Group risks: Websites won't have to keep registration records for three years
Website owners will not need to store user registration and authorization data for three years. This provision has disappeared from the new edition of the package of anti—fraud bills "Antifraud 3.0" - it was prepared and sent to the interdepartmental coordination of the Ministry of Finance. The norm could help fraud investigations, but would create additional risks of data leakage, experts say. At the same time, site owners would incur additional costs for their storage infrastructure. About what other provisions have disappeared from the new version of the anti—fraud bill, see the Izvestia article.
What provisions have disappeared from "Anti-fraud 3.0"
Owners of websites, information systems, and applications will not need to store user registration, cancellation, and authorization data linked to a phone number for three years. This follows from the new edition of the third package of anti-fraud laws (the so-called "Anti-fraud 3.0"), which was reviewed by Izvestia. The Ministry of Finance has already sent out the document for interdepartmental approval, according to a source familiar with officials from several relevant departments.
At the moment, the third package of anti-fraud measures is under discussion, so it's too early to talk about which measures will be included in the final version, the Ministry of Finance told Izvestia.
— All proposals are being worked out taking into account the balance between security and user rights. Their goal is to strengthen the protection of citizens from fraudsters on the Internet," they said.
The storage of registration data is not the only provision excluded from the new version of the draft law. For example, the block disappeared from it, which provided, in particular, for the transfer of consent to the processing of personal data and its withdrawal through Gosuslugi. The rule requiring Internet sites to verify the accuracy of information about the phone numbers of foreigners working for them, and a number of others, has been removed.
The storage of registration data was conceived as a tool for a retrospective analysis of the actions of intruders, said Akhmetzhan Makhmutov, ex-Minister of Digital Development of the Vologda Region.
"In cases of cyberbullying, phishing, and illegal use of personal data, it is critically important to restore the chronology: who accessed the resource, when, and from which device, what actions they performed, and how they changed credentials," he explained to Izvestia. — Scammers often cover their tracks by deleting accounts, using disposable SIM cards, proxy servers, and virtual machines. If the site owner has been storing information about registration and authorization for several years, the investigation has the opportunity to go back and build an evidence base.
But such a practice would lead to an increased burden on business, Makhmutov continued. Storing a three—year array of data on user actions requires not only technical infrastructure, but also compliance with information security requirements - this is often unaffordable for small Internet resources, he explained. It is also not always clear what exactly is considered "registration data" — only the login and email address, or also the IP address, digital fingerprints of the device, and the password change history. The accumulation of such volumes of personal data in itself creates huge risks of leaks. The regulator probably concluded that the benefits of storage do not outweigh the potential harm, Akhmetzhan Makhmutov believes.
What is the danger of storing registration data?
The meaning of the provision on storing registration data in conjunction with the provision on authorization by subscriber number proposed in the draft law is obvious: to get a solid log of entries that allows you to quickly connect the account, phone number and actual user, said Yaroslav Shitsle, head of the It&Ip Dispute Resolution department at Rustam Kurmaev & Partners.
But legally, the norm had vulnerabilities, he believes. In particular, it covered any resource with a personal account, from an online store to an industry service, and not just sites of interest to law enforcement officers. The financial and economic justification for the bill stated only that there were no budget expenditures, but the business costs of building and protecting log storage facilities were not estimated at all, Yaroslav Shitsle added. Finally, a log leak is more dangerous than a leak of personal data, since it allows you to restore the range of services and the daily routine of a particular person, and when comparing several arrays, deanonymize the user, the lawyer pointed out.
Most likely, the initial norm was considered too broad, according to Oleg Yablokov, NTI expert on wireless communications. It covered any website with registration or authorization, although some services already store similar information according to the current rules for information dissemination organizers. For a small store, a forum, and a large social network, the same three-year obligation was established, and the composition of the data that needed to be saved was not clearly defined, the expert lists.
— The longer the information is stored, the greater the potential damage from its leakage. Log-in logs can contain account IDs, phone numbers, email addresses, IP addresses, and activity times. Such information simplifies targeted phishing, allows you to link different accounts and restore the history of human actions. For resource owners, these are additional costs for secure storage, access control, and secure deletion of data," Oleg Yablokov pointed out.
Leonid Konik, a partner at ComNews Research, noted that users often leave fictitious names and secondary email addresses on free hosting sites to register on information sites. Therefore, such information is of little use to identify a specific person, he believes.
In general, the risks of data leakage outweigh the need to introduce new anti-fraud tools, says Vladimir Zykov, editor-in-chief of the analytical publication Runet. Now these risks have become more acute in the light of the development of AI, he notes. Recent weeks have shown that AI models are increasingly hacking into secure information systems of companies. Therefore, the refusal to create an additional source of possible leaks is a logical step, the expert concluded.
Переведено сервисом «Яндекс Переводчик»