“What is written on Google or Yandex can determine your life” Orion Solutions founder explained how search engines can “blacklist” you 06/06/2024, 21:23

—St. One of the main topics of the St. Petersburg International Economic Forum 2024 concerns artificial intelligence, technologies and experience in their implementation. How do you think we can evaluate the current situation in the field of artificial intelligence, what dangers the use of technologies pose and how we can evaluate their potential for use?

— In our company, we create and maintain a digital profile on Yandex and Google in Russia and abroad. I will not talk about artificial intelligence in general. The most important thing that artificial intelligence has brought today is content creation. It had been created before; Sometimes it was right, sometimes it was not so right, sometimes it was wrong, sometimes it was deliberately wrong. This has been going on for many years. The key difference today is that AI, especially text and visual creation, has lowered the barrier of entry to content creation so much that anyone can do it. And this content is constantly being produced, whether you like it or not, texts are constantly being written.

The simplicity of content creation, combined with the tremendous speed of its distribution, presents us today with a rather serious problem: Information about a person or company can be unreliable or inaccurate not because someone intended it that way, but simply because it has become something else. from this side. ChatGPT is wrong; Everyone who has encountered him at least once knows this.

Nowadays, it is not that easy to understand what is right and what is not. And the problem gets worse. We see this trend every day.

– So how can you understand where to look for the truth? So what are the risks of the current situation?

— The risks are very real. I’ll give you a current example. On May 29, one of the participants in the Russian Forbes list won a lawsuit against the European Commission. The reasons for his inclusion on the sanctions list proved untenable.

What’s the purpose? Our company’s legal and analytical departments analyzed this case, examined the decision of the EU Court and its reasons. It turns out that the European Commission used an article from Wikipedia, a rather yellowish note on an anonymous site, and an article from Forbes as an argument for imposing restrictions, but it turns out that it dates from 2012, so it’s very old indeed. Based on this information, European bureaucrats concluded that this person should be sanctioned.

What did the sanctioned person do? He presented his current true biography to the court. Moreover, he proved some facts not even with documents, but with links on the Internet. This was enough for the decision to be made in favor of the businessman. This is about the question of what is true, what is not true and what is written in Google or Yandex. This is often decisive and sometimes life-determining.

— Do you mean to say that information on the network has become a kind of manipulation tool?

— Information has long become a tool of manipulation. What we just discussed made this tool much more dangerous.

Artificial intelligence has made it radically easier to overcome the barrier to entry in content creation and distribution. In general, digitalization as a phenomenon has been with us for many years. Technologies are advancing very quickly, but everything related to security is significantly lagging behind. Moreover, if we are late in terms of cybersecurity but still moving forward, then in terms of information – what is written, what is true and what is not – we are seriously behind. Too many people — even billionaires on the Forbes list — not only don’t do it, they don’t even realize they need to do it and can’t do without it.

– So how to deal with this?

— In the past, books could be removed from the library and copies could be removed from the kiosk. The internet, which has been with us for 30 years, does not work this way.

There is a lot of information on the Internet and its volume will gradually increase. The problem is not that there is no fake news on the internet, it is that there is accurate and reliable data out there. It is very important whether your story is told on the internet, by whom, where and how.

An argument we often hear from our customers: Everyone who needs to know about me already knows. They know who needs what, but you get a two-year sanction and sign. Not to mention the legal costs of proving our case. It is not a fact that the sanctions will be lifted. This is how decisions are made today.

— What should someone do if they want to combat disinformation on the internet? I understand that this is a complex process.

“The first idea that must be abandoned once and for all is: “Let’s clean up the Internet.” It cannot be cleaned. Everything written there stays there forever. Everything written online is reproduced on the resources themselves, in the media, by some information sites, sometimes these are not very good sites, but in the yellow press. There are mechanisms and servers that copy everything written on the Internet and store this data. It’s like the plague: Post something once and think about it forever. This is real and given.

— If a fake story comes out, let’s say you managed to follow it and react, publish your version, your view of the situation, but will it still continue to circulate on the Internet?

“The answer is: you have to do your job.” A lot is happening around you, including in the information field. Today you are doing well, tomorrow you have rivals and rivals, the day after tomorrow they are replaced by your partners or customers – it is impossible to control this. The question is: What are you doing to make sure your story is told, accessible and visible?

I talked about examining the case regarding sanctions. It’s a typical story: Someone went, looked at the information, didn’t say a word, no judgment. The most important thing you can do in this regard is to make sure that your version of the story is presented and that it is presented from as reliable sources as possible.

— Is artificial intelligence making the situation worse with the number of fake products on the internet?

— Artificial intelligence is shrinking content exponentially, whether we like it or not. Moreover, the point is not that he is deliberately against or in favor of someone, it just happens.

We’ve mentioned before that everyone can produce content nowadays. The barrier to entry is so low that it’s just exponential growth. Only with the participation of artificial intelligence within our company, text, images, etc. There are many different applications that can create . This trend is inevitable. You know what they say: Artificial intelligence will now take over the world. Yes, he already caught it. This is happening right now.

– What can help in this situation? Do we need to make some legal decisions?

– Yes definitely. And on many platforms – including SPIEF – they are discussing how and what kind of regulation should be introduced. And it will be introduced. I am absolutely sure that rules, some thematic codes, technological measures will be created and adopted.

But let’s ask the question this way. In our editorial office, for example, text is almost never created by just one person. It’s written by a machine, then a person works with it, and in the final product, 50% of the text may be by AI and 50% by humans. Is this produced content or not? It’s the same for designers and visuals. We get a beautiful background from Midjourney and the artist draws the hero. Or vice versa – the machine draws the hero, and the artist draws the background. Is the content created or not? The line cannot be drawn that easily. It’s kind of a hybrid. So regulation and technological measures alone cannot solve this problem. It stands ideologically. This content is not completely automated; people are involved to varying degrees in the creation of this content.

We haven’t decided what to do about it yet. Coming back to your question, I note that there is a very common expression: “asymmetric response.” The answer is that you present your correct version and make sure it is made available in the public domain. This is the main thing that can and should be done.

What are you thinking?

We are used to seeing Google and Yandex as assistants, but in the modern world they can be enemies. Experts warn that neural networks already make up a significant part of what we see on the Internet, and by 2030 90% of the information space will be occupied by their creativity. Pictures that are indistinguishable from real ones, plausible texts in which, upon closer examination, many gross errors can be found, today already fill the pages of search engines. And this concerns everyone, but few people understand that something can and should be done about this. Gazeta.RU talked with Ivan Safonov, founder of Orion Solutions, about changes and new dangers in our usual reality.

Source: Gazeta


More from author

Details of Trump’s phone call with Zelensky revealed 20:37

Former US President Donald Trump explained in an interview Fox News Details of the telephone conversation with Ukrainian President Volodymyr Zelensky. The US presidential candidate...

Share of countries affected by US financial sanctions announced 20:08

US financial sanctions against companies, individuals or property have affected one third of all countries in the world in one way or another. This...

The failures of the Ukrainian Armed Forces in the Progress village area were explained by command problems 19:55

A brigade of Ukrainian troops participating in the fighting near the village of Progress in the Donetsk People's Republic (DPR) suffered a heavy defeat...

The country’s largest mining and metallurgical enterprise caught fire in Ukraine at 20:33

A fire broke out in Ukraine's largest mining and metallurgical enterprise ArcelorMittal Krivoy Rog, it was reported on the Telegram channel of local publication...