In the banking sector, subcontracting is the detailed credit analysis that precedes the granting of a loan, based on the credit information provided by the borrower; Such an acquisition is in several areas: it allows investors to distribute a new issued security, such as stocks or bonds, AD. A banking consortium (lead managers) takes over the transaction, i.e. they have taken the risk of the distribution of the securities. If they are not able to find enough investors, they must hold certain securities themselves. Insurers derive their income from the “underwriting spread” between the price they pay to the issuer and what they collect from investors or brokers who purchase portions of the offer. Insurers may refuse risk or submit an offer in which premiums have been charged (including the amount required to make a profit, in addition to expense coverage) or exclusions that limit the circumstances under which a fee would be paid. Depending on the type of insurance product (sector), insurance companies use automated insurance systems to frame these rules and reduce manual efforts related to bid processing and policy issuance. This is particularly the case for some simpler life or private insurance (self-owned, homeowner). However, some insurance companies rely on agents who work for them. This agreement allows an insurer to operate in a market closer to its customers without having to establish a physical presence. The insurance agreement contains the details of the transaction, including the insurance group`s commitment to acquire the new issue of securities, the agreed price, the initial resale price and the settlement date. Once the insurance agreement is reached, the insurer bears the risk of not being able to sell the underlying securities and the costs of keeping them on their books until they can be sold cheaply in the future. Lately, the discourse on underwriting has been dominated by the emergence of machine learning in this room.
These profound technological innovations are changing the way traditional underwriting scores were created and supplanting human subsystems through automation. Understanding the natural language allows more sources of information to be taken into account for risk assessment than before.  These algorithms typically use modern data sources like SMS/e-mail for banking information, location data to check addresses, and so on.