Start free trial
Trending New Genres Authors
EnglishEnglish
EspañolSpanish
简体中文Chinese
繁體中文Chinese (Traditional)
FrançaisFrench
DeutschGerman
日本語Japanese
PortuguêsPortuguese
ItalianoItalian
한국어Korean
РусскийRussian
NederlandsDutch
العربيةArabic
PolskiPolish
हिन्दीHindi
Tiếng ViệtVietnamese
SvenskaSwedish
ΕλληνικάGreek
TürkçeTurkish
ไทยThai
ČeštinaCzech
RomânăRomanian
MagyarHungarian
УкраїнськаUkrainian
IndonesiaIndonesian
DanskDanish
SuomiFinnish
БългарскиBulgarian
עבריתHebrew
NorskNorwegian
HrvatskiCroatian
CatalàCatalan
SlovenčinaSlovak
LietuviųLithuanian
SlovenščinaSlovenian
СрпскиSerbian
EestiEstonian
LatviešuLatvian
فارسیPersian
മലയാളംMalayalam
தமிழ்Tamil
اردوUrdu
Searching...
SoBrief
Third Millennium Thinking

Third Millennium Thinking

Creating Sense in a World of Nonsense
Better decisions begin by exposing uncertainty before trusting expertise.
by Saul Perlmutter 2024 320 pages
3.66
339 ratings
Listen
Immersive
V2.3
Amazon Kindle Audible
Try Full Access for 3 Days
Unlock listening & more!
Continue

Key Takeaways

Give experts the facts, but keep values and authority with affected people

Good decisions need three parts. In a hospital emergency, doctors can estimate which treatment is likeliest to keep a patient alive. They cannot decide how that patient should weigh survival against the risks of surgery, and they should not take the final choice away from a capable patient. Reliable information, personal values, and legitimate decision-making authority each matter. Overweight expertise and people become subjects of technocratic rule. Reject expertise in the name of autonomy and decisions can be driven by misinformation or whatever feels right.

That balance is 3MT. Third Millennium Thinking is the book’s name for using scientific habits to check claims, acknowledge uncertainty, and make decisions with others. It is not a demand to defer to scientists. It is a way for ordinary people to assess evidence and ask experts to explain how they reached a conclusion.

Analysis

The framework usefully distinguishes who can answer a factual question from who has the right to choose among outcomes. That distinction is essential in medicine, where a risk estimate is not itself a treatment decision. It also exposes a recurring political error: presenting a value judgment as if technical expertise settled it. The harder cases arise when decisions affect people who cannot speak for themselves, or spill costs onto others. Then authority needs justification too, not merely a confident expert or a personal preference.

Treat correlations as clues; intervene before calling something a cause

Association leaves options open. A survey might find that heavier drinkers have more osteoporosis. Drinking could contribute to bone loss, poor bone health could alter drinking, or a third factor, such as inactivity, could influence both. A correlation alone cannot identify which explanation is right. If the question matters, researchers need a way to distinguish these causal stories.

Experiments create leverage. In a randomized experiment, people are assigned by chance to a treatment or comparison group. This tends to balance known and unknown differences, letting researchers compare what happened with a useful estimate of what would have happened otherwise. When an experiment is impossible or unethical, evidence such as a large association, consistent results across settings, the cause preceding the outcome, a dose-response pattern, and a plausible mechanism can strengthen a causal claim. Percival Pott used this kind of reasoning to link chimney soot with a rare cancer among sweeps.

Analysis

The intervention idea turns causation from a seductive story into a testable contrast: change one factor and observe whether the outcome changes. The same logic applies outside laboratories, from product trials to workplace policies, though real settings rarely allow perfect control. Random assignment helps guard against confounding, but it does not guarantee that results transfer to every population or situation. The practical discipline is to ask both what the evidence shows and what alternative causal account could produce the same pattern.

Put probabilities on beliefs, then check whether your confidence earns them

Confidence has degrees. Probabilistic thinking means treating beliefs as more or less likely, rather than forcing every claim into true or false. A prediction that a major Bay Area earthquake has a 72 percent chance within 30 years communicates uncertainty while still supporting planning. A person is calibrated when predictions assigned a given probability come true at roughly that rate over many cases.

Authority can misstate risk. Before the Challenger disaster, NASA’s public estimate suggested a failure rate of one in 100,000 launches. Earlier evidence about the rockets used pointed to a far higher historical failure rate, roughly one failure per 28 or 29 launches under the same assumptions. The lesson is not that experts must be right every time. It is that they should show their uncertainty, and be judged by how well their confidence matches their record.

Analysis

Putting numbers on confidence makes disagreement more informative: two people may agree about the evidence yet differ over how strongly it supports a conclusion. Calibration also gives audiences a better standard than forceful delivery. However, precise probabilities can create a false impression of precision when the underlying evidence is weak or the question is poorly defined. A useful estimate therefore needs its scope, time horizon, and basis made clear, along with a record that permits later checking.

Before celebrating a pattern, ask how much noise you searched

Randomness makes convincing shapes. People often expect coin flips to alternate neatly, but genuine random sequences contain long runs of heads or tails. Searching noisy data for a pattern can therefore produce apparent discoveries that are coincidences. The risk grows when researchers examine many possible outcomes or locations and report only the most striking one. This is the look elsewhere effect.

A false planet appeared. Saul Perlmutter’s team detected a rapid, pulsing signal from a supernova remnant. After correcting for Earth’s motion, the signal looked like a planet was orbiting a pulsar. The apparent signal came instead from a nearby instrument leaking a similar frequency. By contrast, the Higgs claim gained strength when two independent detector teams saw a similar bump in their data. Before trusting a pattern, ask what else was searched and whether an independent check found it too.

Analysis

The central challenge is not simply detecting a pattern, but estimating how often a search would produce something equally persuasive by chance. That question matters in science, intelligence, investing, and everyday interpretation of other people’s behavior. Independent replication is powerful because it makes a shared artifact or analytic choice less likely to explain the result. Still, replication is not a magic stamp: researchers need to compare methods and assumptions, since different teams can share the same hidden blind spot.

Average random errors; hunt systematic errors with different methods

Two uncertainties need different fixes. Statistical uncertainty is random scatter around the true value, like readings from several hotel scales that are sometimes high and sometimes low. More measurements can reduce that scatter. Systematic uncertainty pushes readings consistently in one direction, like a home scale that is always five pounds too low. Repeating the same measurement will not correct that bias.

Check the measuring process. Climate researchers have to account for changes in where and how temperatures were recorded across the century. Some ocean readings shifted from water collected in buckets to measurements from ship intakes, which can produce different results. Researchers examine such changes and compare methods with different possible biases. Agreement across genuinely different approaches makes it less likely that one systematic error is driving the conclusion.

Analysis

A measurement can be highly repeatable and still be wrong, which is why consistency alone cannot establish accuracy. This distinction is valuable wherever records are compared across time, from health monitoring to institutional performance. Triangulation works best when methods have different failure modes, not merely different labels. It is also resource-intensive: some biases cannot be fully reconstructed after the fact. Honest reporting should distinguish what has been corrected, what has only been bounded, and what remains unknown.

Stay with hard problems by measuring progress in small, testable steps

Optimism keeps inquiry moving. Scientific optimism is the working expectation that a difficult problem may be solvable, enough to keep testing ideas through setbacks. Perlmutter’s team expected to find distant supernovae in three years. It took five years to find the first, seven to develop a reliable method, and ten to analyze the results, which revealed that the universe’s expansion is accelerating.

Progress must be visible. Long efforts survive through iterative advancement: each attempt should teach something or improve the next one. Fermat’s claim that he had proved his famous theorem helped mathematicians keep working on it for 358 years, until a proof was completed in 1995 using mathematics unavailable in Fermat’s time. But effort already spent is not evidence that a project deserves more. Continuing makes sense when learning or progress continues, not merely because quitting feels costly.

Analysis

This is optimism with a feedback loop, not cheerleading. It gives teams permission to begin before success is guaranteed while requiring them to notice whether their work is producing knowledge. That distinction is especially useful for research and public policy, where outcomes can take years to emerge. The danger is that visible progress can be manufactured by redefining milestones or ignoring costs. A sound effort therefore needs both a plausible path of incremental tests and agreed conditions for pausing or changing direction.

Use rough estimates to find the dominant cause before debating small ones

Approximation is a reality check. Fermi estimation breaks an unfamiliar quantity into factors that are easier to approximate, then combines them to see whether a claim is in the right ballpark. To estimate US gasoline spending, the authors used about 165 million cars, 10,000 miles per car each year, 20 miles per gallon, and $4 per gallon. The result was about $330 billion annually, close to the roughly $400 billion figure they found for 2012.

Scale helps rank causes. The same approach can compare public spending. Their rough estimates put Social Security above education, and education above incarceration, matching the actual order even though individual totals were imperfect. This kind of estimate does not decide what government ought to fund. It helps reveal whether a proposed explanation is large enough to matter before attention is spent on fine details.

Analysis

Approximation is a defense against numerical theater: a number can sound authoritative while being impossible at the scale claimed. Breaking it into components also reveals which assumptions drive the answer, making disagreement concrete. Fermi estimates are not substitutes for data when decisions demand precision, and rough inputs can share the same bias. Their best use is as a screening tool: identify implausible claims, rank likely drivers, and determine what deserves more careful measurement.

Hide outcomes while making judgment calls that could favor your preferred answer

Expectations shape analysis. Confirmation bias leads people to seek supporting evidence and scrutinize unwelcome evidence more harshly. In scientific work, analysts may reject bad data or fix software errors until a result looks familiar, then stop searching. Blind analysis reduces this risk by concealing the key outcome while researchers choose data, methods, and exclusions. Once those choices are fixed, the result is revealed.

The water-memory test illustrates why. Researcher Jacques Benveniste claimed that extreme dilutions of an antibody still produced a biological effect. In a supervised rerun, labels were hidden from the experimenters, and the striking result did not hold. The procedure exposed how expectations could influence the measurement. Related safeguards include preregistering a study’s planned tests and analysis before collecting data, and inviting independent teams to examine the same question.

Analysis

Blinding changes the decision environment rather than asking people to become perfectly impartial. That is its strength: it makes motivated selection harder at the point where choices are made. The technique does not remove every judgment call, and concealment can be impractical when analysts need outcome information to detect technical failures. Preregistration and independent analysis can complement it by making flexibility visible. The broader lesson is to design procedures around predictable human weakness, rather than treating good intentions as a sufficient safeguard.

Separate what policies do from what communities value

Evidence cannot choose priorities. In a Denver dispute over police ammunition, opposing sides argued about which bullet was best without agreeing on what best meant. A team separated three outcomes: injury to the person hit, the bullet’s ability to stop a threat, and danger to bystanders. Ballistics experts estimated how available bullets performed on those facts. Community members and stakeholders indicated how much each outcome mattered to them.

The split made comparison possible. Once the technical effects and public priorities were made explicit, the group identified a bullet that had not been central to the original argument and that better fit the stated criteria. This method does not make value judgments scientific. It lets people see where disagreement lies: in predictions about consequences, in the weight assigned to those consequences, or in both.

Analysis

Explicitly separating consequences from priorities can lower the temperature of a dispute because participants need not pretend that one technical answer settles every moral question. The Denver process also shows why experts should not define public values by default. Yet numerical weights can make a contested moral choice look more objective than it is. Who selects the outcomes, whose experiences are represented, and which harms are omitted remain substantive questions. Transparency improves deliberation; it does not abolish politics.

Build procedures where disagreement can improve the answer, not merely signal identity

Crowds need the right conditions. Averaging independent estimates can cancel random errors: students guessed an ostrich’s weight at an average of 326 pounds, near the actual 345. But discussion can make a group less accurate if people simply follow a majority or repeat a shared bias. Groups do better when members can challenge one another and share ways to test an answer.

Structure makes listening possible. Deliberative Polling brings a representative group of citizens together to study balanced briefing materials, discuss the issue in moderated small groups, and question experts with competing views. The goal is informed judgment, not forced unanimity. For uncertain future events, the Good Judgment Project combined quantified predictions, feedback on accuracy, and collaboration, with top forecasters often drawn from outside traditional expert ranks. Across these methods, trust rests on visible openness to evidence and correction.

Analysis

The book’s strongest civic insight is that trust can be supported by procedures, not just personal faith in a leader or institution. A well-designed process gives disagreement a constructive job: expose assumptions, test claims, and reveal which values are in conflict. Such formats require time, inclusion, and skilled facilitation, and they cannot guarantee good faith or erase power differences. Their promise is more modest and practical: make it harder for identity, status, or charisma alone to determine what a group believes.

Analysis

The book’s central argument is a form of fallibilist realism: a shared world exists, yet people reach it through imperfect instruments, uncertain inference, and institutions that can fail. Its most valuable move is to treat scientific thinking not as a body of expert conclusions but as a set of error-correcting practices. Random assignment, calibration, blinding, triangulation, and structured deliberation all make private judgment answerable to evidence beyond itself.

That shift also reframes the familiar conflict between expertise and democracy. Expertise improves factual estimates; affected people retain a claim to define relevant harms and acceptable risks. This is not a neat division. Choices about which outcomes to measure already reflect values, and a community’s preferences can be distorted by unequal voice or misleading information. The authors’ own methods imply a remedy: make those choices visible, contestable, and revisable rather than pretending they are neutral.

A deeper implication is that trust should attach less to a person’s certainty than to the correction mechanisms surrounding their claims. An expert who exposes methods, acknowledges limits, and permits independent checks may be more dependable than a fluent authority whose answer happens to sound reassuring. The same applies to institutions. Their credibility is partly a record of how they respond when the evidence changes.

The project is ambitious, but not a universal algorithm. Some decisions are urgent, some values are irreconcilable, and some communities lack the time or power to deliberate on equal terms. Even so, the framework offers a demanding standard for public reasoning: use evidence to constrain stories, use values to choose among consequences, and build procedures that make both open to challenge. Its optimism is credible precisely because it treats clear thinking as collective maintenance, not a permanent human achievement.

Summary in 30 Seconds
Correlation suggests possibilities, but random assignment and intervention provide stronger causal evidence. Calibrated forecasts should match their stated odds over time: a 72 percent prediction should succeed about 72 times in 100. More measurements reduce random error, not consistent bias, which requires independent methods. Rough estimates reveal dominant causes, while blinding, preregistration, and structured deliberation limit bias and separate evidence from values.
Contains spoilers

Last updated:

Report Issue

Review Summary

3.66 out of 5
Average of 339 ratings from Goodreads and Amazon.

Third Millennium Thinking receives mixed reviews, with an average rating of 3.63 out of 5. Readers appreciate its focus on critical thinking and analyzing information in a world full of misinformation. Many find it useful for those new to the topic, praising its insights on biases and decision-making. However, some critics find it repetitive and lacking in novel ideas. The book's interdisciplinary approach, combining perspectives from physics, philosophy, and psychology, is generally well-received. Some readers note that similar concepts are covered more effectively in other works.

Your rating:
4.19
342 ratings
Want to read the full book?

Glossary

Third Millennium Thinking

Shared scientific reasoning

The authors’ name for a practical set of habits drawn from science and other traditions. It includes testing claims against evidence, expressing uncertainty, checking for bias, and working with people who disagree. The aim is to make better individual and collective decisions in a world crowded with technical, conflicting, and sometimes misleading information.

probabilistic thinking

Reasoning in degrees

A way of treating claims as more or less likely rather than insisting they are simply true or false. It makes room for partial evidence, changing confidence, and decisions before certainty is available. A person using this approach states how confident they are and remains willing to revise that confidence when new evidence arrives.

interactive exploration

Testing through interaction

A way of gaining confidence that a phenomenon is real by acting on it and observing a responsive change. A billiard ball moves when pushed, and a sound display changes when a whistle changes pitch. Such interaction can make an instrument’s reading feel connected to the world, though many important measurements allow less direct testing.

Hill's criteria

Evidence for causal claims

A set of provisional considerations for judging whether an observed association is likely to be causal when decisive experiments are unavailable. They include the strength and consistency of an association, whether the cause precedes the effect, dose-response patterns, plausible mechanisms, and analogy. They guide judgment but do not function as a mechanical proof checklist.

calibration

Confidence matched to accuracy

The degree to which stated confidence corresponds to actual success across many predictions. Someone who is correct about half the time when expressing 50 percent confidence is calibrated at that level. Calibration is different from being right on one occasion. It is a measure of metaknowledge, or how well someone understands the limits of their own knowledge.

look elsewhere effect

False patterns from broad searching

The increased chance of finding an apparently meaningful result when many outcomes, subgroups, or patterns are searched. A study designed to test one outcome can produce misleading evidence if researchers also examine many unplanned outcomes and highlight whichever looks strongest. Planning tests in advance and accounting for the number of comparisons can reduce the risk.

statistical uncertainty

Random measurement scatter

Uncertainty caused by measurement results varying randomly around a true value. Repeated measurements may be high or low by chance, so averaging more observations can reduce this form of uncertainty. It differs from systematic uncertainty, which pushes measurements in a consistent direction and cannot be corrected simply by repeating the same procedure.

systematic uncertainty

Consistent measurement bias

Uncertainty caused by a measurement process that consistently shifts results higher or lower than the true value. A miscalibrated scale is a simple example. Repeating the same biased measurement does not make the error disappear. Researchers look for possible sources, compare different methods, or design procedures that cancel or quantify the bias.

scientific optimism

Persistence through uncertainty

The working stance that a difficult problem may be solvable, which helps people persist long enough to make progress. It is not a guarantee that any particular effort will succeed. The approach depends on incremental learning and improvement, and should not be confused with continuing a project solely because much time or money has already been spent.

Fermi estimation

Rough scale calculation

A method for approximating a difficult quantity by breaking it into simpler, more familiar estimates and combining them. The goal is usually to get the scale roughly right, not to produce a precise answer. Fermi estimates can test whether a public claim is plausible and help identify which factors dominate a problem.

blind analysis

Concealing results during analysis

A method for limiting confirmation bias by hiding the key outcome while researchers make choices about data, exclusions, or analytical procedures. Once those decisions are fixed, the result is revealed. Blinding reduces the temptation to make judgment calls that favor an expected answer, though it may require alternative ways to check technical problems.

Deliberative Polling

Informed citizen deliberation

A process that gathers a representative sample of citizens to consider a policy issue after studying balanced briefing materials, discussing the issue in moderated small groups, and questioning experts with differing views. Participants are polled before and after the event. The aim is not compulsory consensus, but a better-informed picture of public judgment after structured deliberation.

reflective equilibrium

Testing principles against cases

A method of moral reasoning that moves back and forth between judgments about particular cases and the general principles that might explain them. If a principle produces an unacceptable judgment, it may need revision, or the initial judgment may need reconsideration. The process aims for a considered fit between rules and specific decisions.

FAQ

What's Third Millennium Thinking about?

  • Focus on Decision-Making: The book explores how to make sound decisions amidst the overwhelming information available today, using scientific thinking.
  • Interdisciplinary Approach: It combines insights from physics, philosophy, and psychology to provide tools for clearer thinking and effective problem-solving.
  • Coping with Uncertainty: Emphasizes understanding uncertainty and using probabilistic thinking to improve decision-making in complex situations.
  • Scientific Thinking Tools: Highlights the importance of scientific thinking to navigate misinformation and biases in a complex world.

Why should I read Third Millennium Thinking?

  • Practical Tools: Offers practical tools and frameworks applicable to everyday decision-making, relevant for a wide audience.
  • Addressing Modern Challenges: Tackles issues like misinformation, polarization, and collaborative decision-making challenges in a complex world.
  • Empowerment Through Knowledge: Empowers readers to evaluate information and expertise confidently, crucial in today’s information-rich environment.
  • Promoting Collaborative Solutions: Emphasizes collaboration and collective intelligence for more effective and inclusive societal solutions.

What are the key takeaways of Third Millennium Thinking?

  • Probabilistic Thinking: Assign probabilities to beliefs and decisions for a nuanced understanding of uncertainty.
  • Signal vs. Noise: Learn to differentiate between meaningful signals and distracting noise in data for accurate decision-making.
  • Collaborative Problem-Solving: Encourages diverse perspectives to enhance understanding and improve collective decision-making processes.
  • Understanding Cognitive Biases: Recognize and mitigate biases like confirmation bias and groupthink to improve judgment and collaboration.

What is the significance of "probabilistic thinking" in Third Millennium Thinking?

  • Confidence Levels: Allows individuals to express confidence in statements, aiding informed decisions based on varying certainty levels.
  • Iterative Learning: Promotes continuous learning and adaptation, revising decisions as new information becomes available.
  • Real-World Applications: Applicable in fields like medicine and policy-making, where uncertainty is a constant factor.

How do the authors define "signal" and "noise" in Third Millennium Thinking?

  • Signal: Refers to meaningful information or patterns indicating a specific phenomenon or trend.
  • Noise: Represents irrelevant or random data that can obscure the signal, complicating true information discernment.
  • Context-Dependent: The distinction between signal and noise varies based on the observer's goals and context.

What is Fermi estimation, and how is it used in Third Millennium Thinking?

  • Quick Estimation Technique: A method for making rough calculations based on known quantities and logical reasoning.
  • Differentiating Factors: Helps distinguish between first-order (primary) and second-order (secondary) factors in complex problems.
  • Empowerment Through Estimation: Fosters confidence in understanding and engaging with the world, promoting a sense of agency.

What is the "look elsewhere effect" in Third Millennium Thinking?

  • Pattern Recognition: Occurs when searching for multiple patterns in data, increasing the likelihood of finding false signals by chance.
  • Statistical Implications: Highlights the importance of predefining hypotheses and limiting variables to avoid misleading conclusions.
  • Real-World Consequences: Can lead to erroneous claims in research, emphasizing the need for rigorous study design and analysis.

How does Third Millennium Thinking address the issue of expert overconfidence?

  • Calibration of Confidence: Discusses the importance of experts aligning their confidence levels with their accuracy.
  • Consequences of Overconfidence: Overconfidence can lead to poor decision-making and significant errors, as seen in historical examples.
  • Encouraging Humility: Advocates for intellectual humility among experts, promoting a culture where uncertainty is acknowledged.

What is Deliberative Polling, and how is it applied in Third Millennium Thinking?

  • Structured Public Engagement: Brings together a representative sample of citizens to discuss and deliberate on policy issues.
  • Combining Facts and Values: Allows participants to weigh factual information against values, leading to informed decision-making.
  • Real-World Applications: Provides examples of successful events leading to meaningful policy changes, demonstrating effective collective thinking.

What are the dangers of groupthink as described in Third Millennium Thinking?

  • Pathologies of Group Decision-Making: Groupthink occurs when consensus desire overrides realistic appraisal of alternatives.
  • Symptoms of Groupthink: Includes illusions of invulnerability, collective rationalizations, and self-censorship.
  • Preventive Measures: Suggests strategies like encouraging dissenting opinions and appointing a devil's advocate to prevent groupthink.

How does Third Millennium Thinking propose to rebuild trust in society?

  • Trust as a Foundation: Argues that trust is essential for effective collaboration and problem-solving in society.
  • Engaging Diverse Perspectives: Emphasizes engaging with differing viewpoints to foster understanding and bridge societal divides.
  • Creating Trust Networks: Advocates for developing networks prioritizing open-mindedness and learning, enhancing collective decision-making.

What are the best quotes from Third Millennium Thinking and what do they mean?

  • "We can get a handle on the world": Emphasizes empowerment through scientific thinking tools to understand complex issues.
  • "Truth will win when we each stop to 'do the math'": Highlights the importance of rigorous analysis and critical thinking for accurate conclusions.
  • "We have to stop doing this study right now": Underscores the need for self-awareness in decision-making, advocating for reflective approaches.

About the Author

Saul Perlmutter PhD is a renowned physicist and author of "Third Millennium Thinking." As a Nobel Prize-winning scientist, Perlmutter brings his expertise in physics to explore critical thinking and decision-making in the modern world. His background in scientific research and analysis likely informs the book's approach to evaluating information and addressing contemporary challenges. Perlmutter's collaboration with experts from other fields, including philosophy and psychology, demonstrates his commitment to interdisciplinary problem-solving. His work at UC Berkeley, where he teaches a course related to the book's topics, further establishes his credibility in the field of critical thinking and scientific reasoning.

Download PDF

To save this Third Millennium Thinking summary for later, download the free PDF. You can print it out, or read offline at your convenience.
Download PDF
File size: 0.42 MB     Pages: 21

Download EPUB

To read this Third Millennium Thinking summary on your e-reader device or app, download the free EPUB. The .epub digital book format is ideal for reading ebooks on phones, tablets, and e-readers.
Download EPUB
File size: 1.42 MB     Pages: 27
Want to read the full book?
Follow
Listen6 mins
Home
Swipe
Library
Get App
Try Full Access for 3 Days
Listen, bookmark, and more
Compare Features Free Pro
📖 Read Summaries
Read unlimited summaries. Free users get 3 per month
🎧 Listen to Summaries
Listen to unlimited summaries in 40 languages
—
❤️ Unlimited Bookmarks
Free users are limited to 4
—
📜 Unlimited History
Free users are limited to 4
—
📥 Unlimited Downloads
Free users are limited to 1
—
Risk-Free Timeline
Today: Get Instant Access
Listen to full summaries of 26,000+ books. That's 12,000+ hours of audio!
Day 2: Trial Reminder
We'll send you a notification that your trial is ending soon.
Day 3: Your subscription begins
You'll be charged on Nov 24,
cancel anytime before.
Consume 2.8× More Books
2.8× more books Listening Reading
Our users love us
1M+ readers
Trustpilot Rating
TrustPilot
4.6 Excellent
This site is a total game-changer. I've been flying through book summaries like never before. Highly, highly recommend.
— Dave G
Worth my money and time, and really well made. I've never seen this quality of summaries on other websites. Very helpful!
— Em
Highly recommended!! Fantastic service. Perfect for those that want a little more than a teaser but not all the intricate details of a full audio book.
— Greg M
Save 62%
Yearly
$119.88 $44.99/year/yr
$3.75/mo
Monthly
$9.99/mo
Start a 3-Day Free Trial
3 days free, then $44.99/year. Cancel anytime.
Unlock a world of fiction & nonfiction books
26,000+ books for the price of 2 books
Read any book in 10 minutes
Discover new books like Tinder
Request any book if it's not summarized
Read more books than anyone you know
#1 app for book lovers
Lifelike & immersive summaries
30-day money-back guarantee
Download summaries in EPUBs or PDFs
Cancel anytime in a few clicks

Why one summary?

Scanner
Find a barcode to scan

⭐
⭐
⭐
We have a special gift for you
Open
38% off
DISCOUNT FOR YOU
⭐
⭐
⭐
$79.99
$49.99/year
Only $4.16 per month
Continue
2 taps to start, super easy to cancel
Settings
General
Widget
Loading...