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
The Year in Tech 2022

The Year in Tech 2022

The Insights You Need from Harvard Business Review: The Insights You Need from Harvard Business Review
The next tech boom depends less on what machines can do than who sets their rules.
by Harvard Business Review 2021 176 pages
3.56
176 ratings
Listen
Immersive
V2.3
Amazon Kindle Audible
Try Full Access for 3 Days
Unlock listening & more!
Continue

Key Takeaways

Turn AI ethics from abstract principles into funded, product-level risk controls

Ethics must reach operations. AI can magnify existing discrimination, as alleged in an Optum health-care algorithm that directed more attention toward white patients than sicker Black patients. Broad promises such as fairness do little unless teams decide what fairness means for a particular product, who owns the decision, and how outcomes will be checked. The proposed approach draws on existing governance bodies, sets industry-specific standards, gives product managers practical decision tools, trains and rewards staff, and monitors effects after launch.

Build the system around decisions. Health care offers a useful model: informed consent becomes meaningful through clear explanations and real procedures, not a slogan. Amazon reportedly abandoned a hiring model it could not make non-discriminatory; Sidewalk Labs ended its Toronto smart-city project after public backlash over data practices. Both cases show that ethical failures can waste money as well as trust.

Analysis

The central move is from values statements to accountable routines, much as safety engineering converts a general aim into inspections, incident reporting, and design constraints. That translation also reveals unavoidable trade-offs: a model may be more accurate overall while producing worse results for a subgroup, and no single fairness measure settles what is acceptable. The strongest governance therefore makes the choice explicit, assigns decision rights, and includes affected people. It should also distinguish preventing harm from managing reputational exposure, since a program built only to protect the brand can reproduce the very distrust it claims to address.

Let employees rehearse hard conversations before customers pay the price

Practice without real-world fallout. Virtual reality (VR) training places employees in simulated interactions with digital avatars, letting them try difficult conversations and receive feedback without risking a customer relationship. In a PwC study, employees completed VR training up to four times faster than classroom training and felt almost four times more emotionally connected to it. More than two-thirds of surveyed learning leaders had implemented, or planned to implement, VR soft-skills training.

Measure behavior, not novelty. H&R Block used avatar-based calls to help new representatives practice empathy and conflict resolution. After introducing the program, it reported 50% fewer dissatisfied customers and 9.9% shorter customer-handling times. Cognizant used simulated client presentations with automated feedback, while HPE Financial Services recorded sales role-plays so managers could direct training where it was needed.

Analysis

Simulation is useful because communication skill is not merely information to memorize; it is behavior practiced under social pressure. The format resembles flight simulators, where safe repetition helps learners encounter unusual situations before lives are at stake. But emotional engagement and speed are not proof of durable learning. Organizations should check whether performance transfers to real conversations, whether automated scoring rewards the right behaviors, and whether employees can challenge inaccurate evaluations. VR is most persuasive as a rehearsal and coaching tool, not as a substitute for human judgment.

Reimburse digital health apps only when evidence keeps accruing in practice

Germany made payment a test. Its Digital Healthcare Act created a route for approved digital health applications, or DiGA, to be prescribed and reimbursed by statutory insurers covering about 90% of the population. The regulator can place an app on a central registry through an accelerated review. If the app lacks full evidence of benefit at entry, its maker has up to 12 months to demonstrate positive effects, such as better health outcomes or improved treatment adherence.

Study real use. The system permits real-world evidence: findings drawn from routine care data rather than only controlled trials. That can show whether patients and clinicians actually use an app and whether it helps outside a tightly managed study. Data from the registry may also help other health systems decide what they should pay for.

Analysis

The policy links innovation to reimbursement, a powerful way to make adoption answerable to evidence rather than marketing. Real-world data can reveal practical effects that trials miss, including who stops using an app and how it fits into ordinary care. Yet observational data can be misleading: patients who choose an app may differ from those who do not, and changing products complicate comparisons. The sound approach is not to replace randomized trials wholesale, but to combine rigorous trials with ongoing monitoring and clear standards for what counts as meaningful patient benefit.

Use blockchain to connect supply-chain records, but verify the physical goods

One shared record can close blind spots. A blockchain is a distributed digital ledger that participating organizations maintain together, recording transactions in a tamper-resistant sequence. In supply chains, it can connect information, inventory, and financial activity that separate company systems often fail to link. A bank could verify a supplier’s order, provide working capital, and track delivery against payment. Smart contracts, computer instructions that act when recorded conditions are met, can automate steps such as releasing payment.

Start with a narrow use. Drug or food traceability can help firms identify which batches and suppliers are involved in a recall. Walmart Canada used a shared ledger with trucking companies to synchronize shipment information and payments. The technology does not guarantee that the original item or scan is genuine: firms still need physical checks, secure tags, and clear rules about who can join and what information is shared.

Analysis

The practical promise is coordination among organizations that do not want to hand control of all their data to one central operator. That makes the technology less a magical database than a shared rulebook for recording events. The important limitation is the gap between a trustworthy record and a trustworthy world: a ledger can preserve a false entry perfectly. A pilot should therefore test data quality, partner incentives, and dispute procedures, as well as transaction speed. If existing systems already share reliable information cheaply, adding a blockchain may only add complexity.

Protect brain data before employers turn attention into a performance metric

A brain-computer interface has two sides. A BCI links brain activity to an external device, often using noninvasive sensors that detect electrical signals. It could help a worker avoid operating machinery while drowsy, support focus training, or let someone control a computer through brain signals. But a manager might also use the same data to monitor attention or stress. The technology cannot reliably tell whether a person is focused on a meeting, a phone, or a private thought.

The measurements have limits. The article reports that 15% to 30% of people may not produce signals robust enough to operate a BCI. Age, lifestyle, and other individual characteristics can also affect readings. Inaccurate signals could lead to unfair evaluations, while stolen or altered data could threaten privacy and even physical safety.

Analysis

Brain signals are unusually intimate, but they are not transparent windows into a person’s intentions. The measurement problem matters as much as the privacy problem: a noisy proxy for attention can become a consequential employment judgment simply because it looks scientific. This is a familiar danger in management metrics, where a measurable indicator can displace the quality it was meant to represent. Any workplace use needs meaningful consent, strict limits on access and retention, and a ban on treating uncertain readings as objective evidence of effort or character.

Prepare for quantum security before quantum computers become practical

Quantum computing uses different rules. A qubit, the quantum counterpart of a classical bit, can occupy a probabilistic combination of 0 and 1. This property may make quantum machines useful for particular tasks, including simulating molecules, optimizing complex systems, and searching unsorted data. Quantum communication also offers a different basis for encryption: the laws of physics make it possible to detect attempts to copy quantum information.

The promise is not readiness. Existing quantum machines are small, and qubits are fragile enough that small disturbances can destroy information. Scaling them while correcting errors remains a major engineering challenge. The practical advice is to plan for encryption changes, bring quantum researchers together with industry specialists to identify valuable uses, and address access and social impacts while the field is still developing.

Analysis

Quantum technology is best understood as a specialized tool, not a faster replacement for every computer. That distinction guards against both hype and complacency: broad commercial use is uncertain, while the prospect of future machines breaking some current encryption makes long-term security planning prudent. The analogy to infrastructure migration is apt because changing encryption across large systems takes time. Still, organizations should avoid buying vague quantum readiness projects. A useful plan identifies vulnerable data, follows credible standards work, and tests whether a proposed use actually benefits from quantum computation.

Build the customers in space before betting on an economy there

Space has two different markets. The space-for-Earth economy sells satellite services and other space-produced benefits to customers on Earth. In 2019, it accounted for 95% of the estimated $366 billion space-sector revenue. A space-for-space economy would instead supply people and businesses operating in space, through services such as transportation, construction, manufacturing, and eventually resource use.

Demand must come before scale. The book describes private companies beginning to carry people into space, while government agencies such as NASA remain crucial customers. SpaceX’s launch business and Made In Space’s work on manufacturing illustrate how firms can use government contracts to build capabilities that might later serve private activity. The transition depends not only on technology, but also on rules for safety, property rights, shared resources, and international cooperation.

Analysis

The distinction between selling space services to Earth and building commerce for people in space clarifies why earlier ventures struggled: they pursued supply before a nearby customer base existed. This is a classic market-formation problem. Public contracts can absorb early risk and create capabilities, but they do not prove that private demand will become large enough to sustain the next stage. Policy must also guard against treating orbital space or lunar resources as ownerless opportunities. Commercial growth depends on legitimate institutions as much as launch prices.

Trade growth-at-any-cost tech models for businesses that can survive scrutiny

The old bargain is weakening. The book identifies four pressures on Silicon Valley: dominant firms crowding out rivals, investors demanding clearer paths to profit, public and regulatory scrutiny of harms, and rising expectations that companies act responsibly. A survey cited in the article found that 85% of respondents thought Big Tech had too much power. The author expects microtargeted advertising and weak gig-worker protections to face greater pressure.

Growth is not a business model by itself. Some direct-to-consumer brands struggle with costly customer acquisition, limited scale, and products that are not meaningfully better than established alternatives. The article sees room for service platforms that help many brands sell online, as well as companies that pursue slower, more sustainable growth. Legacy firms may have an advantage in physical operations and established brands, though digital transformation is no guarantee of success.

Analysis

This argument treats regulation and public trust as constraints that reshape competitive advantage, rather than as public-relations issues to be managed after growth. That is a useful strategic lens: a model dependent on hidden costs, cheap capital, or weak worker protections can look efficient only because someone else bears the bill. Still, slower growth does not automatically mean better conduct, and ethical branding can become another marketing layer. The more credible signal is whether a company changes incentives, product design, and accountability when responsible choices reduce short-term revenue.

Make inclusion measurable, and never outsource the burden to Black employees

Underrepresentation shapes daily work. The author describes being one of very few Black employees in technology workplaces and facing exclusion, racial jokes, and assumptions about competence. Figures cited in the article show Black employees made up 2.8% of Google’s technical workforce in 2018, while just 1% of venture-capital funding went to Black founders and employees. These are not only hiring gaps; they affect who feels welcome, who advances, and whose work is trusted.

Individual advice cannot repair an institution. Barton recommends maintaining personal integrity, speaking up about offensive assumptions, and seeking help when needed. He also urges managers to take complaints seriously, document incidents, and escalate them when supervisors fail to act. Leaving a workplace can be a valid choice when staying requires compromising one’s values.

Analysis

The essay makes a crucial distinction between helping individuals navigate a hostile workplace and assigning them responsibility for fixing it. Personal strategies can protect agency, but they cannot substitute for leadership accountability, fair promotion systems, or consequences for discrimination. The invitation to speak up also carries unequal risk: people with less organizational power may face retaliation for the same candor others are praised for. A serious inclusion effort therefore tracks outcomes and protects complainants, while ensuring that the labor of educating colleagues does not fall by default on those already harmed.

Make platforms regulate together before public trust collapses into government control

Self-regulation means firms set and enforce rules themselves. The article argues that social platforms should act before governments impose broader controls on harmful content. The danger is a tragedy of the commons: each company protects short-term engagement or revenue, while collectively eroding trust in the online environment all depend on. The book points to movie and video-game ratings, broadcast advertising rules, and airline reservation standards as examples of industries that developed shared safeguards.

Coalitions can curb free riding. If one platform moderates content while competitors do not, the first may bear the costs alone. Shared standards reduce that disadvantage. The authors argue that self-regulation has often become serious when government intervention is a credible possibility, and that transparent, collective rules can be preferable to leaving every decision to lawmakers.

Analysis

Platform governance is a collective-action problem: rules that are costly for one firm can become viable when competitors adopt them together. Historical industry codes show that private standards can help, but they are not automatically democratic or effective. The key tests are transparency, independent scrutiny, appeal mechanisms, and evidence that rules are applied consistently. Self-regulation should complement public oversight, not serve as a reason to block it. A coalition that writes convenient standards and exempts its own members would merely reproduce the trust problem under a different name.

Measure software by energy consumed, not just speed and accuracy

Code has a physical footprint. Software consumes energy through the hardware and data centers that run it. The authors cite an estimate that information and communications technology could account for 14% of global carbon emissions by 2040, compared with about 1.5% in 2007. In one example, improving an AI model’s accuracy from 96.17% required a sharp increase in energy use, with the final 0.08 percentage point costing nearly four times the energy of the first stage.

Put efficiency into the whole lifecycle. Set environmental targets early, compare algorithms and tools, and measure energy use during operation. Cloud providers and users can reduce waste through data compression and deduplication, and by using architectures that scale computing resources to actual demand. Energy efficiency belongs alongside security, accessibility, and functionality when teams assess software quality.

Analysis

The key insight is that digital products are not immaterial: computation moves environmental costs into electricity generation, hardware production, and data-center infrastructure. Measuring energy alongside accuracy exposes diminishing returns that conventional product metrics hide. Yet efficiency alone is not the whole climate ledger; the source of electricity and the hardware lifecycle also matter. Organizations should therefore track energy per useful outcome, not simply per computation, and disclose the boundaries of their estimates. That makes it harder to celebrate a more efficient model whose total usage grows enough to increase emissions overall.

Analysis

The collection’s argument is less that technology is inherently liberating or dangerous than that its consequences depend on the institutions surrounding it. Across AI, workplace monitoring, supply chains, digital health, social platforms, and space commerce, the recurring question is who sets the rules, who bears the risk, and who can contest a decision. Technical capability often arrives before the social agreement needed to use it responsibly.

That pattern makes governance a practical management issue, not an abstract ethical add-on. The AI chapter asks firms to make principles operational; the digital-health example ties market access to evidence; blockchain requires partner rules and trusted inputs; and platform self-regulation depends on collective incentives. Even apparently technical subjects such as quantum security and green software demand early decisions about infrastructure, access, and measurement. The book’s most useful strategic lesson is therefore to build mechanisms for learning and correction alongside new systems, rather than assuming a launch completes the work.

The anthology also supplies a needed counterweight to technology boosterism. VR training is illustrated with business outcomes, but those results do not guarantee lasting learning. Real-world health data can accelerate evaluation, but it can also mislead without careful comparison. A secure ledger cannot make a false physical record true, and brain sensors cannot turn ambiguous signals into reliable judgments about a worker’s attention. These limits are not reasons to reject innovation; they are reasons to match claims to evidence and preserve human accountability.

Because the book gathers short articles written in 2020 and 2021, some forecasts are proposals or expectations, not established outcomes. Its lasting value lies less in predicting which technology wins than in showing how to interrogate a deployment: identify the affected people, state the trade-offs, test real-world effects, and assign responsibility when harm appears. The future arrives as technology, but whether it becomes broadly beneficial is an institutional choice.

Summary in 30 Seconds
Technology's effects depend on who sets the rules, who bears the risks, and whether outcomes are checked after launch. AI principles need product-level safeguards and assigned responsibility; digital health apps need evidence from routine care. A shared blockchain cannot verify that physical goods are genuine, and brain sensors cannot reliably establish attention. Software also has an environmental cost: ICT could reach 14% of global carbon emissions by 2040.
Contains spoilers

Last updated:

Report Issue

Review Summary

3.56 out of 5
Average of 176 ratings from Goodreads and Amazon.

The Year in Tech 2022 receives mixed reviews, with an average rating of 3.55 out of 5. Readers appreciate its overview of emerging technologies and thought-provoking ideas, covering topics like ethical AI, blockchain, and space travel. Some find it accessible and informative, while others criticize its superficial treatment of complex subjects. The book's forward-looking approach is both praised and questioned. Reviewers note its value for staying updated on tech trends but highlight inconsistencies in depth and relevance across chapters.

Your rating:
4.1
62 ratings
Want to read the full book?

Glossary

Data and AI ethics program

Operational ethical risk controls

An organization-wide system for identifying, evaluating, reducing, and monitoring ethical risks in data use and AI products. It connects senior leadership, product teams, governance bodies, and affected stakeholders through standards, decision procedures, training, incentives, and post-launch review. The aim is to make ethical commitments actionable in specific business and regulatory settings.

Digitale Gesundheitsanwendungen (DiGA)

Prescribable digital health applications

Germany’s category of digital health applications, including software, mobile apps, and browser-based tools that can be prescribed by physicians or psychotherapists. Eligible applications can enter a central registry and receive reimbursement through statutory health insurance. They must meet requirements that include data protection, security, interoperability, and evidence of positive effects.

Fast-Track Process

Accelerated app review

The regulatory pathway created in Germany for qualifying digital health applications. The Federal Institute for Drugs and Medical Devices reviews applications for registry inclusion. An app with preliminary evidence of benefit may be listed while its maker gathers further evidence, which must be provided within 12 months if it was not sufficient at the start.

blockchain

Shared tamper-resistant ledger

A distributed digital record in which multiple participants maintain copies of a chronological sequence of transactions. In the supply-chain use described here, known organizations can record movements of goods, orders, and payments, making the history easier to inspect and harder to alter retrospectively. The ledger does not by itself ensure that data entered about physical goods is accurate.

smart contracts

Code-triggered contract actions

Computer code that uses recorded data to determine whether specified contractual conditions have been met and then initiates an action. In a supply chain, a smart contract might release payment after delivery is recorded or flag an exception for human review. Its reliability depends on accurate inputs and well-designed rules.

brain-computer interfaces (BCIs)

Brain signals linked to devices

Systems that connect brain activity to an external device, commonly by using sensors to detect electrical signals and algorithms to interpret patterns. Noninvasive versions include wearable headbands and earbuds. Potential uses include assistive control, neurofeedback, and safety monitoring, but readings can be inaccurate and raise serious questions about privacy, consent, and workplace power.

qubits

Quantum information units

The basic units of information in quantum computing, analogous to bits in classical computers. A qubit can exist in a probabilistic combination of 0 and 1, enabling quantum operations that differ from ordinary binary computation. Qubits are highly sensitive to disturbances, so maintaining and correcting their information is a major engineering challenge.

space-for-space economy

Commerce serving space activity

An economy in which goods and services are produced in space for use by people and organizations in space. Potential activities include transportation, manufacturing, construction, and infrastructure services. It differs from the space-for-Earth economy, which uses space-based assets, such as satellites, to provide products or services to customers on Earth.

network effects

Value grows with participation

A dynamic in which a platform becomes more useful as more people or organizations use it. Social platforms can attract users, advertisers, and other participants in mutually reinforcing cycles. These effects can help successful platforms grow quickly, but they can also concentrate power and amplify harmful content or business practices.

self-regulation

Industry-created operating rules

Rules and enforcement practices developed by companies or industry associations to guide their own conduct, often before or alongside government regulation. It can include monitoring, standards, and content policies. The book emphasizes that shared industry action can reduce the disadvantage faced by a single firm that adopts costly safeguards alone.

green software

Software designed for lower energy use

Software whose environmental effects are considered during design, development, and operation. Practices include measuring energy consumption, choosing efficient algorithms and tools, reducing unnecessary data storage or transfer, and matching computing capacity to demand. The goal is to make energy use a core measure of software quality alongside performance, security, and functionality.

About the Author

Harvard Business Review is a renowned publication known for its management and business insights. While specific author information for "The Year in Tech 2022" is not provided, HBR typically features contributions from leading experts, scholars, and industry professionals. The publication's annual tech series, started in 2021, aims to provide readers with a comprehensive overview of emerging technologies and their potential impact on business and society. HBR's approach often combines academic rigor with practical applications, making complex topics accessible to a wide audience of business leaders, entrepreneurs, and technology enthusiasts.

Download PDF

To save this The Year in Tech 2022 summary for later, download the free PDF. You can print it out, or read offline at your convenience.
Download PDF
File size: 0.36 MB     Pages: 19

Download EPUB

To read this The Year in Tech 2022 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: 1.39 MB     Pages: 24
Want to read the full book?
Follow
Listen
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 25,
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...