Telecommunication fraud identification and prevention and control method based on artificial intelligence assistance

Abstract

The invention discloses a telecommunication fraud identification and prevention and control method based on artificial intelligence assistance, and relates to the technical field of telecommunication fraud prevention and control, and the method comprises the steps: comparing an identification result with official data, and generating a structured verification report; when the information displayed by the structured verification report is inconsistent or the suspicious degree score exceeds a suspicious threshold value, initiating an interactive question to the user and obtaining confirmation feedback, updating the risk level and triggering a blocking decision; payment control is executed, a block chain smart contract is automatically created for a smart contract generation instruction, and a user is guided to pay to a contract escrow account and loan to a payee only when a condition is reached; and for a credit protection account trusteeship instruction, transferring user funds to a financial institution guarantee account and sending a template filling instruction, and performing fund operation after artificial intelligence verification of material authenticity and secondary authorization of the user. According to the method, long-acting prevention and control are realized by constructing fraud recognition-risk blocking-long-acting prevention and control full-link closed-loop precise interception.

Description

Technical Field

[0001] This invention relates to the field of telecommunications fraud prevention and control technology, and in particular to an artificial intelligence-assisted method for the identification and prevention of telecommunications fraud. Background Technology

[0002] Current telecom fraud prevention technologies mainly focus on single-point defense mechanisms: financial institutions widely adopt risk control models based on static rules (such as transaction amount thresholds and interception triggered by abnormal geographical locations). These technologies rely on matching historical fraud patterns and are poorly adaptable to the evolution of new rhetoric. Biometric verification technologies (such as facial recognition payment) improve the accuracy of identity verification, but cannot cover the early communication inducement stage of fraud. Blockchain smart contracts applied in the field of trade settlement (such as IBM's goods delivery payment contract) provide a conditional fund release approach, but have not yet been deeply integrated with fraud identification scenarios.

[0003] The existing technology system has significant flaws: First, its dynamic semantic recognition capability is weak, and traditional rule bases are unable to analyze the semantic feature variations of fraudulent phrases such as impersonating customer service for refunds; second, the fragmented cross-system verification leads to response delays, as the public security identity database, bank account system, and communication monitoring module are in separate architectures, resulting in a disconnect between risk assessment and fund control; third, there is a lack of long-term prevention and control mechanisms, as verified accounts lack continuous trust management and dynamic control over payment limits, and there is a lack of ability to curb excessive payment behavior. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides an artificial intelligence-assisted method for the identification and prevention of telecommunications fraud, which solves the problems of lack of dynamic semantic recognition capabilities, response delays caused by fragmented cross-system verification, and the risk of overpayment caused by the absence of a long-term trust mechanism in existing telecommunications fraud prevention technologies.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, this invention provides an artificial intelligence-assisted method for identifying and preventing telecommunications fraud. The method includes: a user's smart device monitoring communication information in real time; when the AI ​​detects payment instructions, unfamiliar accounts, or preset risk keywords, automatically parsing voice, text, website links, and image content, and outputting identification results; the AI ​​calling public security and financial institution databases, comparing the identification results with official data, and generating a structured verification report; when the structured verification report shows inconsistent information or the suspiciousness score exceeds a suspiciousness threshold, initiating interactive questions to the user and obtaining confirmation feedback on the relationship with the payee and the transaction background, updating the risk level, and triggering a blocking decision; executing payment control based on the updated risk level; for smart contract generation instructions in the blocking payment process, extracting transaction commitment terms to automatically create a blockchain smart contract, guiding the user to pay to a contract escrow account, and releasing funds to the payee only when conditions are met; for credit protection account escrow instructions, transferring user funds to a financial institution guarantee account, sending template filling instructions, and executing fund operations after AI verifies the authenticity of the materials and the user grants secondary authorization; updating the account tag library based on the payee verification results, adding payee tags and dynamic credit limits matching the business type, and executing tag rules in subsequent transactions.

[0007] As a preferred embodiment of the AI-assisted telecommunications fraud identification and prevention method of the present invention, the output of the identification result includes the following steps: The user's smart device continuously monitors the communication channels of calls, text messages and social applications. When it detects that the communication source is an unknown number or an unknown social account, it automatically triggers the activation of artificial intelligence. Artificial intelligence performs classification and analysis operations on communication content; The parsed text content is compared with the fraud feature rule set in the telecommunications fraud case database, and the recognition result is output.

[0008] As a preferred embodiment of the AI-assisted telecommunications fraud identification and prevention method of the present invention, wherein: the generation of the structured verification report includes, Artificial intelligence extracts the account and identity information to be verified from the recognition results; A request for identity information verification is initiated to the population database of the public security system through a pre-configured interface; Verify account information through the financial institution's account verification system; When it involves corporate recipients, supplementary business registration and tax information verification is performed to generate a structured verification report.

[0009] As a preferred embodiment of the AI-assisted telecommunications fraud identification and prevention method of the present invention, the step of updating the risk level and triggering a blocking decision includes the following steps: When the structured verification report shows that the account consistency is not a match, or the suspiciousness score exceeds the suspiciousness threshold, the human-computer interaction process is triggered. Artificial intelligence generates targeted question options and displays them on the user's smart device interface, forcing the user to make a selection and obtain confirmation feedback; The risk level is updated based on user confirmation feedback, and a blocking decision is triggered based on the updated risk level.

[0010] As a preferred embodiment of the AI-assisted telecommunications fraud identification and prevention method of the present invention, the payment control based on the updated risk level refers to terminating the transaction and marking the abnormal account if it is confirmed as fraud; if there is doubt, a blocking payment process is initiated by generating a smart contract, managing a credit protection account, or redirecting to a trusted third-party account.

[0011] As a preferred embodiment of the AI-assisted telecommunications fraud identification and prevention method of the present invention, the step of automatically creating a blockchain smart contract by extracting transaction commitment terms from the smart contract generation instruction in the blocking payment process, guiding the user to make payments to the contract escrow account, and releasing funds to the payee only when conditions are met includes the following steps. When a blocking payment process instruction requires the smart contract to be generated, artificial intelligence scans and parses the text content to identify keywords in the transaction commitment terms; Extract complete commitment clauses and automatically generate a smart contract code framework; Deploy smart contracts on the Ethereum blockchain and generate unique contract escrow account addresses; Guide users to transfer funds to the contract escrow account through electronic payment channels; The smart contract continuously monitors the status of the conditions, and automatically transfers funds to the recipient's address only when the monitored data meets the contract's triggering conditions.

[0012] As a preferred embodiment of the AI-assisted telecommunications fraud identification and prevention method of the present invention, the step of transferring user funds to a financial institution's guarantee account in response to the credit protection account escrow instruction, sending a template filling instruction, and executing the fund operation after AI verifies the authenticity of the materials and the user grants secondary authorization includes the following steps. When a blocked payment process instruction requires the credit protection account to be held in escrow, the user's payment funds will be immediately transferred to the user's financial institution guarantee account. Select a verification template that matches the transaction type from the preset template library; Send template filling instructions to the payee through official channels of financial institutions; Artificial intelligence performs material authenticity verification and generates a verification report that includes the status of each verification result; When the verification report shows that all verifications have passed, the verification results and the recipient's application content are pushed to the user's smart device, requiring the user to complete a second authorization. Fund operations will be carried out based on the authorization results.

[0013] As a preferred embodiment of the AI-assisted telecommunications fraud identification and prevention method of the present invention, the step of updating the account tag library based on the recipient's verification result, adding payment tags and dynamic credit limits matching the business type, and executing the tag rules in subsequent transactions includes the following steps: Once the credit protection account verification process is complete, extract the payee's identity and business type from the structured verification report; Query the historical transaction database of financial institutions to obtain the main business revenue data of the payee; Create new tag records or update tag records in the account tag library; The storage verification validity period is set, and the tag rules are executed when the payee initiates a new payment request.

[0014] In a second aspect, the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, it implements any step of the artificial intelligence-assisted telecommunications fraud identification and prevention method described in the first aspect of the present invention.

[0015] Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the artificial intelligence-assisted telecommunications fraud identification and prevention method described in the first aspect of the present invention.

[0016] The beneficial effects of this invention are as follows: By matching a dynamic feature library in real time, the false alarm rate of fraudulent statements is reduced; and by using optical character recognition (OCR) technology, fraudulent accounts in images are accurately extracted, avoiding delays in manual review and reducing the risk of user error. The pre-set interface returns verification results in seconds, shortening the response time for high-risk transactions. By supplementing verification with enterprise taxpayer identification numbers, the problem of fraudsters using shell companies to collect payments is solved, improving the accuracy of identifying fake enterprises. By automatically extracting transaction commitment terms to generate Solidity smart contracts and deploying them to the blockchain, verbal agreements are transformed into executable code on the chain, resolving disputes over "unfulfilled promises" or "failure to meet payment targets" in contract fraud, and achieving conditional release of funds. The use of templated material verification and biometric secondary authorization balances security and efficiency. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart of an AI-assisted method for identifying and preventing telecommunications fraud.

[0019] Figure 2 Generate flowcharts for smart contracts.

[0020] Figure 3 Flowchart for verifying loan disbursements for credit protection accounts. Detailed Implementation

[0021] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0022] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0023] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0024] Reference Figures 1-3 This is one embodiment of the present invention, which provides an artificial intelligence-assisted method for identifying and preventing telecommunications fraud, comprising the following steps: S1. The user's smart device monitors communication information in real time. When the artificial intelligence detects payment instructions, unfamiliar accounts, or preset risk keywords, it automatically parses voice, text, website links, and image content and outputs the recognition results.

[0025] The user's smart device continuously monitors the communication channels of calls, text messages and social applications. When it detects that the communication source is an unknown number or an unknown social account, it automatically triggers the activation of artificial intelligence. Specifically, the user's smart device monitors call voice streams, SMS text messages, and social application communication messages in real time through the operating system's underlying API. When it detects that the communication source number or social account does not exist in the user's address book and historical contact records, it automatically activates the artificial intelligence process.

[0026] Artificial intelligence performs classification and analysis operations on communication content; Furthermore, the artificial intelligence process calls the speech recognition engine to convert call audio into text, extracts keywords from SMS / social text, initiates a sandbox environment to scan for potential malicious code in website links, and uses optical character recognition technology to parse text within images, generating a structured set of parsed results.

[0027] The parsed text content is compared with the fraud feature rule set in the telecommunications fraud case database, and the recognition result is output.

[0028] Specifically, the artificial intelligence process loads a predefined set of fraud feature rules from the telecom fraud case database (containing 500+ keywords and logical combinations of 12 types of fraud templates such as impersonating public security, procuratorate and court officials, high-return investments, and emergency transfers), performs multi-dimensional similarity matching between the parsed result set and the feature rule set, and outputs recognition results including a suspiciousness score (0-100 points), a list of payment accounts to be verified, fields of identity information to be verified, and high-risk marker types (such as "impersonating customer service for refunds").

[0029] S2. Artificial intelligence calls on databases of public security and financial institutions, compares the identification results with official data, and generates a structured verification report.

[0030] Artificial intelligence extracts the account and identity information to be verified from the recognition results: The accounts to be verified include bank card numbers and payment platform accounts; the identity information to be verified includes personal names, ID card numbers or company names, and unified social credit codes.

[0031] The system initiates an identity information verification request to the public security system's population database through a pre-configured interface, and verifies account information through the financial institution's account verification system. Specifically, the system submits identity information fields to the public security system's population database interface through an encrypted channel to initiate a verification request, and obtains the validity status of the ID card and the consistency result of the name in real time; simultaneously, it submits the account field to be verified to the financial institution's account verification system to obtain the account's real-name authentication name and account status (normal / frozen / cancelled).

[0032] When it involves corporate recipients, supplementary business registration and tax information verification is performed to generate a structured verification report.

[0033] Furthermore, when the identity information field contains the unified social credit code, it is confirmed as a corporate payee. The system then queries the State Administration for Market Regulation's database to check the company's business registration status and verifies the validity of the taxpayer identification number with the tax system. Finally, it integrates the identity consistency status returned by the public security system, the account real-name matching status returned by financial institutions, and the corporate qualification status returned by the industry and commerce and tax authorities to generate a structured verification report containing the account consistency conclusion (yes / no), account risk label (normal / abnormal freezing), corporate qualification status (valid / invalid), and business attribute prediction (financial management / loan, etc.).

[0034] S3. When the structured verification report shows inconsistent information or the suspicion score exceeds the suspicion threshold, initiate interactive questions to the user and obtain confirmation feedback on the relationship with the payee and the transaction background, update the risk level and trigger a blocking decision.

[0035] When the structured verification report shows that the account consistency is mismatched, or the suspiciousness score exceeds the suspiciousness threshold, the human-computer interaction process is triggered; artificial intelligence generates targeted question options and displays them on the user's smart device interface, forcing the user to complete the selection and obtain confirmation feedback; Specifically, when the structured verification report shows that the account consistency conclusion is "mismatch" or the suspiciousness score exceeds the suspiciousness threshold (example value: 80 points), the artificial intelligence extracts the key contradictions in the structured verification report and generates two-option questions; the option window is forcibly popped up through the user's smart device interface, and confirmation feedback is obtained after the user makes a selection.

[0036] For example, a user refers to the caller as "uncle," and the uncle asks the user for money. Throughout the process, the AI ​​cannot identify the uncle's real name. Using the account provided by the uncle, the AI ​​searches online and finds the account holder's real name is Zhang San. At this point, the AI ​​generates a question: "The uncle's name is Zhang San," offering a yes or no option. The user selects a yes or no, confirming the uncle's name is Zhang San. The AI ​​then determines that Zhang San is indeed the user's uncle, the loan is credible, and prompts the user to make payment to Zhang San. The user confirms the AI's identification result, believing the loan is not a scam, and transfers the money to the account provided by the uncle.

[0037] It should be noted that for some ambiguous information, information that is difficult to identify or verify, artificial intelligence generates relevant question options for users to choose from, thereby achieving more accurate identification of the information and more effectively judging whether the content of the information is telecommunications fraud.

[0038] The risk level is updated based on user confirmation feedback, and a blocking decision is triggered based on the updated risk level.

[0039] Furthermore, the risk level is updated based on user confirmation feedback (it is downgraded to medium risk when it matches the public security data, and upgraded to high risk when it contradicts it). Based on the updated risk level, a blocking decision instruction is triggered: when it is high risk, a "terminate payment instruction" is output and the abnormal account is marked and reported to the financial institution; when it is medium risk, a "block payment process instruction" is output and carries the business attribute tag value (such as "financial investment" or "hardware loan").

[0040] S4. Based on the updated risk level, execute payment control, generate instructions for smart contracts in the blocking payment process, extract transaction commitment terms to automatically create blockchain smart contracts, guide users to make payments to the contract escrow account and release funds to the payee only when the conditions are met.

[0041] Among them, payment control based on updated risk levels means that if fraud is confirmed, the transaction will be terminated and the abnormal account will be marked; if there is doubt, a blocking payment process will be initiated, which includes smart contract generation, credit protection account custody, or redirection to a third-party trusted account.

[0042] When a blocking payment process instruction requires the generation of a smart contract, artificial intelligence scans and parses the text content, identifies keywords in the transaction commitment terms, extracts the complete commitment terms, and automatically generates the smart contract code framework. Furthermore, when the blocking payment process instruction carries a business attribute tag and requires the generation of a smart contract, the AI ​​scans and parses the text content to identify keywords in the transaction commitment clauses such as "payment after acceptance" and "payment after obtaining qualifications," and extracts complete conditional statements (such as "Party A will pay after Party B completes the project application service"). Based on the statement logic, the Solidity smart contract code framework is automatically generated: the payer is set to the address of the contract creator, the payee is a valid address verified by a financial institution, and the commitment clauses are transformed into executable trigger conditions (such as calling the national project application system API to verify the qualification acquisition status).

[0043] Deploy smart contracts on the Ethereum blockchain to generate unique contract escrow account addresses; guide users to transfer funds to the contract escrow account through electronic payment channels; the smart contract continuously monitors the status of condition fulfillment, and automatically executes the fund transfer to the recipient's address only when the monitored data meets the contract trigger conditions.

[0044] Specifically, the contract is deployed on the Ethereum testnet and its deployment status is verified through the Etherscan API, generating a contract escrow account address starting with 0x. Users are then guided to transfer funds to this contract escrow account address via bank transfer or cryptocurrency payment channels. The smart contract continuously monitors the conditions through preset IoT sensor data streams (engineering acceptance), government platform API return values ​​(qualification status), or manual confirmation instructions, and automatically executes the fund transfer to the recipient address only when a valid data signature indicating that the conditions have been met is received.

[0045] S5. For instructions regarding the custody of credit protection accounts, the user's funds will be transferred to the financial institution's guarantee account, and a template filling instruction will be sent. After the authenticity of the materials is verified by artificial intelligence and the user grants secondary authorization, the fund operation will be executed.

[0046] When a blocked payment process instruction requires custody of a credit protection account, the user's payment funds will be immediately transferred to the user's financial institution guarantee account. Specifically, when a blocked payment process instruction requires custody of a credit protection account, the user's payment funds will be immediately transferred from the original account to the user's credit protection account opened at the financial institution via the bank transfer API, and a custody certificate will be generated.

[0047] Select a verification template that matches the transaction type from the preset template library and send template filling instructions to the payee through the official channels of the financial institution; Furthermore, based on the business attribute tags (such as "financial investment" or "hardware payment") carried in the blocked payment process instructions, the system matches the corresponding verification template from the preset template library (the personal template requires ID card + facial recognition video, and the corporate template requires business license + corporate account + legal representative's ID card); the system sends a template filling link to the payee via official SMS and APP push from the financial institution, forcing them to upload materials through an encrypted page within 24 hours; Artificial intelligence performs material authenticity verification and generates a verification report that includes the status of each verification result; Specifically, the artificial intelligence calls the public security facial recognition interface to verify the biometric features of ID card photos and videos, verifies the authenticity of business license codes through the API of the State Administration for Market Regulation, and verifies the matching of corporate accounts through the financial institution account system, generating a verification report that includes the status of each verification result (pass / fail).

[0048] When the verification report shows that all verifications have passed, the verification results and the recipient's application content are pushed to the user's smart device, requiring the user to complete a second authorization; the fund operation is executed according to the authorization result.

[0049] Furthermore, when the verification report shows that all verification statuses are passed, the complete verification result data package and the original application of the payee are pushed to the user's smart device, triggering the payment password input or fingerprint / facial biometric collection window; after the user completes the biometric verification or password input, the fund release operation is immediately executed to transfer the funds from the credit protection account to the payee's account; otherwise, the original delayed payment period (such as 30 days) is maintained and the payee is notified to supplement the materials.

[0050] S6. Update the account tag library based on the recipient's verification results, and add payment tags and dynamic credit limits that match the business type. Then, execute the tag rules in subsequent transactions.

[0051] Once the credit protection account verification process is complete, extract the payee's identity identifier (18-digit ID card number for individuals / 20-digit unified social credit code for enterprises) and the verified business type (such as salary income, financial investment, hardware payment, etc.) from the structured verification report.

[0052] The system queries the historical transaction database of financial institutions to obtain the main business revenue data of the payee; it creates new tag records or updates tag records in the account tag library; it stores the validity period of the verification and executes the tag rules when the payee initiates a new collection request.

[0053] Furthermore, query the historical transaction database of financial institutions for the past 12 months of loan transaction records matching the business type of the payee, and calculate the average monthly income value as the base after excluding non-core income such as salary refunds; create or update a record with identity identifier as the primary key in the account tag library, and write the business type tag, base value, single payment limit (base × 3), monthly cumulative payment limit (base × 5), and expiration date (current date + 180 days for personal accounts / current date + 365 days for corporate accounts); When the payee initiates a new payment request through the payment channel, the latest record in the account tag library is retrieved by the identity identifier. The system verifies whether the business type fully matches the tag value, whether the amount of this payment is less than or equal to the single transaction limit, and whether the cumulative payments for this month are less than or equal to the monthly limit. If all conditions are met, the verification process is automatically waived and the funds are released. Otherwise, the transaction is immediately intercepted and an over-limit interception notification is sent to the user (including the specific over-limit type and amount).

[0054] This embodiment also provides a computer device applicable to the situation of a telecommunications fraud identification and prevention method based on artificial intelligence, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to realize the telecommunications fraud identification and prevention method based on artificial intelligence as proposed in the above embodiment.

[0055] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0056] This embodiment also provides a storage medium storing a computer program. When executed by a processor, the program implements the artificial intelligence-assisted telecommunications fraud identification and prevention method proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0057] In summary, this invention reduces the false alarm rate of fraudulent statements by: real-time matching with a dynamic feature library; accurately extracting fraudulent accounts from images using Optical Character Recognition (OCR) technology, avoiding delays in manual review and reducing the risk of user error; shortening the response time for high-risk transactions by returning verification results in seconds through a pre-set interface; improving the accuracy of identifying fake companies by supplementing verification with enterprise taxpayer identification numbers, thus solving the problem of fraudsters using shell companies to collect payments; automatically extracting transaction commitment terms to generate Solidity smart contracts and deploying them to the blockchain, transforming verbal agreements into on-chain executable code, resolving disputes over "unfulfilled promises" or "failure to meet payment targets" in contract fraud, and achieving conditional release of funds; and balancing security and efficiency by using templated material verification and biometric secondary authorization.

[0058] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for identifying and preventing telecommunications fraud based on artificial intelligence, characterized in that: include, The user's smart device monitors communication information in real time. When the artificial intelligence detects payment instructions, unfamiliar accounts or preset risk keywords, it automatically parses voice, text, website links and image content and outputs recognition results. Artificial intelligence accesses databases from public security and financial institutions, compares the identification results with official data, and generates a structured verification report. When the structured verification report shows inconsistent information or the suspicion score exceeds the suspicion threshold, an interactive question is initiated to the user to obtain confirmation feedback on the relationship with the payee and the transaction background, the risk level is updated and a blocking decision is triggered; Payment control is implemented based on the updated risk level. Instructions are generated for smart contracts in the blocking payment process. Transaction commitment terms are extracted to automatically create blockchain smart contracts, guide users to make payments to the contract escrow account, and release funds to the payee only when the conditions are met. In response to the credit protection account custody instruction, user funds are transferred to the financial institution's guarantee account, and a template filling instruction is sent. After the authenticity of the materials is verified by artificial intelligence and the user authorizes the transaction a second time, the fund operation is executed. Update the account tag library based on the recipient's verification results, add payment tags and dynamic credit limits that match the business type, and enforce the tag rules in subsequent transactions.

2. The method for identifying and preventing telecommunications fraud based on artificial intelligence as described in claim 1, characterized in that: The output recognition result includes the following steps: The user's smart device continuously monitors the communication channels of calls, text messages and social applications. When it detects that the communication source is an unknown number or an unknown social account, it automatically triggers the activation of artificial intelligence. Artificial intelligence performs classification and analysis operations on communication content; The parsed text content is compared with the fraud feature rule set in the telecommunications fraud case database, and the recognition result is output.

3. The method for identifying and preventing telecommunications fraud based on artificial intelligence as described in claim 2, characterized in that: The generation of the structured verification report includes, Artificial intelligence extracts the account and identity information to be verified from the recognition results; A request for identity information verification is initiated to the population database of the public security system through a pre-configured interface; Verify account information through the financial institution's account verification system; When it involves corporate recipients, supplementary business registration and tax information verification is performed to generate a structured verification report.

4. The method for identifying and preventing telecommunications fraud based on artificial intelligence as described in claim 3, characterized in that: The process of updating the risk level and triggering a blocking decision includes the following steps: When the structured verification report shows that the account consistency is not a match, or the suspiciousness score exceeds the suspiciousness threshold, the human-computer interaction process is triggered. Artificial intelligence generates targeted question options and displays them on the user's smart device interface, forcing the user to make a selection and obtain confirmation feedback; The risk level is updated based on user confirmation feedback, and a blocking decision is triggered based on the updated risk level.

5. The method for identifying and preventing telecommunications fraud based on artificial intelligence as described in claim 4, characterized in that: The payment control based on the updated risk level refers to terminating the transaction and marking the abnormal account if fraud is confirmed; and initiating a blocking payment process by generating a smart contract, escrowing a credit protection account, or redirecting to a trusted third-party account if there is any doubt.

6. The method for identifying and preventing telecommunications fraud based on artificial intelligence as described in claim 5, characterized in that: The aforementioned instructions for generating smart contracts in a blocked payment process extract transaction commitment terms to automatically create a blockchain smart contract, guiding users to make payments to a contract escrow account and releasing funds to the recipient only when conditions are met. This includes the following steps: When a blocking payment process instruction requires the smart contract to be generated, artificial intelligence scans and parses the text content to identify keywords in the transaction commitment terms; Extract complete commitment clauses and automatically generate a smart contract code framework; Deploy smart contracts on the Ethereum blockchain and generate unique contract escrow account addresses; Guide users to transfer funds to the contract escrow account through electronic payment channels; The smart contract continuously monitors the status of the conditions, and automatically transfers funds to the recipient's address only when the monitored data meets the contract's triggering conditions.

7. The method for identifying and preventing telecommunications fraud based on artificial intelligence as described in claim 6, characterized in that: The aforementioned instruction for the custody of the credit protection account involves transferring user funds to a financial institution's guarantee account and sending a template completion instruction. After AI verifies the authenticity of the materials and the user grants secondary authorization, the fund operation is executed, including the following steps. When a blocked payment process instruction requires the credit protection account to be held in escrow, the user's payment funds will be immediately transferred to the user's financial institution guarantee account. Select a verification template that matches the transaction type from the preset template library; Send template filling instructions to the payee through official channels of financial institutions; Artificial intelligence performs material authenticity verification and generates a verification report that includes the status of each verification result; When the verification report shows that all verifications have passed, the verification results and the recipient's application content are pushed to the user's smart device, requiring the user to complete a second authorization. Fund operations will be carried out based on the authorization results.

8. The method for identifying and preventing telecommunications fraud based on artificial intelligence as described in claim 7, characterized in that: The process of updating the account tag library based on the payee's verification results, adding payment tags and dynamic credit limits that match the business type, and executing the tag rules in subsequent transactions includes the following steps: Once the credit protection account verification process is complete, extract the payee's identity and business type from the structured verification report; Query the historical transaction database of financial institutions to obtain the main business revenue data of the payee; Create new tag records or update tag records in the account tag library; The storage verification validity period is set, and the tag rules are executed when the payee initiates a new payment request.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the artificial intelligence-assisted telecommunications fraud identification and prevention method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the artificial intelligence-assisted telecommunications fraud identification and prevention method as described in any one of claims 1 to 7.

Images