AI快速发展之下，口译从业者的出路在哪里

explainability.AI systems should display levels of explainability that aredomains,we anticipate that developing joint guidance will be a priority forregulators.Rationalefor theprincipleIn certain circumstances,AI can have a significant impact on people’s lives,including insurance offers,credit scores,and recruitment outcomes.AI-enabled decisions with high impact outcomes should not be arbitrary andshould be justifiable.In order to ensure a proportionate and context-specific approach regulatorsshould be able to describe and illustrate what fairness means within theirsectors and domains,and consult with other regulators where multiple remitsare engaged by a specific use case.We expect that regulators’interpretations of fairness will include consideration of compliance withrelevant law and regulation,including:4)Relevant sector-specific fairness requirements,such as the FinancialConduct Authority(FCA)Handbook.PrincipleAccountability and governanceDefinitionandexplanationGovernance measures should be in place to ensure effective oversight of thesupply and use of AI systems,with clear lines of accountability establishedacross the AI life cycle.AI life cycle actors should take steps to consider,incorporate and adhere tothe principles and introduce measures necessary for the effectiveThe Data Protection and Digital Information(No.2)Bill reforms the UK’s data protection regime(Data Protection Act 2018and the UK GDPR).Guidance on vulnerability includes:FCA guidance on vulnerable consumers,FCA,2019; Consumer vulnerabilityprotections,Ofgem,2020; Vulnerable consumers,CMA,2018.A pro-innovation approach to AI regulationimplementation of the principles at all stages of the AI life cycle.Regulators will need to look for ways to ensure that clear expectations forregulatory compliance and good practice are placed on appropriate actors inthe AI supply chain,and may need to encourage the use of governance

A pro-innovation approach to AI regulationCase Study 3.4:Explainable AI in practiceThe level of explainability needed from an AI system is highly specific to its context,including the extent to which an application is safety-critical.The level and type ofexplainability required will likely vary depending on whether the intended audience of theexplanation is a regulator,technical expert,or lay person.For example,a technical expert designing self-driving vehicles would need to understandthe system’s decision-making capabilities to test,assess and refine them.In the samecontext,a lay person may need to understand the decision-making process only in orderto use the vehicle safely.If the vehicle malfunctioned and caused a harmfuloutcome,105aregulator may need information about how the system operates in order to allocateresponsibility – similar to the level of explainability currently needed to hold human driversaccountable.While AI explainability remains a technical challenge and an area of active research,regulators are already conducting work to address it.In 2021,the ICO and the Alan TuringInstitute issued co-developed guidance on explaining decisions made withAI,106givingorganisations practical advice to help explain the processes,services and decisionsdelivered or assisted by AI to the individuals affected by them.The audience for an explanation of AI’s outcomes will often be a regulator,who mayrequire a higher standard of explainability depending on the risks represented by an

interpret and apply to AI within their remits.This collaborative and iterative approach can keeppace with a fast moving technology that requires proportionate action to balance risk andopportunity and to strengthen the UK’s position as a global leader in AI.Our agile approachaligns with Sir Patrick Vallance’s Regulation for Innovationreport,78which highlights that flexibleregulatory approaches can better strike the balance between providing clarity,building trust andenabling experimentation.Our framework will provide more clarity to innovators by encouragingcollaboration between government,regulators,industry and civil society.