Did all Columbia University students cheat using AI? A young man achieved an "entrepreneurial comeback": not using AI means starting the race at a disadvantage.

Is almost everyone at Columbia University using AI to cheat? Roy Lee's astonishing revelation hits the pain point of education. From the viral video of cheating in an Amazon interview to Cluely's invisible interface, this young entrepreneur is leading the AI revolution.

Roy Lee, the founder of Cluely, claims that almost every student at Columbia University uses AI tools to cheat.

In response, a netizen said, "When I was in college, I never used AI. Back then, we used our brains!"

"This has been the case for 23 years, but the school turns a blind eye because admitting it would damage their business model. They stopped caring about students a long time ago."

Is the AI cheating scandal at Columbia just the tip of the iceberg?

A Reddit user, a recent graduate from the University of California, Berkeley, admitted that AI tools are extremely common among students.

He said, "Every student will use AI as much as possible. It doesn't make sense not to use it. If you compete with someone using AI, you can't win without it."

However, simply categorizing all AI use as "cheating" is clearly an oversimplification.

As a netizen said, the popularization of AI is similar to the rise of the Internet—students were also accused of cheating when they used the Internet to search for information back then.

Some people pointed out that current education is more like "imitating the appearance of learning," overly relying on papers and exam scores rather than true internalization of knowledge.

The popularity of AI tools exposes the vulnerability of this system: when a seemingly perfect paper can be generated by AI, traditional grading standards lose their meaning.

"Cheat on Everything": Cluely is going big!

Roy Lee is the founder and CEO of Cluely, and he recently received a $15 million investment.

What made him famous is the explosive startup slogan "Cheat on Everything."

The core of "Cheat on Everything" is to use AI to maximize your performance and productivity in various scenarios.

Roy Lee's explosive declaration has caused a stir in the industry.

Garry Tan, the president of Y Combinator, even said on the X platform that he would directly block the word "Cluely."

Roy sincerely responded, "I have big dreams for this company. You may think my marketing method is outrageous and that I'm a stupid young man doing something unethical. But the ultimate goal of this company will have a huge positive impact on humanity."

The goal is to create a utopian world where people can freely pursue what they really want to do instead of being forced to work by life.

A16Z explained the reason for the investment in a blog post and posed a provocative question: "What if breaking the rules is the key?"

What exactly is Cluely up to at the center of the storm?

In fact, it is a deeply integrated desktop AI assistant that can access system audio and the microphone to provide real-time information.

Through a screen overlay similar to Apple's "Liquid Glass," Cluely can provide real-time information and suggestions during meetings, sales calls, and even interviews without being noticed by others.

If you use the screenshot function built into a Mac, the Cluely overlay won't appear in the screenshot.

However, if you use the more underlying screen recording function on a Mac, Cluely won't be able to hide.

On the eve of Jarvis: Does "privacy" give way to efficiency?

Cluely's interaction method is basically the same as Apple's "Liquid Glass." All interfaces are covered with a semi-transparent glass effect.

Roy believes this is the future way of using AI, and now is the critical moment to seize the market.

Cluely aims to make users get used to the seamless interaction mode:

Instead of going to the ChatGPT website, users can press "Command+\" to directly use the AI assistant on the desktop.

In this way, the company can naturally collect user data.

When a multimodal supermodel with an extremely long context window appears in the future, Cluely will have enough data to create an all - around AI assistant like Jarvis.

As for the privacy issues caused by such an AI assistant, Roy doesn't care.

He asserts that if people can significantly improve efficiency by sacrificing data privacy, they will choose the latter without hesitation.

If handing over their data can make the model more personalized and efficient, people will be willing to give all their data to it.

Correspondingly, copyright laws and data protection laws will gradually disappear, and we will move towards a completely open - source world where all technologies can be shared.

In the AI era, the scarcest resource is a positive vision for the future.

Just like going back 600 years and asking a blacksmith, "This is a steam engine. Can you imagine what the world will be like because of it?"

He couldn't imagine that there would be data analysts in the future.

All he could see was, "Wow, this invention will automate my work. People won't need to learn how to hammer metal with a hammer anymore. This will make us lose the core critical thinking that defines humans."

The blacksmith would feel that people would lose the art of physical labor, which is the core of the blacksmith's job.

He would think it was completely negative and couldn't understand the positive impact of technology.

In fact, if AI has super - intelligence and surpasses humans cognitively and does better than humans, it will surely accelerate scientific progress. Cancer will no longer exist, and Alzheimer's disease will also disappear.

Imagine if you had all the money in the world and could get anything you wanted just by snapping your fingers.

This is the world brought by super - intelligence.

In this world, you won't be a mindless puppet.

You'll do what you really want to do. You won't have to do a hated mechanical job for eight hours a day just to become a data analyst.

Instead, you'll drink coffee just to take a walk and enjoy the process. You'll open a coffee shop because you love running it.

You'll meet interesting people, discuss interesting things, think about the philosophy of life, explore nature, and do everything you want to do.

Facts have proved that even if AI automates difficult jobs, humans still tend to do challenging things.

Take chess as an example. When the first chess - playing robot that could solve chess problems appeared, everyone thought it would be the end of chess.

But in fact, chess is more popular than ever.

Because human instinct isn't to produce just for the sake of production.

We do things because we naturally want to.

When technology allows us to get rid of all unnecessary chores and only do what we really enjoy, every day will be completely different.

Of course, this is on the premise that super - intelligence appears.

If super - intelligence doesn't appear, all these discussions are just jokes, and AI may just be a tool to automate 10% to 20% of white - collar jobs.

Cheating in an Amazon interview and being expelled from Columbia

Now, Roy thinks he's living the most fulfilling life.

At the beginning of this fall semester, Roy considered finding a job at a big tech company.

In another world, he might be an intern at Meta, doing a soulless software engineer job, worrying about work and the future, and preparing for the summer courses in his junior year at Columbia University.

That kind of life would be completely unfulfilling. Now might be the most exciting moment in human history.

Roy truly believes that he's working hard to push human society in the most positive direction.

His goal is to cure cancer, conquer Alzheimer's disease, and make us all live forever and lead the most ideal life.

The more people use AI, the higher their productivity will be.

If this tool can increase everyone's productivity by 10%, it's equivalent to adding the productivity of 700 million people to the world.