In world first, EU's sweeping AI law enters into force
Agence France-Presse
August 1, 2024
The EU's sweeping risk-based rules will cover all types of artificial intelligence (Fabrice COFFRINI/AFP)
The European Union's landmark law on artificial intelligence came into force on Thursday, which Brussels vows will drive innovation while protecting citizens' rights.
The EU earlier this year adopted the world's first sweeping rules to govern AI, especially powerful systems like OpenAI's ChatGPT after difficult and tense negotiations.
Although the rules were first proposed in 2021, they took on greater urgency when ChatGPT burst onto the scene in 2022, showing generative AI's human-like ability to churn out eloquent text within seconds.
Other examples of generative AI include Dall-E and Midjourney, which can generate images in nearly any style with a simple input in everyday language.
"With our artificial intelligence act, we create new guardrails not only to protect people and their interests, but also to give business and innovators clear rules and certainty," European Commission President Ursula von der Leyen said.
Companies will have to comply by 2026 but rules covering AI models like ChatGPT will apply 12 months after the law enters into force.
Strict bans on using AI for predictive policing based on profiling and systems that use biometric information to infer an individual's race, religion or sexual orientation will apply six months after the law enters into force.
The law known as the "AI Act" takes a risk-based approach: if a system is high-risk, a company has a stricter set of obligations to fulfil to protect citizens' rights.
The higher the risk to Europeans' health or rights, for example, the greater the companies' requirements to protect individuals from harms.
"The geographic scope of the AI Act is very broad, so organisations with any connections to the EU in their business or customer base will need an AI governance programme in place to identify and comply with their obligations," said Marcus Evans, partner at law firm Norton Rose Fulbright.
Companies in violation of the rules on banned practices or data obligations face fines of up to seven percent of worldwide annual revenue.
The EU in May established an "AI Office" of tech experts, lawyers and economists under the new law to ensure compliance.
August 1, 2024
The EU's sweeping risk-based rules will cover all types of artificial intelligence (Fabrice COFFRINI/AFP)
The European Union's landmark law on artificial intelligence came into force on Thursday, which Brussels vows will drive innovation while protecting citizens' rights.
The EU earlier this year adopted the world's first sweeping rules to govern AI, especially powerful systems like OpenAI's ChatGPT after difficult and tense negotiations.
Although the rules were first proposed in 2021, they took on greater urgency when ChatGPT burst onto the scene in 2022, showing generative AI's human-like ability to churn out eloquent text within seconds.
Other examples of generative AI include Dall-E and Midjourney, which can generate images in nearly any style with a simple input in everyday language.
"With our artificial intelligence act, we create new guardrails not only to protect people and their interests, but also to give business and innovators clear rules and certainty," European Commission President Ursula von der Leyen said.
Companies will have to comply by 2026 but rules covering AI models like ChatGPT will apply 12 months after the law enters into force.
Strict bans on using AI for predictive policing based on profiling and systems that use biometric information to infer an individual's race, religion or sexual orientation will apply six months after the law enters into force.
The law known as the "AI Act" takes a risk-based approach: if a system is high-risk, a company has a stricter set of obligations to fulfil to protect citizens' rights.
The higher the risk to Europeans' health or rights, for example, the greater the companies' requirements to protect individuals from harms.
"The geographic scope of the AI Act is very broad, so organisations with any connections to the EU in their business or customer base will need an AI governance programme in place to identify and comply with their obligations," said Marcus Evans, partner at law firm Norton Rose Fulbright.
Companies in violation of the rules on banned practices or data obligations face fines of up to seven percent of worldwide annual revenue.
The EU in May established an "AI Office" of tech experts, lawyers and economists under the new law to ensure compliance.
July 30, 2024
Artificial intelligence is becoming more common in everyday life. Some companies are exaggerating the use of AI when marketing their products and services, which could impact our relationship with AI technology.
Artificial intelligence is becoming more common in everyday life. Some companies are exaggerating the use of AI when marketing their products and services, which could impact our relationship with AI technology.
AI washing may frustrate investors as they miss out on funding and real genuine projects
Marc Asensio/NurPhoto/IMAGO
Imagine this: You want to buy the latest vacuum cleaner, you go to your favorite online store and begin to scroll through your cellphone or tablet. Then, you see a product that claims it is powered by advanced artificial intelligence (AI) technology, which makes it sound super high-tech and smart.
But really, it just has a simple sensor button to make sure it doesn't bump into furniture. Nothing special or groundbreaking. That's AI washing. It means making products sound way cooler or smarter by claiming they have AI when they are just doing regular stuff.
Why is AI washing a problem?
First of all, real AI innovation can get buried under all the hype, making it hard to spot real breakthroughs. Secondly, it makes people skeptical about AI claims, so they stop trusting the technology altogether.
Christoph Lütge, from the Munich-based Institute for Ethics in Artificial Intelligence, says that since the term AI is used in very different ways, there are so many concepts which make it easier for companies to hype it up.
"The challenge is that it's hard to deal with AI from a legal perspective because the term is too vague. From an ethical point of view, it can work. It's good to have an expert who comes in and guides regulatory bodies and civil society from an ethical point of view," he told DW.
It's also a headache for investors. They might miss genuine projects because of all the noise. Companies may also begin to aim for impossible goals, believing that AI could do more than it really could.
Earlier this year, the US Securities and Exchange Commission (SEC) announced settled charges against two investment firms, Delphia and Global Predictions, for making false and misleading statements about their purported use of artificial intelligence. The firms agreed to settle the SEC's charges and pay a total of $400,000 (€368,000) in fines.
According to the SEC, Toronto-based Delphia made false claims that it "put[s] collective data to work to make our artificial intelligence smarter so it can predict which companies and trends are about to make it big and invest in them before everyone else" when in fact it did not have the AI and machine-learning capabilities that it claimed.
Meanwhile, San Francisco-based Global Predictions had wrongly claimed on its website and on social media that it was the "first regulated AI financial advisor" and misrepresented that its platform provided "expert driven forecasts."
"We've seen time and again that when new technologies come along, they can create buzz from investors as well as false claims by those purporting to use those new technologies," SEC chair Gary Gensler said in a press statement. "Investment advisers should not mislead the public by saying they are using an AI model when they are not. Such AI washing hurts investors."
Major companies like Coca-Cola and Amazon have also faced backlash for allegedly engaging in so-called AI washing. In September, Coca-Cola launched the Y3000 drink, a futuristic new flavor of cola that it claims was "co-created" by AI.
Imagine this: You want to buy the latest vacuum cleaner, you go to your favorite online store and begin to scroll through your cellphone or tablet. Then, you see a product that claims it is powered by advanced artificial intelligence (AI) technology, which makes it sound super high-tech and smart.
But really, it just has a simple sensor button to make sure it doesn't bump into furniture. Nothing special or groundbreaking. That's AI washing. It means making products sound way cooler or smarter by claiming they have AI when they are just doing regular stuff.
Why is AI washing a problem?
First of all, real AI innovation can get buried under all the hype, making it hard to spot real breakthroughs. Secondly, it makes people skeptical about AI claims, so they stop trusting the technology altogether.
Christoph Lütge, from the Munich-based Institute for Ethics in Artificial Intelligence, says that since the term AI is used in very different ways, there are so many concepts which make it easier for companies to hype it up.
"The challenge is that it's hard to deal with AI from a legal perspective because the term is too vague. From an ethical point of view, it can work. It's good to have an expert who comes in and guides regulatory bodies and civil society from an ethical point of view," he told DW.
It's also a headache for investors. They might miss genuine projects because of all the noise. Companies may also begin to aim for impossible goals, believing that AI could do more than it really could.
Earlier this year, the US Securities and Exchange Commission (SEC) announced settled charges against two investment firms, Delphia and Global Predictions, for making false and misleading statements about their purported use of artificial intelligence. The firms agreed to settle the SEC's charges and pay a total of $400,000 (€368,000) in fines.
According to the SEC, Toronto-based Delphia made false claims that it "put[s] collective data to work to make our artificial intelligence smarter so it can predict which companies and trends are about to make it big and invest in them before everyone else" when in fact it did not have the AI and machine-learning capabilities that it claimed.
Meanwhile, San Francisco-based Global Predictions had wrongly claimed on its website and on social media that it was the "first regulated AI financial advisor" and misrepresented that its platform provided "expert driven forecasts."
"We've seen time and again that when new technologies come along, they can create buzz from investors as well as false claims by those purporting to use those new technologies," SEC chair Gary Gensler said in a press statement. "Investment advisers should not mislead the public by saying they are using an AI model when they are not. Such AI washing hurts investors."
Major companies like Coca-Cola and Amazon have also faced backlash for allegedly engaging in so-called AI washing. In September, Coca-Cola launched the Y3000 drink, a futuristic new flavor of cola that it claims was "co-created" by AI.
Is it a thing in Germany?
Some companies may advertise AI in their products before they launch it. Even though they intend to include AI in the product at some point, they may advertise as if the product already contains AI, which may not be the case.
According to Joerg Heidrich, an AI lawyer and certified AI specialist based in Hannover, Germany, it's not that people are not aware of AI washing in Germany, but rather the concept of how its applied in products and services.
"We are in an era of false advertising, with companies claiming that they do have something which in fact does not exist. I see that really a lot in two areas where I work. First one is in the area of legal tech, where every legal-tech company is advertising that it uses AI which is largely false and another area where I see it, is in cyber security, where I work a lot," Heidrich told DW.
A 2019 report by London-based venture capital firm MMC Ventures found that out of more than 2,830 startups in Europe are classified as AI companies. However, 40% percent had nothing to do with AI at all.
The EU AI Act
The EU AI Act comes into effect on August 1, 2024, with provisions coming into force gradually over the following six to 36 months. It has a limited risk category, which ensures that users are informed that they are interacting with AI systems, allowing them to make informed decisions.
However, Heidrich says: "Although the AI Act is not responsible to regulate the area of AI washing, it might be helpful because the act ensures organizations are very transparent about AI so this might help indirectly limit false advertising from companies due to transparency requirements."
The European Consumer Center (ECC) in Germany is the first contact point for all German consumers with cross-border issues and for foreign consumers living in Germany. Having dealt with disputes to do with green and social washing, they are sure that the topic of AI washing will come up.
"[But] consumers must be able to prove this claim and that there is reliable proof of that because now everyone tries to use AI as a buzzword," ECC spokesperson Karolina Wojtal told DW.
Edited by: Uwe Hessler
Rachel Nduati A multimedia journalist with a focus on food security, climate change, technology and human rights.
No comments:
Post a Comment