- 1、原创力文档(book118)网站文档一经付费(服务费),不意味着购买了该文档的版权,仅供个人/单位学习、研究之用,不得用于商业用途,未经授权,严禁复制、发行、汇编、翻译或者网络传播等,侵权必究。。
- 2、本站所有内容均由合作方或网友上传,本站不对文档的完整性、权威性及其观点立场正确性做任何保证或承诺!文档内容仅供研究参考,付费前请自行鉴别。如您付费,意味着您自己接受本站规则且自行承担风险,本站不退款、不进行额外附加服务;查看《如何避免下载的几个坑》。如果您已付费下载过本站文档,您可以点击 这里二次下载。
- 3、如文档侵犯商业秘密、侵犯著作权、侵犯人身权等,请点击“版权申诉”(推荐),也可以打举报电话:400-050-0827(电话支持时间:9:00-18:30)。
LIFE
FLIAISafetyIndex2024
IndependentexpertsevaluatesafetypracticesofleadingAIcompaniesacrosscriticaldomains.
11thDecember2024
Availableonlineat:/index
Contactus:policy@
Firm
Overall
Grade
Score
Risk
Assessment
Current
Harms
Safety
Frameworks
Existential
SafetyStrategy
GovernanceAccountability
TransparencyCommunication
Anthropic
C
2.13
C+
B-
D+
D+
C+
D+
DeepMind
D+
1.55
C
C+
D-
D
D+
D
OpenAI
D+
1.32
C
D+
D-
D-
D+
D-
ZhipuAI
D
1.11
D+
D+
F
F
D
C
x.AI
D-
0.75
F
D
F
F
F
C
Meta
F
0.65
D+
D
F
F
D-
F
Grading:UsestheUSGPAsystemforgradeboundaries:A+,A,A-,B+,[...],Flettervaluescorrespondingtonumericalvalues4.3,4.0,3.7,3.3,[...],0.
KeyFindings
•Largeriskmanagementdisparities:Whilesomecompanieshaveestablishedinitialsafetyframeworksorconductedsomeseriousriskassessmentefforts,othershaveyettotakeeventhemostbasicprecautions.
•Jailbreaks:Alltheflagshipmodelswerefoundtobevulnerabletoadversarialattacks.
•Control-Problem:Despitetheirexplicitambitionstodevelopartificialgeneralintelligence(AGI),capableofrivalingorexceedinghumanintelligence,thereviewpaneldeemedthecurrentstrategiesofallcompaniesinadequateforensuringthatthesesystemsremainsafeandunderhumancontrol.
•Externaloversight:Reviewersconsistentlyhighlightedhowcompanieswereunabletoresistprofit-drivenincentivestocutcornersonsafetyintheabsenceofindependentoversight.WhileAnthropicscurrentandOpenAI’sinitialgovernancestructureswerehighlightedaspromising,expertscalledforthird-partyvalidationofriskassessmentandsafetyframeworkcomplianceacrossallcompanies.
Introduction
RapidlyimprovingAIcapabilitieshaveincreasedinterestinhowcompaniesreport,assessandattempttomitigateassociatedrisks.TheFutureofLifeInstitute(FLI)thereforefacilitatedtheAISafetyIndex,atooldesignedtoevaluateandcomp
原创力文档


文档评论(0)