading and re-reading this one for a while.

[Mark Fallon · Rating 4 out of 5]

We know that we're stronger together. research has proven that we're also smarter together. Now add the power of computer networking and AI.....

[João · Rating 4 out of 5]

Is a group conscious? How can we improve collective intelligence? The author offers non intuitive and tought provoking answers. Nice job.

[Jerrod Carter · Rating 2 out of 5]

This felt like a book that was published under the "publish or perish" model. An MIT prof throwing in as many references to other MIT prof's studies as he could. Really, maybe the only takeaway I got from this book was that we should look at making computers part of groups, not just as personal assistants. Just as they can increase our individual productivity, they could also increase a group's productivity and decision making by including their insights into a decision making process. Probably could have made that point in a good essay a lot better than in a book.

[Art Tucker · Rating 3 out of 5]

In a post-pandemic, post-Jan 6th, post-ChatGPT world, some of the speculation and references in this book are increasingly dated. It was difficult to get through its first quarter, but picked up pace after that. It sometimes feels like the political views were naively optimistic. That said, it makes some very worthwhile observations about what it calls "Superminds" (interconnected networks of entities working towards goals). I would argue that, at least for the next decade or so, it's a worthwhile read.

[Steve · Rating 1 out of 5]

Apparently I am not the target audience. It must be better suited for the fields of organizational dynamics, management, marketing and perhaps sociology. If you have experience in computer science, machine learning or are in the fields of engineering or biomedical engineering you will not find this book thought provoking or an interesting read.

I was definitely disappointed.

[Douglas Summers-Stay · Rating 4 out of 5]

This brings together a lot of threads about how groups of people can work together to solve problems. The author was the first to try to measure the intelligence quotient of a group. I have issues with how they did it, though-- instead of using an established intelligence test, they made up their own tests, which seems like it would be less useful information. Also, the data that shows that the intelligence of the group depends on the social perceptiveness of the group (and because of strong correlation, the number of women in the group) is sketchy: it looks like random noise with a trend-line drawn through it to me. I would bet it won't replicate.
The writing could have been quite a bit denser and I would have been okay with it, even though this isn't really my field. I did like his point that building an AI is like building an intelligent community, only moreso because the individual units are so dumb.
The thought I had when reading was that the difference between political systems isn't fundamentally about different goals, it's about different choice-making architectures. Democracy is about making choices by ballot. Libertarianism is about making choices by market. Fascism is about making decision by hierarchy. Communitarianism (?) is about making choices by community agreement-- I've been told that Quakers decide by unanimous voice. Communism-- I'm not sure exactly, but it's still a different way of choosing how resources get allocated, and how it is decided.

[Darnell · Rating 2 out of 5]

There are some insights here, but they're buried under a lot of generic chapters. The book's formula is to call a system (like a company or market) a supermind, make some basic observations, then suggest that technology will probably have an impact on it. The number of chapters that actually dealt with computer/human interactions was very small.

I felt many of the takes on other subjects weren't great either. He is bizarrely optimistic about China's social credit score system, and has an enthusiasm for microtask sysems like Mechanical Turk that I can't generate.

Also, not a single chapter about cooperative human/computer chess, which was honestly the main thing I wanted.

[JB Lynn · Rating 5 out of 5]

A highly interesting look at how groups think and work together to problem-solve. Probably my favorite sections of the book were all the different real-world scenarios of some interesting experiments and studies that have been going on in the effort to study and improve upon AI technology and how that's been incorporated into our lives.

There's also a lengthy references section at the end with plenty of options for anyone who might want to read more on a particular idea/study mentioned in the book.

[Peter Skurr · Rating 3 out of 5]

Discusses collective intelligence, or group minds. Including interactions with computers. Started off interesting, but soon ground to a halt. A found the model incomplete and vague, and there were elements i simply wanted to reject out of hand. Should have put it down 100 pages ago.

[Lorenzo Barberis Canonico · Rating 4 out of 5]

Amazing! Full of examples that will blow your mind, while also providing a really robust framework to think about collective intelligence across a variety of domains

[Ken Hamner · Rating 4 out of 5]

Good overall summary of the inevitable and accelerating ability of people to leverage machine learning and artificial intelligence in the advancement of knowledge and ideas. Pretty good book.

[Doug · Rating 3 out of 5]

A well researched and insightful book - just not an enjoyable read.

[D.L. Morrese · Rating 3 out of 5]

The idea of a group mind isn't new. A business, an army, a musical band, a nation... Many such groups can be viewed as unique entities that possesses information, goals, and attitudes that help define them but which are not necessarily the same as those of their constituent members. (For example, ask your heart its opinion of your last bacon cheeseburger.) There is also nothing new about intelligent behavior emerging from a mixed group of humans and non-humans. (A sheriff, a bloodhound, and a map can find an escaped prisoner far better than any of them alone.) And that's pretty much my take on this book. There's nothing new here, no insights, no new information, no fresh perspective.... It's an interesting subject, and if I had chosen to write about it, I'd probably have used the word 'emergent' a lot, and talked considerably more about how difficult it can be to pin down precisely what 'mind' may be generating seemingly intelligent behavior. But, I didn't, and I'm unlikely to. No one would pay me for the effort, and it doesn't sound like enough fun to do it without being paid.... Which makes me wonder why I'm bothering to write this review. Another good question.

[Angel Grimalt · Rating 4 out of 5]

I read sections of Prof. Malone's book initially and have now completed it from cover to cover. I've come to believe that the future of AI and humans is profoundly linked to the concept of "superminds"—the idea that we operate collectively in increasingly interconnected and synchronous ways towards a greater goal. One key insight from Malone's thesis is the potential for these "superminds" to augment collective intelligence, not just in terms of computational power but in enhancing the ability to collaborate and innovate. This aligns with broader discussions in the field of AI and collaborative systems, where the synergy between human competencies and AI's abilities can lead to unprecedented solutions in science, healthcare, and governance. Regarding the negative reviews, I understand that this approach may seem rather general, leaning on philosophical thinking. However, it is the way we think and the concepts we use that can make a significant difference in our decision-making and actions.

[Laurentiu Vasile · Rating 3 out of 5]

The book is touching on our quest to understand how we can survive along with robots, artificial or augmented reality and how our world might look like. The author touches on our hope that 1 human + 1 machine will give a higher output than 2. Couple of case spice up the logic journey towards our future.... P&G, Apple being just a few examples.
What if there will be a global mind? What will this global mind want and do for us? Will it have consciousness?.... It seems consciousness is subjective for individuals and groups and thus really interesting to see if AI can evolve to a global non-harming super mind....
An optimistic book around one of our possible futures .... "If you are a scientific materialist, you may say that understanding all this complexity is good goal for future science"..... "If you are spiritually inclined you may think complexity as God will or Fate and try to harmonize your own actions to it".
Enjoy as you can..... the future will come soon enough

[Daniel · Rating 3 out of 5]

I had never thought of anything as a supermind before reading supermind. I liked the chapter where he explained the different types of superminds. And I also like the chapter where he talks about how the future of work. Or maybe it was how A.I. will effect work now and in the future. Something like that.

I think its interesting to think of communities, markets, ecosystems, governments as superminds. Reading this book made me think about how I think I have to know a lot of things and do everything myself. It made me think that maybe its better to be able to work with other people well. Because me as one person can't do that much. But, if I worked well with a group of people I could get more done. I don't have to know everything and do everything.

Superminds made me think that many people working together can a lot of the time do good things.

[Ryan Vannucci · Rating 4 out of 5]

My biggest takeaway from this book was a useful framework for understanding how decisions can be made collectively through different structures. Noticing that we're able to manipulate these structures to varying degrees and use technology to enrich them feels like a great first step toward positive societal change with the help of technology. Works that I think would pair very well with this framework would include analyses of the limitations of these decision-making structures and works that contemplate how to address diverse cultural problems with greater accuracy. Actively pairing the insights from Superminds with a greater range of social science and systems analysis might help us enrich how we use technology to solve problems without falling prey to the pitfalls of effective accelerationist reductionism.

[Keith Breinholt · Rating 3 out of 5]

Superminds is a fascinating topic. Using large groups of individuals, computers, companies, democracies, societies, etc, as a collectively agent for decision making. I loved the topic and the potential this new concept has.

On the other hand many of the examples used in this book are questionable, or just plain wrong. The the author calls some of them out, but not others. So you are left to discern if the false examples are there to provoke thought or an error in the reasoning. On the whole the ideas are thought provoking and engaging. But I am left questioning many of the conclusions drawn from examples.

Recommended reading. But validating the historical contexts, ideas and suggestions presented with critical thinking and some skepticism is healthy.

[Rhiannon Hanger · Rating 3 out of 5]

Laboured the point a bit, probably would have been more interesting to just read the research papers. Not sure I agree with a company being conscious nor the philosophical reach towards the end about organisms working towards the goals of what the strongest want. It completely ignores how that just isn’t true in cases of opposing wants and needs (eg humans may want clean energy but oil companies want profits)

[Miroslav Mitkov · Rating 5 out of 5]

Super minds are networks of thinking or information processing participants, which generate solutions.
Thomas Malone opens a new world of understanding of how decisions can be successfully made by people who are not experts in the certain field. By engaging with artificial intelligence to guide their discussions and find blind spots, amazing results can be achieved.

[Sambasivan · Rating 3 out of 5]

An ok read. No fresh insights for me. Seems more of a synthesis of current ideas.

[Potehin Ruslan · Rating 1 out of 5]

There's one useful idea in this book - concept of collective intelligence. Everything else - wasting your time. Extremely bad written, not interesting.

[Julia · Rating 3 out of 5]

Nothing ground-breaking here but I did enjoy it. A bit of food for thought and helped me some connections that I never considered in detail before.

[Carlos Bianchi · Rating 3 out of 5]

A simple concept so over explained

[Lobak Thirteen · Rating 4 out of 5]

Informational and its had been a good read.

[Michael Morris · Rating 1 out of 5]

uuuuggggggg