Recently, I found that I have too many scattered ideas and not enough energy to implement them all. Often, coming up with an idea takes just a few minutes, but scaling and implementing an impactful idea could take at least a decade, a difference of five orders of magnitude. For example, John Clifton “Jack” Bogle spent a lifetime realizing and popularizing his idea for index investment funds. In 1975, he created the world’s first index fund accessible to individual investors. He then devoted himself to promoting low-cost, passive investment strategies suitable for individual investors, founding Vanguard Group, which eventually became one of the largest investment fund management companies in the United States.

Thus, it’s incredibly difficult for one person to spend a lifetime realizing an impactful idea. Everyone always has endless important ideas. It would be a waste to keep them all buried inside, so I write some down when I have time over the weekend. Maybe some ideas have already been realized by others, and I’m just unaware. Or, if they are indeed new ideas and I can find like-minded people willing to spend money and collaborate, that would be a major win.

This article lists roughly eight whimsical ideas, and anyone inspired can freely work on them. I believe in Silicon Valley’s motto: Ideas are free, the difference is in the execution.

Category 1: Personal Life

Idea-1 (Biggest Mistakes Sharing Platform) Encyclopedia of Life’s Pitfalls

I am a fan of Charles Thomas Munger. He has a lot of wisdom, one of the most famous being the concept of inversion thinking. He said, “All I want to know is where I’m going to die, so I’ll never go there.” This approach to avoiding major pitfalls is very effective. In life, one has to make many decisions: education, choosing a major, choosing a spouse, starting a business, selecting an industry, investing, deciding whether to go abroad, seeking medical services, etc. While some decisions allow for trial and error, the cost of making mistakes in others can be too high (e.g., risk to life, major financial loss, irreversible medical harm, etc.). If we could know in advance the common pitfalls of major life events, we could make fewer fatal errors and reduce societal losses.

The content of this website should allow users to create categories and entries, share pitfalls, and engage in mutual comments and ratings. If we treat a person’s life as the execution of an algorithm with wisdom, then these pitfalls are like fatal bugs in software, and experts and scholars could standardize names and categories. For example, the Common Vulnerability Scoring System (CVSS) is an open framework for conveying the characteristics and severity of software vulnerabilities. With such an encyclopedia of pitfalls, individuals making significant decisions could scan a list of common pitfalls associated with those decisions to avoid major mistakes.

Idea-2 Automated Critical Thinking Assistant

If I could only learn one skill, I would choose critical thinking. Definitions of critical thinking vary, but a simple one is the analysis of existing facts, evidence, observations, and arguments to form a judgment. In an age of misinformation, everyone must learn to critically receive, verify, and judge the truth of information, and then avoid being misled among many choices, avoid extremes, avoid elementary thinking errors, and find the best decisions balancing various goals.

Unfortunately, people have many cognitive biases and are greatly influenced by emotions and environment, making it difficult to judge impartially. To correct biases, we still need to rely on technology, algorithms, and software. Recent advancements in artificial intelligence and machine learning in natural language processing have been remarkable. Some researchers have already used the latest large language models for reasoning with natural language.

I recently tested several critical thinking evaluation questions with OpenAI’s GPT-3. I found that it often generates correct answers, even providing explanations. This seems like a promising research direction. This research does not imply that all complex systems can be precisely reasoned and predicted but aims to technologically reduce low-level, common thinking errors as much as possible.

Idea-3 Solving Poverty by Paid Learning

Currently, education costs are paid by students/parents (directly or indirectly), which disadvantages students from impoverished families. Why can’t we make education profitable for students from poor families who meet certain standards through exams? Anyone, at any age, wanting to study while achieving certain grades should be able to earn enough to support themselves and even contribute to their family’s income.

To reduce the burden on taxpayers, funding could come from private individuals or organizations’ donations. To prevent the abuse of course content for indoctrination, donors could decide on the courses of study and standards for assessment. Everyone agrees that the best personal investment is in education, and for society, spending on education is the most effective way to improve the quality of the population and grow the economic pie.

With this pathway to earning through learning, anyone could afford to take time off to recharge and seek their next goal.

Idea-4 (Review of Each Person) Personal Review Website

This could be a good idea or a Pandora’s box. Currently, we can find reviews online for products, schools, and companies, making it convenient to know the pros and cons. Why isn’t there a similar public comment system for individuals, with whom interactions are most crucial? The most obvious reason is everyone’s desire for privacy, not wanting their flaws, deficiencies, and stains to be publicly disclosed.

However, such a system has many advantages, potentially reducing scammers and malefactors. Some companies (e.g., Bridgewater Associates founded by Ray Dalio) have begun to experiment with extreme truthfulness and transparency among employees. It’s said that knowing each other’s strengths and weaknesses makes communication and cooperation more effective. Implementing such a system requires some pioneers willing to take the lead, offering benefits for voluntary participation in this open evaluation system.

LinkedIn’s endorsement system is somewhat similar but focuses on publicly highlighting workplace strengths. What people need most might be avoiding major interpersonal pitfalls, but there’s currently no convenient system for publicly sharing such negative information. The U.S. personal credit system evaluates individuals from a borrowing perspective, but the channels for providing and querying information are still very narrow and inconvenient. Some websites allow individuals to set up suggestion boxes, where people reluctant to provide face-to-face feedback can offer it anonymously. There are also niche personal review websites, like those for rating teachers, contract workers, etc.

Category 2: Government and Public Services

The role of government is undeniable, and there are many areas where the government can improve.

Idea-5 (AI Bots for Universal and Direct Voting) Intelligent Direct Collective Decision-Making System

Traditional collective decision-making involves voters electing representatives who then make decisions. The main reason is that there are too many everyday decisions needed for a country or community, making it impractical for every person to understand the issues and vote on them.

For example, the U.S. Congress discusses almost 17,000 bills in a two-year term, of which about 4% eventually pass into law. This is just the decision-making at the congressional level; civil society has many layers of organization (states, counties, cities, school districts, homeowners associations, etc.), all requiring decisions. If every citizen participated in all decisions, they wouldn’t have time to earn a living.

Thus, out of necessity, we elect representatives at various levels to make decisions on our behalf.

This leads to the classic economic and game theory problem: the principal-agent problem, where the interests of citizens and their agents (representatives) often do not align. Elected officials often act against the best interest of their constituents. The same issue arises in company management, where the interests of shareholders and board members frequently diverge.

With current technological advancements, it’s entirely feasible to develop digital proxies for each citizen, precisely modeling each person’s values and producing corresponding decisions. In other words, use artificial intelligence to create personalized decision models, eliminating intermediaries and automating direct participation in all decisions by everyone. The models could also be market-selected through competition, allowing better models to prevail. I believe this is a very impactful research direction. This idea is not unique to me; a similar concept was proposed in a 2018 TED Talk. For those interested, check out:

https://www.ted.com/talks/cesar_hidalgo_a_bold_idea_to_replace_politicians

Idea-6 Automated Judicial Services

One of the major pain points of living in the United States is litigation, which can rank in the top three for inefficiency and high cost, preventing timely resolution of injustice or disputes. Many people, lacking the time and money to hire lawyers, cannot defend their legal rights. A recent example is the dispute over whether colleges can give special consideration or restrict students based on skin color; the lawsuit has been ongoing for over 8 years, with both sides spending tens of millions of dollars.

The development of artificial intelligence makes machine-assisted legal consultations, courtroom debates, and even trials possible. Future lawyer debates might become a contest between artificial intelligence. Free AI legal consultations could allow ordinary people to enjoy high-quality legal services. Developing automation technology for the future is very meaningful, and this is also a very significant research direction. Stanford University’s Human-Centered Artificial Intelligence (HAI) Institute recently established a Center for Research on Foundation Models (CRFM), planning interdisciplinary discussions aimed at making fundamental progress in research, development, and deployment of AI foundational models. Their report mentions driving legal services as a key application scenario. For more details, read the report On the Opportunities and Risks of Foundation Models:

https://crfm.stanford.edu/assets/report.pdf

Category 3: Commercial Products and Services

Idea-7 Low-Cost Medical Treatment and Therapy Sharing Website

The current medical system, driven by economic interests, often promotes treatments that address symptoms rather than root causes, which might not always be in the best interest of patients or at least not easily lead to low-cost therapies.

To break this situation, ordinary people should unite to exchange and research low-cost treatments that address root causes. For each disease, one could listen to possibly conflicted medical experts and also consider remedies discovered by fellow sufferers without a conflict of interest. This website could list various diseases, with patients recommending methods and others commenting and rating. Of course, a disclaimer would be provided.

Idea-8 Open Bounty System for New Services and Products

Scientists and entrepreneurs are constantly trying to understand what customers need and what society most requires to determine their research and development directions. It’s entirely possible to have a platform where everyone can offer bounties with their own money for new products and services they seek. Products and services with the most bounties could be developed and produced by businesses competing for bids, with the platform assisting in contract management, stepwise payments, final payment upon completion, etc.

This way, everyone’s ideas for improving the world can find like-minded people to offer real money as bounties through this platform, attracting capable individuals or teams to realize them. In the end, everyone becomes an investor pushing for social progress and also benefits from it. While there are similar crowdfunding websites where people and teams post projects,  the necessity of such a platform emphasizes the direct financial engagement from the public in fostering innovations.

Summary

These eight rough ideas address some of the biggest pain points today: avoiding pitfalls, optimizing thinking, solving the principal-agent problem, addressing medical issues, improving judicial efficiency, and addressing the challenges of investing in new products/services. Most are approached from an automation and intelligence perspective, while others leverage the power of crowdsourcing to involve the public in beneficial competitions. These are just initial thoughts to spark further discussion and are freely shared for anyone inspired to implement them.

“Ideas are cheap. Execution is expensive. The ability to execute separates people, not the ability to come up with ideas.”

Original Chinese version: https://mp.weixin.qq.com/s/XmQBBJldIfsNfH8wYMh5IQ