Technology must comply with the rights of the people, the EU releases a white paper on artificial intelligence
Daimler: Human beings will still be the creators of technology
In 1814, the British George Stephenson invented the first steam locomotive in human history, known as the "Voyager". It kept fire coming out of the chimney as it moved forward, so it was called a "train".
The first railway was built in Britain in 1825. In September of that year, the "Voyager" locomotive towed more than 30 small carriages for a formal trial and ran 40 kilometers at a speed of 24 kilometers per hour. distance.
At the time, technological skeptics viewed the locomotive as an "evil thing" and predicted that humanity would suffer consequences from this "hellish" invention. For example, they are worried that the smoke on the train will poison passengers, the train airflow will cause pneumonia, and the fast pace of the train will cause brain confusion, etc.
Why did the skeptics change their views again?
Ms. Renata Jungo Brunger, a member of Daimler’s board of directors and responsible for business integrity and legal affairs, put forward two reasons: First, because over time, people have completely adapted to The existence of these technologies has even begun to understand the value of new technologies. Also, sooner or later, the rules governing these new technologies will be formulated.
On the morning of September 10, 2019, German time, when she gave a speech on Daimler’s artificial intelligence guiding principles at the Frankfurt Motor Show that opened that day, she believed that today’s digital transformation is also undergoing the same change, and artificial intelligence The same is true for the rise of intelligence, which poses similar challenges.
Consulting firm PricewaterhouseCoopers (PwC) predicts that by 2030, artificial intelligence will drive GDP growth of approximately 26%, 14%, and 10% in China, North America, and Europe respectively. This technology will not only help overcome the many challenges posed by climate action, but also further optimize the results achieved in fields such as transportation, medicine and agriculture. Automotive Business Review noted that while technology experts, business leaders and government officials all agree that artificial intelligence is one of the most transformative technologies in the world, they generally note that it also brings new changes to personal privacy and daily life. risk.
Bai Yunge said: "Artificial intelligence has been fully integrated into our daily lives, and the field of autonomous driving is a good example. However, every invention in history is full of charm and also makes people feel sad. It creates awe.”
She said that for most artificial intelligence applications, a large amount of data is essential, including personal data. However, many customers do not want their personal information to be leaked. They worry about incomplete data protection, data tampering, or loss of control. Some people even worry that such advanced artificial intelligence systems will one day rule the world.
How to seize the opportunities brought by artificial intelligence while resolving the attendant crises, fears and challenges?
To this end, Daimler has identified four principles: reasonable use; explainability; privacy protection; safety and reliability. Bai Yunge emphasized: "Humans will still be the creators of technology!"
Bosch: Humans must retain the final control over decisions made by artificial intelligence
Local time 2, 2020 On March 19, German industrial giant Bosch also proposed a similar "AI Ethics Code", that is, artificial intelligence must be safe, reliable and explainable, and humans should always retain control over artificial intelligence.
Bosch released this guideline at the opening ceremony of the 7th Bosch Connected World (BCW) held in Berlin that day.
“Artificial intelligence is a tool that serves humans. Any artificial intelligence decision-making that may affect humans requires human supervision.” Bosch CEO Volkmar Denner ) means so.
Previously, Bosch decided to actively respond to moral and ethical issues that arise in the use of artificial intelligence technology based on the values ??embodied in the Universal Declaration of Human Rights.
Dr. Michael Bolle, chief digital officer and chief technology officer of the Bosch Group, said: "Only when people no longer regard artificial intelligence as a mysterious 'black box' will the seeds of trust sprout. And trust It will be an indispensable factor in achieving the interconnected world."
He said that the newly released Bosch "AI Code of Ethics" adheres to the concept of "technology creates the beauty of life" and combines the spirit of innovation with social responsibility. .
According to Bosch’s AI Ethics Code, artificial intelligence must make decisions with some degree of human intervention or influence. It contains three mechanisms, which adhere to the same principle, that is, in artificial intelligence products developed by Bosch, humans must retain the final control over the decisions made by artificial intelligence.
The first mechanism is human-in-command, which is suitable for situations where artificial intelligence only appears as an auxiliary tool. For example, in decision support systems, artificial intelligence assists people in classifying objects or living things.
The second mechanism is human-in-the-loop during the use phase, which is suitable for situations where the artificial intelligence system can make autonomous decisions, but humans can intervene in its decision-making at any time. For example, in the driving assistance system, the driver can directly intervene in the decision-making of the parking assistance system.
The third mechanism is human-on-the-loop during the design phase, which is suitable for applications such as emergency braking systems. When developing such smart products, experts will define parameters as the basis for AI decision-making, while humans do not participate in the decision-making, which is carried out by AI. However, engineers can retroactively check whether the machine complies with the set parameters at any time to make decisions, and modify the parameters if necessary.
Bosch plans to conduct artificial intelligence training for nearly 20,000 employees in the next two years. The AI ??Ethics Code will also be included as part of the training content.
It is said that by 2025, every Bosch product will have artificial intelligence capabilities or use artificial intelligence technology in the development and production process.
A PwC report shows that since 2018, various European organizations*** have published 44 reports making recommendations on "artificial intelligence ethics."
Facebook: Commitment to strengthen response to harmful information on the Internet
Automotive Business Review noted that whether it is Daimler or Bosch, their focus on artificial intelligence is more focused on human beings Control over artificial intelligence, but from Silicon Valley giants, their focus on artificial intelligence is more about privacy.
Artificial intelligence now appears in products such as Apple’s Siri and Face ID, and also powers Google’s search engine, self-driving cars and Facebook’s advertising business. Since January 2020, senior executives from Facebook, Google and Apple have visited Brussels because the EU will introduce policies on artificial intelligence regulation.
First, Sundar Pichai, CEO of Alphabet, Google’s parent company, came to Brussels, Belgium. Then, John Giannandrea, Apple's senior vice president of artificial intelligence, also appeared here.
Since 2018, the European Union has adopted laws and regulations on digital privacy and punished Google and other companies on antitrust issues. This has also prompted other parts of the world to adopt stricter measures against technology industry giants. action.
The European Commission’s new artificial intelligence policy may also become a template for other countries and regions to follow. Technology giants such as Google, Facebook and Apple that have bet heavily on artificial intelligence are very worried.
In a speech during his visit to Brussels, Pichai acknowledged that people are indeed worried about the possible negative impacts of artificial intelligence and the need for appropriate regulation of artificial intelligence, but he believed that "a balance needs to be found" to Make sure rules don’t stifle innovation.
On February 17, German time, Facebook CEO Mark Zuckerberg also joined the group that went to Brussels to meet with European Commission Executive Vice President Margaret ·Meet with local officials such as Margrethe Vestager. He hopes to exchange the EU's commitment to strengthen its response to harmful information online in exchange for lowering market access thresholds for foreign companies. For years, U.S. lawmakers and regulators have largely kept their hands off Silicon Valley companies, allowing them to grow unhindered with little scrutiny of issues such as the spread of false information on social networks.
Vestager likened Europe’s tougher stance on technology regulation to its regulation of agriculture. Many pesticides and chemicals allowed in the United States are banned in Europe.
She said that the European approach is that if something is risky, then the whole society wants to regulate it. "The most important thing for us is to create a system where people feel that they can be sure of what happened." society of things”.
Vestager is responsible for coordinating the first draft of the European artificial intelligence policy. Physicist Ursula von der Leyen, who officially took office as President of the European Commission on December 1, 2019, gave Vestager a 100-day deadline when she took office to meet the deadline. Publish a preliminary proposal on artificial intelligence.
Vestager said she is not worried about artificial intelligence recommending a song on Spotify or a movie on Netflix. She is concerned about deciding who gets a loan or what disease is diagnosed. Artificial Intelligence Algorithms.
She is particularly concerned about the widespread use of facial recognition technology and says new restrictions may be needed before the technology becomes "ubiquitous." At that time, it was also reported that the EU draft white paper on artificial intelligence recommended banning facial recognition technology from being used in public places within 3 to 5 years.
EU: Technology must comply with people’s rights
On February 19, the day Bosch launched the “AI Ethics Code”, European Commission President von der Leyen released the “AI Ethics Code” in Brussels. Artificial Intelligence White Paper,” which includes unprecedented regulations on how companies can use artificial intelligence.
This approximately 30-page white paper consists of two parts: the policy framework and elements of the future regulatory framework. It outlines the European Commission’s AI strategy: it must develop rapidly economically to win the technological competition; Morally consistent with continental European values.
The European Commission pointed out that artificial intelligence is developing rapidly and Europe needs to increase investment in this regard while also dealing with a series of potential risks. Von der Leyen emphasized that technology must comply with people’s rights.
The white paper proposes to establish a "trustworthy artificial intelligence framework", focusing on three major goals: developing human-centered technology; creating a fair and competitive economy; and building an open, democratic and sustainable society. .
Although the restrictions on facial recognition in the next three to five years do not appear in the officially released white paper, the white paper still formulates a series of measures to protect citizens’ privacy and data security.
As for facial recognition, the white paper states that the collection and use of biometric data for remote identification purposes poses specific risks to fundamental rights. EU data protection rules in principle prohibit the processing of biometric data for the purpose of identifying specific natural persons. Special Conditions excluded.
The white paper states: “In order to address the social concerns that may arise from the use of artificial intelligence for this purpose in public places and to avoid fragmentation in the internal market, the European Commission will launch a broad debate in Europe that may be justified. Application status (if any) and the same safeguard measures.”
At the same time, artificial intelligence companies must pass security testing and qualification review by relevant departments before they can enter the EU market. Among them, artificial intelligence companies in "high-risk" industries such as medical equipment, autonomous driving, social security and mobile payments have been listed as key review and regulatory targets.
European Union Justice Commissioner Reynders pointed out: "If a self-driving car causes a traffic accident, who will be held responsible? We must make provisions for similar issues."
The next three Within this month, the EU’s artificial intelligence white paper will accept public consultation from people from all walks of life, and will be revised accordingly based on the feedback.
At the end of 2020, the EU will formulate and promulgate legally binding digital rules such as the EU Digital Services Act, thereby making clear provisions on issues such as regulating market access, strengthening corporate responsibility, and protecting basic rights.
These regulations not only apply to local European companies, but will also have an important impact on third-country digital companies operating in the EU.
White Paper: The Beginning of Europe’s New Digital Strategy
Since 2010, the EU has adopted smart growth as its three major growth goals (smart growth, sustainable growth and inclusive growth). growth) one.
In 2013, it announced the formal implementation of two "future emerging technology flagship plans", the "Human Brain Project" and the "Graphene Project". ?The main research goal of the EU Human Brain Project is to achieve artificial simulation of the human brain. It is the world's first major scientific project in brain science.
This project is not only committed to revealing the activity mechanism of the human brain and the nature of human cognition, but its theoretical results can directly provide the most powerful support for the development of artificial intelligence. Automotive Business Review believes that this may be an important reason why both Daimler and Bosch place greater emphasis on human dominance over artificial intelligence.
With the goal of achieving smart growth, the EU has vigorously promoted the implementation of the "Digital Agenda", focusing on building a "Digital Single Market" and accelerating the "Industry 4.0" process. In 2017, the EU officially announced that its digital single market covering all member states has entered the formal construction stage.
In 2018, the European Union launched the third future emerging technology flagship program "Quantum Technology", actively developing quantum network communications and quantum computers, and providing artificial intelligence in terms of data transmission, analysis, and processing capabilities. lay the foundation for development.
From digital medicine to precision agriculture, from autonomous driving to smart cities, artificial intelligence technology has a wide range of applications and huge economic potential. Despite this, the EU has lagged behind the United States and China in artificial intelligence research and development and application.
The latest statistical report released by the World Intellectual Property Organization shows that currently more than 85% of the world’s artificial intelligence-related technology patents come from Chinese and American companies. The most detailed EU "Artificial Intelligence White Paper" released states that Europe needs to significantly increase the level of investment in artificial intelligence research and innovation. The goal is to attract 20 billion euros in AI technology research and development and application funds in the EU every year in the next 10 years.
The EU estimates that the current economic scale of digital technologies such as artificial intelligence in the EU is 300 billion euros, accounting for 2.4% of the EU's GDP. In five years, this number will be three times the current size. Industries such as artificial intelligence have created 5.7 million jobs in the EU, and this number will at least double in five years.
Currently, Bosch has established artificial intelligence centers in 7 locations around the world. As a founding member of the "CyberValley" research consortium in Baden-Württemberg, Bosch will invest 100 million euros in the construction of an artificial intelligence campus, whose 700 experts will soon be able to collaborate with external researchers and startups Work side by side.
There is no doubt that this white paper represents the beginning of Europe’s new digital strategy, and at the same time puts forward higher requirements for American and Chinese technology companies.
Therefore, Vestager said that the introduction of digital strategies such as artificial intelligence provides the EU with a second opportunity to become the world's leader in advanced technologies.
This article comes from the author of Autohome Chejiahao and does not represent the views and positions of Autohome.