Development of commercial robots: What happens when the market treats the market?
propagandaLas Malicious commercial practices They are not new. Control of the use of distinctive information, as well as against the various forms of market processing, is a long period A battle for the organizers.
However, in recent years, experts have warned New threats For us Financial systems. AI progress makes mechanical negotiation robots not only smarter, but also more independent. While the basic algorithms respond to scheduled orders, New robots They are able to Learn from the experienceQuickly assemble large quantities of information and act independently when performing operations.
According to academics, a The risk scenario meansCooperation between artificial intelligence robots. Imagine: hundreds of personal files Social networks It starts from AI to appear on the Internet, and weave accounts on some companies. The information that is not necessarily wrong, but it can be simply Expanding the current news. In response, the real users of social networks begin to respond, with a highlight of the message chosen by robots.
Since the market tends to the narration shown, roboadvisor From the investor Get the benefitsAfter coordination with Cotillas robots. Other investors, who did not have distinctive information, lose Bad synchronization with the market. The problem is that the investor is that the benefits may not be aware of the plot. This means that the accusations of manipulation of the market should not be effective, although the authorities can see that the operator has benefited from distorted practices.
Social platforms change trade
Alessio Azouti, Professor of Law and Technology attached to the University of Glasgow, told “Euronews” that the previous scenario It is still a hypothesisOn condition There is not enough evidence To show that it happens. However, it shows that the similar and less developed plans occur, especially in “Covering markets And decentralized financial markets. “
Los Microcarbon actors (…) They can be very active in social media platforms and correspondence platforms such as Telegram, where they can encourage their members to invest their money in Defi or in a specific encoded activity, to their “comfort.” We can notice the direct activity of the malicious human actors, but also those who display the robots of Amnesty International. “
He added that the agents who publish wrong information can Not necessarily very sophisticatedBut they still have the ability to “pollute chats through wrong news to Determination of retailers“.
Then the question is, if a young man alone in his local office is able to achieve this type of manipulation, what is it border So that the great actors achieve the same effect, in more advanced markets? “
The way market information is now published online, in a generalized, fast and unaccusable manner, also promotes Various types of negotiation. Private investors more artificial intelligence.
Gamestop Saga, which has been widely martyred, is a good example of negotiating in Manada, when Reddit Forum users are They decided to purchase the shares of Mass Game. Large coverage boxes are betting that the price will decrease, and later lost when it was shot. Many experts claim that it was not a complicity issue, as there was no official agreement.
Spokesman Esma, European Authority for Securities and MarketsHe told “Euronews that the capabilities of artificial intelligence robots to manipulate the markets and benefit from movements are” a realistic concern “, although they have confirmed that they have no” specific or statistical information already occurred. “
“These risks intensify more than one role Social networksAnd that can be like a A quick transmission channel from wrong novels Or you were deceived by the market dynamics. AEVM spokesman said that the degree of human control over these systems, because traditional supervision mechanisms can be insufficient. “
Is it consistent?
One of the challenges facing the organizers is that cooperation between agents of artificial intelligence It cannot be easily tracked. “Email messages are not sent or collected between them. They learn over time He explains the best strategy, so the traditional way to discover collusion with artificial intelligence “explains” Euronews “ITAY Goldstein, Professor of Finance and Economy at the Warton School at the University of Pennsylvania.
“The regulation must be presented forward and found New strategies To deal with this, “argue, adding that reliable data on How to use artificial intelligence The operators.
propagandaFilippo Anounisata, professor of financial markets and banking legislation at the University of Bokonews, said that the current European Union standards “They should not review”In reference to the regulation of market abuse (MAR) and guidance on the II (MIFID II) financial tool markets. However, he argued that “supervisors must be equipped More sophisticated tools To determine as possible Manipulation From the market.
He added: “Until I suggest that we ask those who develop artificial intelligence tools to work in the markets and others that include isolation in these artificial intelligence tools. This will even be forced to stop Before the risk of manipulation occurs“.
As for the current legal framework, there is also a question of responsibility when AF is a harmful manner, regardless of human intention. This is especially important in the case of so – Black monetary operationsAs the robot carries out operations without detecting its internal operation. To deal with this situation, some experts believe that artificial intelligence should be designed to be more transparent, so that the organizers can understand the logic on which the decisions are based.
Another idea is to create new laws on responsibility, so that the actors responsible for spreading artificial intelligence can be considered Responsible for manipulation of the market. This can be applied in cases where they had no intention to deceive investors. “It is somewhat similar to turtles and rabbit, and the supervisors are usually a turtle, but the manipulators who use algorithms are rabbits, and it is difficult to reach,” says Annusata.
propaganda
