Germany is considering blocking ChatGPT, France, Ireland, Spain may also join! Why is Europe "encircling" ChatGPT?

Per reporter: Wen Qiao Per editor: Lan Suying

On April 3, local time, according to the German "Handelsblatt", German data commissioner Ulrich Kelber (Ulrich Kelber) said that Germany may follow Italy's example and block OpenAI's chatbot ChatGPT on the grounds of data security concerns. "In principle, such actions are possible in Germany," Kelber said. But he stressed that it would fall under national jurisdiction.

Kelber did not outline any current plans for such action, but he revealed that Germany had requested more information from Italy on its temporary ban. In addition to Germany, European countries such as France, Ireland and Spain have also begun to consider stricter regulation of artificial intelligence (AI).

In addition to ChatGPT, legal experts say Google's Bard may also be subject to scrutiny, though given Google's long experience with regulators, the company may have taken this into account.

Why did Europe fire the first shot to ban ChatGPT? Some analysts believe that the considerations behind this are closely related to the EU's anti-monopoly crackdown on US technology companies in recent years. In recent years, American tech giants have been hit in Europe in a series of major accusations against Google, Apple, Facebook, Amazon and Microsoft, including misuse of personal data and stifling competition.

The analysis believes that the EU side was originally very dissatisfied with the US technology giants grabbing user data, coupled with the large-scale deployment of ChatGPT is likely to violate EU competition law, it is not surprising to tighten the regulation of AI tools represented by ChatGPT.

Image source: Visual China - VCG111420147041

It's not just ChatGPT, Google is dangerous too

According to Reuters, European lawmakers remain divided on the content and scope of the upcoming EU AI bill, but some regulators have found that existing tools such as the General Data Protection Regulation can be applied to the rapidly emerging generative AI category.

The legal tool cited by Italy's ban on ChatGPT is the European Union's General Data Protection Regulation. Privacy regulators in France, Ireland and Spain have reached out to their Italian counterparts to learn more about the basis for the ban.

It is worth noting that privacy regulators in Europe tend to be independent of the government and have a lot of power over privacy protection. For example, Italian privacy regulator Garante made a decision to ban ChatGPT independently of the government, which the Italian government does not support, and the Italian deputy prime minister criticized the decision of his regulator, calling it "excessive."

In addition, a German government spokesman also told the media that there is no need to ban ChatGPT.

Reuters quoted several legal experts as saying that while Italian regulators have so far only targeted the hottest ChatGPT, other AI chatbots such as Google's Bard may also come under scrutiny.

However, Google has long been a thorn in the side of EU regulators, and Google has extensive experience in how to "deal with" institutions. "Unlike ChatGPT, Google is more likely to have taken this into account due to its history and size in Europe." Defsislava Savova, a partner at law firm Clifford Chance, said.

According to the Corporate Europe Observatory, an independent lobbying group, Google's lobbying of senior European Commission officials topped the list of tech giants and met the most MEPs. OpenAI, an emerging unicorn, doesn't even have a European office. After the ban was made in Italy, it also only took ChatGPT offline in Italy, without directly responding to questions.

Next, Europe and the world are set to tighten the regulation of AI. It is reported that the EU AI bill will be officially voted on from this month, and generative AI will also be included as a key object of supervision.

Behind ChatGPT

Why did Europe fire the first shot to ban ChatGPT?

Some analysts believe that the considerations behind this are closely related to the EU's anti-monopoly crackdown on US technology companies in recent years.

In recent years, American tech giants have taken a series of blows in Europe. Previously, the EU filed a number of major charges against US tech giants such as Google, Apple, Facebook, Amazon and Microsoft, including misusing personal data, stifling competition, failing to pay enough taxes, stealing media content and spreading fake news.

It is worth noting that the large-scale language models represented by ChatGPT, which are now popular around the world, require a large amount of data during the training process, and Italy's consideration of blocking ChatGPT is precisely because of concerns about privacy data. The analysis believes that the EU side was originally very dissatisfied with the US technology companies grabbing European data to gain profits, and in recent years the EU has also carried out more detailed supervision of data.

In 2018, the European Union took the lead in imposing controls on tech giants through the General Data Protection Regulation, which has become an international reference. In 2022, the EU approved the Digital Markets Act and the Digital Services Act, setting new rules for establishing a fair competition in the digital industry.

Zoltán Marosi, co-head of DLA Piper's Competition & Antitrust practice in Hungary, argued in an article that the widespread deployment of ChatGPT is likely to violate EU competition law.

He believes that the first is the possibility of abuse of market dominance. With more than 100 million users in February 2023, ChatGPT could spark disruptive competition on a broader scale, such as competitive pressure on general-purpose internet search engines. For consumers, it can replace or possibly even outperform universal search services.

Marosi said that while ChatGPT has clever disclaimers — OpenAI positions itself as a digital service provider, placing itself in a narrow market for natural language processing and generation. The barriers to this market are high, and the cost of training large language models is increasing exponentially, so ChatGPT or other similar service providers may still dominate the market.

Second, it may constitute unfair competition. Marosi writes, "One problem with generative AI is that it intentionally or unintentionally generates content similar to that of other companies or brands. Tools including ChatGPT, DALL-E can be seen as direct competitors to the artists or designers they can emulate. When AI produces content in an artist's unique style, if that text or image is adopted by the company, it can harm the artist and cause significant losses, which constitutes unfair competition. ”

Third, the possible illegal acquisition, use and disclosure of trade secrets. "If ChatGPT obtains trade secrets during the training process, including information inadvertently obtained on the Internet, users copy-pasting customers' information, these actions may violate the EU ban." He added.

Daily economic news