Image source: @VisualChina
Text | Odd and even studio
The first country in the world to ban ChatGPT has appeared!
On March 31, Italy issued a blocking order on ChatGPT, and from now on, the use of chatbot ChatGPT was completely banned, and a case investigation began.
Within 20 days, if OpenAI fails to come up with a remedy, it will face a fine of up to 20 million euros, equivalent to about 150 million yuan!
This may only be the beginning of Europe's "encirclement" of ChatGPT. Germany is likely to follow Italy's lead by blocking chatbot ChatGPT on data security grounds.
Italy bans ChatGPT
According to the Italian news agency Ansa, on March 31, local time, the Italian Personal Data Protection Agency announced:
The Italian Personal Data Protection Agency, which lists a series of ChatGPT offences:
Although the terms of use published by ChatGPT are aimed at users over the age of 13, there is no age verification system, so minors may be exposed to illegal information.
In addition, ChatGPT does not inform about the collection and processing of user information, and lacks a legal basis for collecting and storing personal information in large quantities.
In addition, the Italian side also specifically pointed out that ChatGPT has had user data leaks.
It is reported that on March 23 this year, there was a serious problem with ChatGPT. Some netizens found that other people's chat history appeared in their chat history column.
According to statistics, the personal payment information of 1.2% of paid version users may be leaked. Some users may see information such as the name, email, payment address, 4 code after the credit card, and credit card expiration time of another active user.
At the time, Sam Altman, the father of ChatGPT, also tweeted an apology.
Late that night on March 31, OpenAI said it had disabled ChatGPT for Italian users at its request.
Sam Altman, the founder of OpenAI, seems to believe he is not at fault, saying he believes his approach complies with European privacy laws and hopes to launch ChatGPT again soon.
"Italy is one of my favourite countries and look forward to visiting again soon".
Since then, Italy has become the first country in the world to announce a ban on ChatGPT.
And right away, the second country may also come.
A fine of 20 million euros, Germany will follow?
The Italian Personal Data Protection Agency, which issued the ban, is not directly subordinate to a government agency, but is appointed by Parliament.
This time, with just one vulnerability, Italy is going to disable AI, is it too harsh?
But as Johanna Björklund, an associate professor of artificial intelligence at Umo University in Sweden, says: "The lack of transparency is the real problem. ”
At present, the world's first ChatGPT "ban" still has room for maneuver. Italy gave OpenAI 20 days to solve the hidden danger of leaking personal privacy.
Otherwise, Italy will impose a fine of 20 million euros, about 150 million yuan, on OpenAI.
This penalty is based on the General Data Protection Regulation (GDPR) passed by the European Parliament in 2016, and the fine can even be as high as 4% of annual turnover.
OpenAI is not alone.
On April 3, Handelsblatt published comments from a spokesman for Germany's Federal Data Protection Commissioner, who said that Germany could follow Italy's lead in blocking the chatbot ChatGPT, citing data security concerns.
The prohibition of the use of ChatGPT will fall within the purview of the German Federal Data Protection Authority.
The European Consumer Rights Coalition (BEUC) is also calling on the European Union and other government agencies to thoroughly investigate ChatGPT and similar chatbots.
I just don't know, EU companies, do they have the ability to launch their own ChatGPT-like AI?
The European Union is currently working on the world's first legislation on artificial intelligence, but the European Consumer Rights Coalition is concerned that the AI Act will take years to come into force, putting consumers at risk of being harmed by inadequate regulation.
A spokesman for the European Commission said:
It's worth noting that ChatGPT is now unavailable in some countries, often because of OpenAI blockades.
For example, OpenAI has never opened ChatGPT services for China, Russia, Iran, etc.
Samsung and Amazon have disabled ChatGPT
This isn't the first time ChatGPT has sounded the alarm about data security.
According to media sources, Amazon's lawyers have warned employees not to share with ChatGPT: "Any confidential information about Amazon, including the code you are writing, because they may use it to train future models." ”
U.S.: Bank of America, Citigroup, Deutsche Bank, Goldman Sachs, Wells Fargo, JPMorgan Chase and Verizon have banned employees from using ChatGPT for work tasks.
Japanese companies such as SoftBank, Fujitsu, Mizuho Financial Group, Mitsubishi UFJ Bank, Sumitomo Mitsui Banking Corporation, and others have also restricted the commercial use of ChatGPT.
However, these internal warnings seem to be of little use.
Within Samsung, there have been 3 cases of abuse of ChatGPT, involving semiconductor equipment measurement data, product yield and other content. It is said to have been stored in ChatGPT's database.
Due to the hidden dangers of using ChatGPT, Samsung is considering developing a Samsung-exclusive internal AI service through an internal innovation center.
Perhaps, when the Samsung chip leaks enough data, we can find ChatGPT to teach us to make high-end chips.
Data security firm Cyberhaven, which surveyed 1.6 million employees in different industries about ChatGPT usage, reported that 2.3% of employees copied confidential company data to ChatGPT.
On average, employees leak sensitive data to ChatGPT hundreds of times a week, and sensitive data currently accounts for 11% of the content employees paste into ChatGPT.
It seems that foreigners are often willing to exchange personal privacy for the convenience of AI.
epilogue
OpenAI's CEO Sam announced this week that he will embark on a six-continent tour in May to discuss OpenAI's technology with users and developers.
Some analysts believe that OpenAI is trying to clear obstacles for the global adoption of ChatGPT.
When a high-intensity regulatory ban comes out, it is naturally accompanied by a high-density lobbying process.
Musk is one of the founders of OpenAI, but has now exited the company.
He has repeatedly expressed the threat posed by AI, and in February he said that AI is one of the greatest risks to future human civilization.
"It has a positive side and a negative side, it is promising, it is powerful, but it also comes with great risks."