1/30
Looks like no tags are added yet.
Access control
n, /ˈækses kənˈtrəʊl/, Kiểm soát truy cập (Access control), A security technique that regulates who or what can view or use resources in a computing environment., "Role-based access control prevents unauthorized employees from viewing confidential customer records.", permission management, security authorization, access restriction
Access risks
n, /ˈækses rɪsks/, Rủi ro truy cập, Vulnerabilities and potential threats that arise when unauthorized entities gain entry to sensitive digital systems., "Granting excessive administrative privileges creates serious access risks across cloud databases.", authorization vulnerabilities, unauthorized entry hazards
Accountability
n, /əˌkaʊntəˈbɪləti/, Trách nhiệm giải trình (Accountability), The obligation or willingness of organizations and AI developers to accept responsibility for actions and outputs., "Deploying autonomous decision systems requires clear frameworks for legal accountability.", responsibility, answerability, liability
Agent
n, /ˈeɪdʒənt/, Tác tử phần mềm / Tác tử AI (Agent), An autonomous software entity that perceives its environment, reasons, and executes actions to accomplish specific goals., "The AI agent checked flight prices, reserved the seat, and sent the confirmation to my calendar.", autonomous agent, intelligent bot, software agent
Agent hijacking
n, /ˈeɪdʒənt ˈhaɪdʒækɪŋ/, Chiếm đoạt quyền điều khiển tác tử AI (Agent hijacking), A cyberattack where an adversary manipulates or subverts an AI agent’s instructions, memory, or tool access., "Indirect prompt injections can lead to agent hijacking and unauthorized data transmission.", agent takeover, prompt compromise, adversarial control
Agentic AI
n, /eɪˈdʒentɪk ˌeɪ ˈaɪ/, Trí tuệ nhân tạo có tính chủ động / AI tự chủ (Agentic AI), Artificial intelligence systems that exhibit agency by proactively pursuing multi-step objectives with tool use., "Agentic AI represents a paradigm shift from passive text generators to goal-directed problem solvers.", proactive AI, autonomous agent technology
Artificial intelligence
n, /ˌɑːtɪˈfɪʃl ɪnˈtelɪdʒəns/, Trí tuệ nhân tạo (AI), The simulation of human intelligence processes by computer systems, including learning, reasoning, and self-correction., "Artificial intelligence is accelerating discoveries in genomics, renewable energy, and robotics.", AI, machine intelligence, cognitive computing
At scale
phrase, /æt skeɪl/, Ở quy mô lớn / Trên quy mô rộng, Operating or deploying a system in a large, comprehensive, and cost-effective manner to handle massive demand., "Cloud computing allows startups to deploy enterprise software at scale without owning physical servers.", extensively, in large volumes, massively
Audit trails
n, /ˈɔːdɪt treɪlz/, Dấu vết kiểm toán (Audit trails), Step-by-step sequential records providing documentary evidence of the sequence of activities that have affected a system., "Detailed audit trails record every tool invocation and database query executed by the automated agent.", activity logs, tracking records, event histories
Autonomous
adj, /ɔːˈtɒnəməs/, Tự chủ / Tự động độc lập, Having the freedom and capability to act independently without direct external human intervention., "Autonomous delivery drones navigate suburban corridors using onboard lidar and computer vision.", self-governing, independent, self-operating
Autonomy
n, /ɔːˈtɒnəmi/, Tính tự chủ / Quyền tự quyết, The capacity of a system or agent to make decisions and execute operations on its own initiative., "Engineers carefully calibrate the level of autonomy granted to automated trading algorithms.", independence, self-determination, operational freedom
Bias
n, /ˈbaɪəs/, Định kiến / Sự thiên vị (trong thuật toán), Systematic prejudice, favoritism, or unfair skew present in training datasets or algorithmic models., "Algorithmic bias in hiring software can unintentionally exclude qualified minority candidates.", prejudice, algorithmic skew, unfairness
Command
n / v, /kəˈmɑːnd/, Lệnh / Chỉ lệnh, A specific instruction or directive given to a computer program or machine to perform an operation., "Entering a CLI command restarts the cloud server container immediately.", instruction, directive, order
Customer support
n, /ˈkʌstəmər səˈpɔːt/, Hỗ trợ khách hàng / Chăm sóc khách hàng, Services and interactive assistance provided to consumers before, during, and after purchasing products., "AI-driven customer support bots handle routine refund requests around the clock.", client assistance, helpdesk service
Digital assistant
n, /ˈdɪdʒɪtl əˈsɪstənt/, Trợ lý kỹ thuật số / Trợ lý ảo, An advanced conversational software agent designed to help users perform personal tasks and queries., "She asked her digital assistant to summarize her morning emails and draft replies.", virtual assistant, AI helper, smart assistant
Final output
n, /ˈfaɪnl ˈaʊtpʊt/, Kết quả đầu ra cuối cùng, The ultimate deliverable, calculated result, or finished artifact produced by a process or model., "Always verify the final output before publishing automated financial reports to clients.", end result, finished deliverable, concluding artifact
Goal
n, /ɡəʊl/, Mục tiêu, The desired outcome, target state, or objective that a person or autonomous system strives to achieve., "The optimization goal of the routing agent is to minimize delivery fuel consumption.", objective, target, aim, purpose
Governance
n, /ˈɡʌvənəns/, Quản trị (Governance) / Quản trị công nghệ, The overarching framework of rules, ethical principles, and regulations governing the development and deployment of technology., "Robust AI governance ensures compliance with international privacy and human rights laws.", regulation framework, institutional oversight, policymaking
Human oversight
n, /ˈhjuːmən ˈəʊvəsaɪt/, Sự giám sát của con người (Human oversight), The active supervision and intervention capability maintained by humans over autonomous machines., "Critical medical diagnostic tools must always require human oversight before prescriptions are issued.", human-in-the-loop, manual supervision, operator control
Inappropriate reliance
n, /ˌɪnəˈprəʊpriət rɪˈlaɪəns/, Sự phụ thuộc không thích hợp / Quá tin tưởng mù quáng vào máy móc, Over-trusting or excessively depending on an automated system when it is inadequate or prone to hallucination., "Inappropriate reliance on code generators without thorough testing introduces severe security flaws.", over-reliance, automation bias, blind trust
Limited supervision
n, /ˈlɪmɪtɪd ˌsuːpəˈvɪʒn/, Sự giám sát hạn chế, Operating with minimal human checkpoints, where the system handles intermediate steps autonomously., "Under limited supervision, the agent executed complex data cleaning tasks across three servers.", semi-autonomous operation, minimal monitoring
Modernization
n, /ˌmɒdənaɪˈzeɪʃn/, Hiện đại hóa, The transformation of legacy systems, workflows, and infrastructures using cutting-edge digital technologies., "The bank’s modernization program involved migrating on-premise servers to microservices in the cloud.", digital transformation, upgrading, system overhaul
Public services
n, /ˈpʌblɪk ˈsɜːvɪsɪz/, Dịch vụ công trực tuyến, Government-administered digital portals and amenities provided to all citizens., "Digitalizing public services allows citizens to renew passports and pay taxes in minutes.", civic portals, e-government utilities
Scheduling
n, /ˈʃedjuːlɪŋ/, Lập lịch trình / Điều phối thời gian, The automated planning and allocation of tasks, appointments, and resources along a chronological timeline., "AI-assisted calendar scheduling eliminates the back-and-forth emails needed to book meetings.", timetable coordination, calendar management, time planning
Sensitive data leakage
n, /ˈsensətɪv ˈdeɪtə ˈliːkɪdʒ/, Rò rỉ dữ liệu nhạy cảm, The unauthorized accidental or malicious exposure of confidential information to external environments., "Strict data loss prevention filters prevent sensitive data leakage into public LLM prompts.", data exfiltration, confidential breach, information leak
Software development
n, /ˈsɒftweə dɪˈveləpmənt/, Phát triển phần mềm, The iterative process of conceiving, specifying, designing, programming, documenting, testing, and bug fixing applications., "Modern software development incorporates automated CI/CD pipelines and unit test suites.", application engineering, coding, software creation
Take action
phrase, /teɪk ˈækʃn/, Thực hiện hành động, To execute a concrete step, command, or operational procedure to achieve an outcome., "When the security threshold is breached, the firewall agent will immediately take action.", execute operation, proceed, intervene
Tool misuse
n, /tuːl ˌmɪsˈjuːs/, Lạm dụng công cụ / Sử dụng sai công cụ, The improper, unsafe, or malicious invocation of software tools and APIs by automated agents or bad actors., "Sandboxing prevents tool misuse by limiting the commands an agent can execute on the operating system.", API abuse, improper tool invocation, tool exploitation
Trustworthy
adj, /ˈtrʌstwɜːði/, Đáng tin cậy (Trustworthy AI), Demonstrating reliability, safety, fairness, transparency, and consistency in operation and output., "Building trustworthy AI requires rigorous red-teaming and verifiable safety guardrails.", reliable, dependable, verifiable, safe
Workflow
n, /ˈwɜːkfləʊ/, Quy trình công việc (Workflow), The sequence of industrial, administrative, or computational processes through which a piece of work passes from initiation to completion., "The approval workflow routes customer invoices to the finance manager before disbursement.", business process, operational pipeline, work sequence
Workflow automation
n, /ˈwɜːkfləʊ ˌɔːtəˈmeɪʃn/, Tự động hóa quy trình làm việc (Workflow automation), The technological orchestration of repeatable business tasks without manual human effort., "Workflow automation cut employee onboarding processing time from three days to thirty minutes.", process automation, pipeline orchestration, RPA