Copilot defendants, LeCun autonomous AI, AlphaFold discovers all proteins...exciting AI news in 2022

2022 is a roller coaster year for artificial intelligence.

From art, video, medicine...artificial intelligence is impacting the mainstream, and the public is increasingly discussing AI.

What will happen in 2023? As we look to the future, let's look back to the past.

The following are some of the big AI news in 2022.

Programmers sue Copilot

Last month, programmer and author Matthew Butterick sued Microsoft and some of its partners, GitHub and OpenAI Wait to file a class action lawsuit.

It’s all because of Copilot, a first-of-its-kind artificial intelligence program that can generate code.

The purpose of Copilot is to assist development teams by creating simple code, thus saving man-hours for other tasks.

The program does not acknowledge how it learned from existing code on the data training set.

This will be a decisive lawsuit for those who care about artificial intelligence, training sets, and the original creators who find their work used in training programs.

How this problem is solved may have a lasting impact on the field for many years to come.

AI app Lensa is causing a stir in the art world

Lensa is an app powered by AI that has taken the internet by storm recently.

On almost any social media platform, you can see photos generated by netizens using Lensa.

As long as users upload their own images, Lensa will turn the photos into stylized works of art.

This may sound innocuous, but many people in the art world are unhappy because Lensa and other programs are trained on readily available photos and artwork.

Learning from images created by humans certainly enhances the capabilities of AI. But to many in the art world, this is plagiarism.

On Twitter, netizens erupted into heated discussions over whether Lensa and similar AI programs are good or bad for art.

It is undeniable that AI is likely to continue to exist.

The question is, how will AI impact the art world in the long term?

AI algorithms are running Washington

Imagine a city run by machines and programs.

Although this sounds like science fiction, it is close to reality.

Washington, D.C. and other metropolitan areas are leveraging machine learning programs and other artificial intelligence tools to manage their entire cities, according to a report.

From housing screening, sentencing, education and budgeting, cities seem to be quietly applying algorithms to make operations more efficient.

Still, some question the effectiveness of current AI plans to empower bureaucracies and any ethical issues that might arise.

The latest ally to help humans fight cockroaches is...laser artificial intelligence?

Not all artificial intelligence-related news this year is worrying.

Scientists this year created a new open source artificial intelligence tool to help rid us of those cockroaches.

Heriot-Watt University researcher Ildar Rakhmatulin and colleagues combined machine learning with machine vision to conduct a series of experiments on cockroaches.

During the experiment, the artificial intelligence was able to detect cockroaches and kill them at a distance of 1.2 meters.

If Rakhmatulin and his team succeed, humanity will not only get a wonderful laser show, but also a cockroach-killing show!

The British House of Lords received the artificial intelligence robot Ai-Da

As early as October, the history of AI was created in the British House of Lords .

#For the first time ever, an artificial intelligence robot named Ai-Da sat in the seats and answered questions in the Upper House of Parliament.

From creativity to technology, Ai-Da is at your fingertips. This is a glorious moment in the history of artificial intelligence.

Previously, Ai-Da was famous for her wonderful portraits of Billie Eilish, the late Queen Elizabeth and Paul McCartney.

Such public inquiries are likely to grow as governments around the world face the rapid rise of artificial intelligence.

Boston Dynamics and other companies have promised not to weaponize robots

Many people are worried that Skynet will become a reality one day. AI-driven robots will swarm onto the surface of the earth, completely replacing humans and becoming the rulers of the earth.

At least for now, they don't have to worry.

Boston Dynamics and other major robotics companies have pledged not to weaponize robots.

They made the surprising announcement this year, a decade after drones and other weapons dominated the modern battlefield .

In a letter to Axios, Boston Dynamics CEO Robert Playter said, "We are concerned about recent efforts to weaponize commercial robots."

This move by these private companies is gratifying. But how long can this unity last? Time will give us the answer.

LeCun, head of Meta AI, published a paper on autonomous AI

Artificial intelligence in science fiction and the gap between AI in the real world The differences seem to be getting smaller and smaller.

LeCun, the head of Meta AI, published a paper proposing a way to better train AI architecture to teach it to predict or plan for changes in real-world environments.

In short, it is an AI that can learn like humans and animals.

If this becomes a reality, it will change the entire game.

The relationship between AI and humans will once again spark debate.

White House unveils Artificial Intelligence Bill of Rights

As the scale of technology continues to grow, more and more people are beginning to discuss responsible AI .

Back in October, the White House anticipated this and unveiled an AI Bill of Rights.

The purpose of this bill is to protect individuals’ personal data and limit surveillance. However, the bill does not provide for operational or enforcement mechanisms.

Instead, the purpose is to take the first step for U.S. federal agencies to start figuring out how to respond to emerging technologies.

The bill also encourages companies to develop a set of "core principles" to expand user control over data and prevent bias.

DeepMind’s AlphaFold discovered nearly every known protein

At the end of summer, DeepMind discovered nearly every known protein currently known to science Every protein.

This open source program is based on AlphaFold, developed in 2018, and uses machine learning algorithms to predict the three-dimensional structure of proteins.

"Determining a protein's 3D structure used to take months or years, but now it takes seconds," explains cardiologist Eric Topol of the Scripps Research Translational Institute. ...These newly added structures can illuminate almost the entire protein world, and we have reason to expect that more biological mysteries will be solved every day."

This is a huge breakthrough that AI has brought to the microbiology and medical communities. Only time can tell how big a change this is.

Google AI engineer Blake Lemoine claims that LaMDA is sentient

One of the most hotly discussed topics in the AI ​​circle last summer was the former The story of Google AI engineer Blake Lemoine.

He claims that Google’s chatbot LaMDA has developed sentience.

He was very sure of this conclusion. "I studied philosophy of mind in graduate school. I have told people at Harvard, Stanford, and Berkeley that LaMDA is conscious."

Although Google denies this statement , but scientists have been concerned about AI’s sentient capabilities for years, and even more so now as AI advances.

Is this true or false? We can't know yet.

Okay, those are the hottest AI news stories of 2022.

What big things do you think will happen in the field of AI in 2023?

The above is the detailed content of Copilot defendants, LeCun autonomous AI, AlphaFold discovers all proteins...exciting AI news in 2022. For more information, please follow other related articles on the PHP Chinese website!