wlmbrown commited on
Commit
c089e62
·
1 Parent(s): f4fb0fb

changes to model and data cards

Browse files
Files changed (2) hide show
  1. data_cc.yaml +34 -40
  2. model_cc.yaml +74 -78
data_cc.yaml CHANGED
@@ -54,142 +54,136 @@ intended_purpose:
54
  verbose: 'This dataset is appropriate to use for AI projects intended to be used by a judicial authority or on their behalf to assist a judicial authority in researching and interpreting facts and the law and in applying the law to a concrete set of facts, or to be used in a similar way in alternative dispute resolution'
55
  value: !!bool false
56
 
57
- data_and_data_governance:
58
- data_governance:
59
  article: 'Art. 10(1)-(2)'
60
  verbose: 'The dataset was subject to data governance and management practices appropriate to the intended use case'
61
  value: !!bool false
62
- design_choices:
63
  article: 'Art. 10(2)(a)'
64
  verbose: 'The dataset has been subject to data governance and management practices as regards its relevant design choices'
65
  value: !!bool false
66
- data_origin:
67
  article: 'Art. 10(2)(b)'
68
  verbose: 'The dataset has been subject to data governance and management practices as regards its data collection processes and the origin of data, and in the case of personal data, the original purpose of the data collection'
69
  value: !!bool false
70
- data_preparation:
71
  article: 'Art. 10(2)(c)'
72
  verbose: 'The dataset has been subject to data governance and management practices as regards its data-preparation processing operations, such as annotation, labelling, cleaning, updating, enrichment and aggregation'
73
  value: !!bool false
74
- data_assumptions:
75
  article: 'Art. 10(2)(d)'
76
  verbose: 'The dataset has been subject to data governance and management practices as regards its formulation of assumptions, in particular with respect to the information that the data are supposed to measure and represent'
77
  value: !!bool false
78
- data_quantity:
79
  article: 'Art. 10(2)(e)'
80
  verbose: 'The dataset has been subject to data governance and management practices that include an assessment of the availability, quantity and suitability of the data sets that are needed'
81
  value: !!bool false
82
- data_bias_examination:
83
  article: 'Art. 10(2)(f)'
84
  verbose: 'The dataset has been subject to data governance and management practices that include an examination of possible biases that are likely to affect the health and safety of persons, have a negative impact on fundamental rights or lead to discrimination prohibited under Union law, especially where data outputs influence inputs for future operations'
85
  value: !!bool false
86
- data_bias_mitigation:
87
  article: 'Art. 10(2)(g)'
88
  verbose: 'The dataset has been subject to data governance and management practices that include appropriate measures to detect, prevent and mitigate possible biases that are likely to affect the health and safety of persons, have a negative impact on fundamental rights or lead to discrimination prohibited under Union law, especially where data outputs influence inputs for future operations'
89
  value: !!bool false
90
- data_compliance:
91
  article: 'Art. 10(2)(h)'
92
  verbose: 'The dataset has been subject to data governance and management practices that include identification of relevant data gaps or shortcomings that prevent compliance with this Regulation, and how those gaps and shortcomings can be addressed'
93
  value: !!bool false
94
- data_relevance:
95
  article: 'Art. 10(3); Rec. 67'
96
  verbose: 'Training data is relevant'
97
  value: !!bool false
98
- data_representativity:
99
  article: 'Art. 10(3); Rec. 67'
100
  verbose: 'Training data is sufficiently representative'
101
  value: !!bool false
102
- data_errors:
103
  article: 'Art. 10(3); Rec. 67'
104
  verbose: 'Training data is, to the best extent possible, free of errors'
105
  value: !!bool false
106
- data_completeness:
107
  article: 'Art. 10(3); Rec. 67'
108
  verbose: 'Training data is complete in view of the intended purpose of system'
109
  value: !!bool false
110
- statistical_properties:
111
  article: 'Art. 10(3)'
112
  verbose: 'Training data possesses the appropriate statistical properties, including, where applicable, as regards the people in relation to whom the system is intended to be used'
113
  value: !!bool false
114
- contextual:
115
  article: 'Art. 10(4)'
116
  verbose: 'Training data takes into account, to the extent required by the intended purpose, the characteristics or elements that are particular to the specific geographical, contextual, behavioural or functional setting within which the system is intended to be used'
117
  value: !!bool false
118
- personal_data_necessary:
119
  article: 'Art. 10(5)'
120
  verbose: 'Where special categories of personal data have been used to ensure the detection and correction of possible biases that are likely to affect the health and safety of persons, have a negative impact on fundamental rights or lead to discrimination prohibited under Union law, especially where data outputs influence inputs for future operations, the use of this data was strictly necessary'
121
  value: !!bool false
122
- personal_data_safeguards:
123
  article: 'Art. 10(5)'
124
  verbose: 'Where special categories of personal data have been used to ensure the detection and correction of possible biases that are likely to affect the health and safety of persons, have a negative impact on fundamental rights or lead to discrimination prohibited under Union law, especially where data outputs influence inputs for future operations, the use complied with appropriate safeguards for the fundamental rights and freedoms of natural persons'
125
  value: !!bool false
126
- personal_data_gdpr:
127
  article: 'Art. 10(5)'
128
  verbose: 'Where special categories of personal data have been used to ensure the detection and correction of possible biases that are likely to affect the health and safety of persons, have a negative impact on fundamental rights or lead to discrimination prohibited under Union law, especially where data outputs influence inputs for future operations, the use of this data satisfied the provisions set out in Regulations (EU) 2016/679 and (EU) 2018/1725 and Directive (EU) 2016/680'
129
  value: !!bool false
130
- personal_data_other_options:
131
  article: 'Art. 10(5)(a)'
132
  verbose: 'Where special categories of personal data have been used to ensure the detection and correction of possible biases that are likely to affect the health and safety of persons, have a negative impact on fundamental rights or lead to discrimination prohibited under Union law, especially where data outputs influence inputs for future operations, the bias detection and correction was not effectively fulfilled by processing other data, including synthetic or anonymised data'
133
  value: !!bool false
134
- personal_data_limitations:
135
  article: 'Art. 10(5)(b)'
136
  verbose: 'Where special categories of personal data have been used to ensure the detection and correction of possible biases that are likely to affect the health and safety of persons, have a negative impact on fundamental rights or lead to discrimination prohibited under Union law, especially where data outputs influence inputs for future operations, the special categories of personal data were not subject to technical limitations on the re-use of the personal data, and state-of-the-art security and privacy-preserving measures, including pseudonymisation'
137
  value: !!bool false
138
- personal_data_controls:
139
  article: 'Art. 10(5)(c)'
140
  verbose: 'Where special categories of personal data have been used to ensure the detection and correction of possible biases that are likely to affect the health and safety of persons, have a negative impact on fundamental rights or lead to discrimination prohibited under Union law, especially where data outputs influence inputs for future operations, the special categories of personal data were subject to measures to ensure that the personal data processed are secured, protected, subject to suitable safeguards, including strict controls and documentation of the access, to avoid misuse and ensure that only authorised persons have access to those personal data with appropriate confidentiality obligations'
141
  value: !!bool false
142
- personal_data_access:
143
  article: 'Art. 10(5)(d)'
144
  verbose: 'Where special categories of personal data have been used to ensure the detection and correction of possible biases that are likely to affect the health and safety of persons, have a negative impact on fundamental rights or lead to discrimination prohibited under Union law, especially where data outputs influence inputs for future operations, the special categories of personal data were not to be transmitted, transferred or otherwise accessed by other parties'
145
  value: !!bool false
146
- personal_data_deletion:
147
  article: 'Art. 10(5)(e)'
148
  verbose: 'Where special categories of personal data have been used to ensure the detection and correction of possible biases that are likely to affect the health and safety of persons, have a negative impact on fundamental rights or lead to discrimination prohibited under Union law, especially where data outputs influence inputs for future operations, the special categories of personal data were deleted once the bias was corrected or the personal data reached the end of its retention period (whichever came first)'
149
  value: !!bool false
150
- personal_data_necessary_105f:
151
  article: 'Art. 10(5)(f)'
152
  verbose: 'Where special categories of personal data have been used to ensure the detection and correction of possible biases that are likely to affect the health and safety of persons, have a negative impact on fundamental rights or lead to discrimination prohibited under Union law, especially where data outputs influence inputs for future operations, the records of processing activities pursuant to Regulations (EU) 2016/679 and (EU) 2018/1725 and Directive (EU) 2016/680 include the reasons why the processing of special categories of personal data was strictly necessary to detect and correct biases, and why that objective could not be achieved by processing other data'
153
  value: !!bool false
154
-
155
- technical_documentation:
156
- general_description:
157
  article: 'Art. 11; Annex IV(2)(d)'
158
  verbose: 'Dataset carries technical documention, such as a dataseet, including a general description of the dataset.'
159
  value: !!bool false
160
- provenance:
161
  article: 'Art. 11; Annex IV(2)(d)'
162
  verbose: 'Dataset carries technical documention, such as a dataseet, including information about its provenance'
163
  value: !!bool false
164
- scope:
165
  article: 'Art. 11; Annex IV(2)(d)'
166
  verbose: 'Dataset carries technical documention, such as a dataseet, including information about scope and main characteristics'
167
  value: !!bool false
168
- origins:
169
  article: 'Art. 11; Annex IV(2)(d)'
170
  verbose: 'Dataset carries technical documention, such as a dataseet, including information about how the data was obtained and selected'
171
  value: !!bool false
172
- labelling:
173
  article: 'Art. 11; Annex IV(2)(d)'
174
  verbose: 'Dataset carries technical documention, such as a dataseet, including information about labelling procedures (e.g. for supervised learning)'
175
  value: !!bool false
176
- cleaning:
177
  article: 'Art. 11; Annex IV(2)(d)'
178
  verbose: 'Dataset carries technical documention, such as a dataseet, including information about data cleaning methodologies (e.g. outliers detection)'
179
  value: !!bool false
180
- cybersecurity:
181
  article: 'Art. 11; Annex IV(2)(h)'
182
  verbose: 'Cybersecurity measures put in place as regards the data (e.g., scanning for data poisoning)'
183
  value: !!bool false
184
-
185
- transparency_and_provision_of_information_to_deployers:
186
- transparency_and_provision_of_information_to_deployers:
187
  article: '# Art. 13(3)(b)(vi)'
188
  verbose: 'Specifications for the input data, or any other relevant information in terms of the training, validation and testing data sets used, taking into account the intended purpose of the AI system'
189
  value: !!bool false
190
-
191
- quality_management_system:
192
- quality_management_system:
193
  article: 'Art. 17(1)(f)'
194
  verbose: 'Systems and procedures for data management, including data acquisition, data collection, data analysis, data labelling, data storage, data filtration, data mining, data aggregation, data retention and any other operation regarding the data that is performed before and for the purposes of the placing on the market or putting into service of high-risk AI systems'
195
  value: !!bool false
 
54
  verbose: 'This dataset is appropriate to use for AI projects intended to be used by a judicial authority or on their behalf to assist a judicial authority in researching and interpreting facts and the law and in applying the law to a concrete set of facts, or to be used in a similar way in alternative dispute resolution'
55
  value: !!bool false
56
 
57
+ high_risk_ai_systems:
58
+ data_and_data_governance_data_governance:
59
  article: 'Art. 10(1)-(2)'
60
  verbose: 'The dataset was subject to data governance and management practices appropriate to the intended use case'
61
  value: !!bool false
62
+ data_and_data_governance_design_choices:
63
  article: 'Art. 10(2)(a)'
64
  verbose: 'The dataset has been subject to data governance and management practices as regards its relevant design choices'
65
  value: !!bool false
66
+ data_and_data_governance_data_origin:
67
  article: 'Art. 10(2)(b)'
68
  verbose: 'The dataset has been subject to data governance and management practices as regards its data collection processes and the origin of data, and in the case of personal data, the original purpose of the data collection'
69
  value: !!bool false
70
+ data_and_data_governance_data_preparation:
71
  article: 'Art. 10(2)(c)'
72
  verbose: 'The dataset has been subject to data governance and management practices as regards its data-preparation processing operations, such as annotation, labelling, cleaning, updating, enrichment and aggregation'
73
  value: !!bool false
74
+ data_and_data_governance_data_assumptions:
75
  article: 'Art. 10(2)(d)'
76
  verbose: 'The dataset has been subject to data governance and management practices as regards its formulation of assumptions, in particular with respect to the information that the data are supposed to measure and represent'
77
  value: !!bool false
78
+ data_and_data_governance_data_quantity:
79
  article: 'Art. 10(2)(e)'
80
  verbose: 'The dataset has been subject to data governance and management practices that include an assessment of the availability, quantity and suitability of the data sets that are needed'
81
  value: !!bool false
82
+ ddata_and_data_governance_ata_bias_examination:
83
  article: 'Art. 10(2)(f)'
84
  verbose: 'The dataset has been subject to data governance and management practices that include an examination of possible biases that are likely to affect the health and safety of persons, have a negative impact on fundamental rights or lead to discrimination prohibited under Union law, especially where data outputs influence inputs for future operations'
85
  value: !!bool false
86
+ data_and_data_governance_data_and_data_governance_data_bias_mitigation:
87
  article: 'Art. 10(2)(g)'
88
  verbose: 'The dataset has been subject to data governance and management practices that include appropriate measures to detect, prevent and mitigate possible biases that are likely to affect the health and safety of persons, have a negative impact on fundamental rights or lead to discrimination prohibited under Union law, especially where data outputs influence inputs for future operations'
89
  value: !!bool false
90
+ data_and_data_governance_data_compliance:
91
  article: 'Art. 10(2)(h)'
92
  verbose: 'The dataset has been subject to data governance and management practices that include identification of relevant data gaps or shortcomings that prevent compliance with this Regulation, and how those gaps and shortcomings can be addressed'
93
  value: !!bool false
94
+ data_and_data_governance_data_relevance:
95
  article: 'Art. 10(3); Rec. 67'
96
  verbose: 'Training data is relevant'
97
  value: !!bool false
98
+ data_and_data_governance_data_representativity:
99
  article: 'Art. 10(3); Rec. 67'
100
  verbose: 'Training data is sufficiently representative'
101
  value: !!bool false
102
+ data_and_data_governance_data_errors:
103
  article: 'Art. 10(3); Rec. 67'
104
  verbose: 'Training data is, to the best extent possible, free of errors'
105
  value: !!bool false
106
+ data_and_data_governance_data_completeness:
107
  article: 'Art. 10(3); Rec. 67'
108
  verbose: 'Training data is complete in view of the intended purpose of system'
109
  value: !!bool false
110
+ data_and_data_governance_statistical_properties:
111
  article: 'Art. 10(3)'
112
  verbose: 'Training data possesses the appropriate statistical properties, including, where applicable, as regards the people in relation to whom the system is intended to be used'
113
  value: !!bool false
114
+ data_and_data_governance_contextual:
115
  article: 'Art. 10(4)'
116
  verbose: 'Training data takes into account, to the extent required by the intended purpose, the characteristics or elements that are particular to the specific geographical, contextual, behavioural or functional setting within which the system is intended to be used'
117
  value: !!bool false
118
+ data_and_data_governance_personal_data_necessary:
119
  article: 'Art. 10(5)'
120
  verbose: 'Where special categories of personal data have been used to ensure the detection and correction of possible biases that are likely to affect the health and safety of persons, have a negative impact on fundamental rights or lead to discrimination prohibited under Union law, especially where data outputs influence inputs for future operations, the use of this data was strictly necessary'
121
  value: !!bool false
122
+ data_and_data_governance_personal_data_safeguards:
123
  article: 'Art. 10(5)'
124
  verbose: 'Where special categories of personal data have been used to ensure the detection and correction of possible biases that are likely to affect the health and safety of persons, have a negative impact on fundamental rights or lead to discrimination prohibited under Union law, especially where data outputs influence inputs for future operations, the use complied with appropriate safeguards for the fundamental rights and freedoms of natural persons'
125
  value: !!bool false
126
+ data_and_data_governance_personal_data_gdpr:
127
  article: 'Art. 10(5)'
128
  verbose: 'Where special categories of personal data have been used to ensure the detection and correction of possible biases that are likely to affect the health and safety of persons, have a negative impact on fundamental rights or lead to discrimination prohibited under Union law, especially where data outputs influence inputs for future operations, the use of this data satisfied the provisions set out in Regulations (EU) 2016/679 and (EU) 2018/1725 and Directive (EU) 2016/680'
129
  value: !!bool false
130
+ data_and_data_governance_personal_data_other_options:
131
  article: 'Art. 10(5)(a)'
132
  verbose: 'Where special categories of personal data have been used to ensure the detection and correction of possible biases that are likely to affect the health and safety of persons, have a negative impact on fundamental rights or lead to discrimination prohibited under Union law, especially where data outputs influence inputs for future operations, the bias detection and correction was not effectively fulfilled by processing other data, including synthetic or anonymised data'
133
  value: !!bool false
134
+ data_and_data_governance_personal_data_limitations:
135
  article: 'Art. 10(5)(b)'
136
  verbose: 'Where special categories of personal data have been used to ensure the detection and correction of possible biases that are likely to affect the health and safety of persons, have a negative impact on fundamental rights or lead to discrimination prohibited under Union law, especially where data outputs influence inputs for future operations, the special categories of personal data were not subject to technical limitations on the re-use of the personal data, and state-of-the-art security and privacy-preserving measures, including pseudonymisation'
137
  value: !!bool false
138
+ data_and_data_governance_personal_data_controls:
139
  article: 'Art. 10(5)(c)'
140
  verbose: 'Where special categories of personal data have been used to ensure the detection and correction of possible biases that are likely to affect the health and safety of persons, have a negative impact on fundamental rights or lead to discrimination prohibited under Union law, especially where data outputs influence inputs for future operations, the special categories of personal data were subject to measures to ensure that the personal data processed are secured, protected, subject to suitable safeguards, including strict controls and documentation of the access, to avoid misuse and ensure that only authorised persons have access to those personal data with appropriate confidentiality obligations'
141
  value: !!bool false
142
+ data_and_data_governance_personal_data_access:
143
  article: 'Art. 10(5)(d)'
144
  verbose: 'Where special categories of personal data have been used to ensure the detection and correction of possible biases that are likely to affect the health and safety of persons, have a negative impact on fundamental rights or lead to discrimination prohibited under Union law, especially where data outputs influence inputs for future operations, the special categories of personal data were not to be transmitted, transferred or otherwise accessed by other parties'
145
  value: !!bool false
146
+ data_and_data_governance_personal_data_deletion:
147
  article: 'Art. 10(5)(e)'
148
  verbose: 'Where special categories of personal data have been used to ensure the detection and correction of possible biases that are likely to affect the health and safety of persons, have a negative impact on fundamental rights or lead to discrimination prohibited under Union law, especially where data outputs influence inputs for future operations, the special categories of personal data were deleted once the bias was corrected or the personal data reached the end of its retention period (whichever came first)'
149
  value: !!bool false
150
+ data_and_data_governance_personal_data_necessary_105f:
151
  article: 'Art. 10(5)(f)'
152
  verbose: 'Where special categories of personal data have been used to ensure the detection and correction of possible biases that are likely to affect the health and safety of persons, have a negative impact on fundamental rights or lead to discrimination prohibited under Union law, especially where data outputs influence inputs for future operations, the records of processing activities pursuant to Regulations (EU) 2016/679 and (EU) 2018/1725 and Directive (EU) 2016/680 include the reasons why the processing of special categories of personal data was strictly necessary to detect and correct biases, and why that objective could not be achieved by processing other data'
153
  value: !!bool false
154
+ technical_documentation_general_description:
 
 
155
  article: 'Art. 11; Annex IV(2)(d)'
156
  verbose: 'Dataset carries technical documention, such as a dataseet, including a general description of the dataset.'
157
  value: !!bool false
158
+ technical_documentation_provenance:
159
  article: 'Art. 11; Annex IV(2)(d)'
160
  verbose: 'Dataset carries technical documention, such as a dataseet, including information about its provenance'
161
  value: !!bool false
162
+ technical_documentation_scope:
163
  article: 'Art. 11; Annex IV(2)(d)'
164
  verbose: 'Dataset carries technical documention, such as a dataseet, including information about scope and main characteristics'
165
  value: !!bool false
166
+ technical_documentation_origins:
167
  article: 'Art. 11; Annex IV(2)(d)'
168
  verbose: 'Dataset carries technical documention, such as a dataseet, including information about how the data was obtained and selected'
169
  value: !!bool false
170
+ technical_documentation_labelling:
171
  article: 'Art. 11; Annex IV(2)(d)'
172
  verbose: 'Dataset carries technical documention, such as a dataseet, including information about labelling procedures (e.g. for supervised learning)'
173
  value: !!bool false
174
+ technical_documentation_cleaning:
175
  article: 'Art. 11; Annex IV(2)(d)'
176
  verbose: 'Dataset carries technical documention, such as a dataseet, including information about data cleaning methodologies (e.g. outliers detection)'
177
  value: !!bool false
178
+ technical_documentation_cybersecurity:
179
  article: 'Art. 11; Annex IV(2)(h)'
180
  verbose: 'Cybersecurity measures put in place as regards the data (e.g., scanning for data poisoning)'
181
  value: !!bool false
182
+ transparency_and_provision_of_information_to_deployers:
 
 
183
  article: '# Art. 13(3)(b)(vi)'
184
  verbose: 'Specifications for the input data, or any other relevant information in terms of the training, validation and testing data sets used, taking into account the intended purpose of the AI system'
185
  value: !!bool false
186
+ quality_management_system:
 
 
187
  article: 'Art. 17(1)(f)'
188
  verbose: 'Systems and procedures for data management, including data acquisition, data collection, data analysis, data labelling, data storage, data filtration, data mining, data aggregation, data retention and any other operation regarding the data that is performed before and for the purposes of the placing on the market or putting into service of high-risk AI systems'
189
  value: !!bool false
model_cc.yaml CHANGED
@@ -54,162 +54,144 @@ intended_purpose:
54
  verbose: 'This model is appropriate to use for AI projects intended to be used by a judicial authority or on their behalf to assist a judicial authority in researching and interpreting facts and the law and in applying the law to a concrete set of facts, or to be used in a similar way in alternative dispute resolution'
55
  value: !!bool false
56
 
57
- risk_management_system:
58
- foreseeable_risks:
 
 
 
 
59
  article: 'Art. 9(2)(a)'
60
- verbose: 'Known or reasonably foreseeable risks the model can pose to health or safety when used for intended purpose'
61
  value: !!bool false
62
- evaluation:
63
  article: 'Art. 9(2)(b)'
64
- verbose: 'Estimation and evaluation of risks when model used for intended purpose'
65
  value: !!bool false
66
- misuse:
67
  article: 'Art. 9(2)(b)'
68
- verbose: 'Estimation and evaluation of risks when model used under conditions of reasonably foreseeable misuse'
69
  value: !!bool false
70
- testing_performance:
71
  article: 'Art. 9(6)'
72
- verbose: 'Testing to ensure model performs consistently for intended purpose'
73
  value: !!bool false
74
- testing_compliance:
75
  article: 'Art. 9(6)'
76
- verbose: 'Testing to ensure model complies with Act'
77
  value: !!bool false
78
- testing_benchmark:
79
  article: 'Art. 9(8)'
80
- verbose: 'Testing against prior defined metrics appropriate to intended purpose'
81
  value: !!bool false
82
- testing_probabilistic:
83
  article: 'Art. 9(8)'
84
- verbose: 'Testing against probabilistic thresholds appropriate to intended purpose'
85
  value: !!bool false
86
-
87
- technical_documentation:
88
- pre_trained_elements:
89
  article: 'Art. 11; Annex IV(2)(a)'
90
  verbose: 'Model has technical documentation that describes pre-trained elements of model provided by third parties and how used, integrated or modified'
91
  value: !!bool false
92
- logic:
93
  article: 'Art. 11; Annex IV(2)(b)'
94
  verbose: 'Model has technical documentation that describes general logic of model'
95
  value: !!bool false
96
- design_choices:
97
  article: 'Art. 11; Annex IV(2)(b)'
98
  verbose: 'Model has technical documentation that describes key design choices including rationale and assumptions made, including with regard to persons or groups on which model intended to be used'
99
  value: !!bool false
100
- classification_choices:
101
  article: 'Art. 11; Annex IV(2)(b)'
102
  verbose: 'Model has technical documentation that describes main classification choices'
103
  value: !!bool false
104
- parameters:
105
  article: 'Art. 11; Annex IV(2)(b)'
106
- verbose: 'What model is designed to optimise for and relevance of its different parameters'
107
  value: !!bool false
108
- expected_output:
109
  article: 'Art. 11; Annex IV(2)(b)'
110
- verbose: 'Description of the expected output and output quality of the system'
111
  value: !!bool false
112
- act_compliance:
113
  article: 'Art. 11; Annex IV(2)'
114
- verbose: 'Decisions about any possible trade-off made regarding the technical solutions adopted to comply with the requirements set out in Title III, Chapter 2'
115
  value: !!bool false
116
- human_oversight:
117
  article: 'Art. 11; Annex IV(2)(e)'
118
- verbose: 'Assessment of the human oversight measures needed in accordance with Article 14, including an assessment of the technical measures needed to facilitate the interpretation of the outputs of AI systems by the deployers, in accordance with Articles 13(3)(d)'
119
  value: !!bool false
120
- validation:
121
  article: 'Art. 11; Annex IV(2)(g)'
122
- verbose: 'Validation and testing procedures used, including information about the validation and testing data used and their main characteristics; metrics used to measure accuracy, robustness and compliance with other relevant requirements set out in Title III, Chapter 2 as well as potentially discriminatory impacts; test logs and all test reports dated and signed by the responsible persons, including with regard to predetermined changes as referred to under point (f)'
123
  value: !!bool false
124
- cybersecurity:
125
  article: 'Art. 11; Annex IV(2)(h)'
126
- verbose: 'Cybersecurity measures put in place'
127
  value: !!bool false
128
-
129
- transparency_and_provision_of_information_to_deployers:
130
- intended_purpose:
131
  article: 'Art. 13(3)(b)(i)'
132
- verbose: 'Intended purpose'
133
  value: !!bool false
134
- metrics:
135
  article: 'Art. 13(3)(b)(ii)'
136
- verbose: 'Level of accuracy, including its metrics, robustness and cybersecurity referred to in Article 15 against which the high-risk AI system has been tested and validated and which can be expected, and any known and foreseeable circumstances that may have an impact on that expected level of accuracy, robustness and cybersecurity'
137
  value: !!bool false
138
- foreseeable_misuse:
139
  article: 'Art. 13(3)(b)(iii)'
140
- verbose: 'Any known or foreseeable circumstance, related to the use of the high-risk AI system in accordance with its intended purpose or under conditions of reasonably foreseeable misuse, which may lead to risks to the health and safety or fundamental rights referred to in Article 9(2)'
141
  value: !!bool false
142
- explainability:
143
  article: 'Art. 13(3)(b)(iv)'
144
- verbose: 'Technical capabilities and characteristics of the AI system to provide information that is relevant to explain its output'
145
  value: !!bool false
146
- specific_groups:
147
  article: 'Art. 13(3)(b)(v)'
148
- verbose: 'Performance regarding specific persons or groups of persons on which the system is intended to be used'
149
  value: !!bool false
150
- data:
151
  article: 'Art. 13(3)(b)(vi)'
152
- verbose: 'Specifications for the input data, or any other relevant information in terms of the training, validation and testing data sets used, taking into account the intended purpose of the AI system'
153
  value: !!bool false
154
- interpretability:
155
  article: 'Art. 13(3)(b)(vii)'
156
- verbose: 'Information to enable deployers to interpret the output of the high-risk AI system and use it appropriately'
157
  value: !!bool false
158
- human_oversight:
159
  article: 'Art. 13(3)(d)'
160
- verbose: 'Human oversight measures referred to in Article 14, including the technical measures put in place to facilitate the interpretation of the outputs of AI systems by the deployers'
161
  value: !!bool false
162
- hardware:
163
  article: 'Art. 13(3)(e)'
164
- verbose: 'Computational and hardware resources needed, the expected lifetime of the high-risk AI system and any necessary maintenance and care measures, including their frequency, to ensure the proper functioning of that AI system, including as regards software updates'
165
  value: !!bool false
166
-
167
- accuracy_robustness_cybersecurity:
168
- accuracy:
169
  article: 'Art. 15(1)'
170
  verbose: 'Model is designed and developed to achieve appropriate level of accuracy'
171
  value: !!bool false
172
- robustiness:
173
  article: 'Art. 15(1)'
174
  verbose: 'Model is designed and developed to achieve appropriate level of robustness'
175
  value: !!bool false
176
- cybersecurity:
177
  article: 'Art. 15(1)'
178
  verbose: 'Model is designed and developed to achieve appropriate level of cybersecurity'
179
  value: !!bool false
180
- accuracy_metrics:
181
  article: 'Art. 15(2)'
182
  verbose: 'Use of relevant accuracy metrics'
183
  value: !!bool false
184
- fault_resilience:
185
  article: 'Art. 15(4)'
186
  verbose: 'Maximum possible resilience regarding errors, faults or inconsistencies that may occur within the system or the environment in which the system operates, in particular due to their interaction with natural persons or other systems. Technical and organisational measures shall be taken towards this regard'
187
  value: !!bool false
188
- attacks:
189
  article: 'Art. 15(5)'
190
  verbose: 'Measures were taken to prevent, detect, respond to, resolve and control for model poisoning attacks, adversarial examples or model evasion attacks (attacks using inputs designed to cause the model to make a mistake), and confidentiality attacks or model flaws'
191
  value: !!bool false
192
-
193
- quality_management_system:
194
- quality_management_system:
195
  article: 'Art. 17(1)(d)'
196
  verbose: 'Examination, test and validation procedures to be carried out before, during and after the development of the high-risk AI system, and the frequency with which they have to be carried out'
197
  value: !!bool false
198
 
199
- transparency_obligations:
200
- generates_media:
201
- article: 'Art. 50(2)'
202
- verbose: 'AI project generates synthetic audio, image, video or text content'
203
- value: !!bool false
204
- marked_as_generated:
205
- article: 'Art. 50(2)'
206
- verbose: 'outputs are marked in a machine-readable format and detectable as artificially generated or manipulated'
207
- value: !!bool false
208
- interoperability:
209
- article: 'Art. 50(2)'
210
- verbose: 'Providers shall ensure their technical solutions are effective, interoperable, robust and reliable as far as this is technically feasible, taking into account specificities and limitations of different types of content, costs of implementation and the generally acknowledged state-of-the-art, as may be reflected in relevant technical standards'
211
- value: !!bool false
212
-
213
  classification_of_gpai_models:
214
  high_impact_capabilities:
215
  article: 'Art. 51(1)(a)'
@@ -318,4 +300,18 @@ obligations_for_providers_of_gpai_models_with_systemic_risk:
318
  cybersecurity:
319
  article: 'Art. 55(1)(d)'
320
  verbose: 'Ensure an adequate level of cybersecurity protection for the GPAI model with systemic risk and the physical infrastructure of the mode'
 
 
 
 
 
 
 
 
 
 
 
 
 
 
321
  value: !!bool false
 
54
  verbose: 'This model is appropriate to use for AI projects intended to be used by a judicial authority or on their behalf to assist a judicial authority in researching and interpreting facts and the law and in applying the law to a concrete set of facts, or to be used in a similar way in alternative dispute resolution'
55
  value: !!bool false
56
 
57
+ high_risk_ai_models:
58
+ risk_management_system_general:
59
+ article: 'Art. 9(2)'
60
+ verbose: 'A risk management system has been planned, run, reviewed, and updated throughout the model lifecycle'
61
+ value: !!bool false
62
+ risk_management_system_foreseeable_risks:
63
  article: 'Art. 9(2)(a)'
64
+ verbose: 'The risk management system that was established, implemented, documented and maintained througout the model lifecycle included the identification and analysis of any known or reasonably foreseeable risks the model can pose to health or safety when used for intended purpose'
65
  value: !!bool false
66
+ risk_management_system_evaluation:
67
  article: 'Art. 9(2)(b)'
68
+ verbose: 'The risk management system that was established, implemented, documented and maintained througout the model lifecycle included the estimation and evaluation of risks when model used for intended purpose'
69
  value: !!bool false
70
+ risk_management_system_misuse:
71
  article: 'Art. 9(2)(b)'
72
+ verbose: 'The risk management system that was established, implemented, documented and maintained througout the model lifecycle included the estimation and evaluation of risks when model used under conditions of reasonably foreseeable misuse'
73
  value: !!bool false
74
+ risk_management_system_testing_performance:
75
  article: 'Art. 9(6)'
76
+ verbose: 'The risk management system that was established, implemented, documented and maintained througout the model lifecycle included testing to ensure model performs consistently for intended purpose'
77
  value: !!bool false
78
+ risk_management_system_testing_compliance:
79
  article: 'Art. 9(6)'
80
+ verbose: 'The risk management system that was established, implemented, documented and maintained througout the model lifecycle included testing to ensure model complies with Act'
81
  value: !!bool false
82
+ risk_management_system_testing_benchmark:
83
  article: 'Art. 9(8)'
84
+ verbose: 'The risk management system that was established, implemented, documented and maintained througout the model lifecycle included testing against prior defined metrics appropriate to intended purpose'
85
  value: !!bool false
86
+ risk_management_system_testing_probabilistic:
87
  article: 'Art. 9(8)'
88
+ verbose: 'The risk management system that was established, implemented, documented and maintained througout the model lifecycle included testing against probabilistic thresholds appropriate to intended purpose'
89
  value: !!bool false
90
+ technical_documentation_pre_trained_elements:
 
 
91
  article: 'Art. 11; Annex IV(2)(a)'
92
  verbose: 'Model has technical documentation that describes pre-trained elements of model provided by third parties and how used, integrated or modified'
93
  value: !!bool false
94
+ technical_documentation_logic:
95
  article: 'Art. 11; Annex IV(2)(b)'
96
  verbose: 'Model has technical documentation that describes general logic of model'
97
  value: !!bool false
98
+ technical_documentation_design_choices:
99
  article: 'Art. 11; Annex IV(2)(b)'
100
  verbose: 'Model has technical documentation that describes key design choices including rationale and assumptions made, including with regard to persons or groups on which model intended to be used'
101
  value: !!bool false
102
+ technical_documentation_classification_choices:
103
  article: 'Art. 11; Annex IV(2)(b)'
104
  verbose: 'Model has technical documentation that describes main classification choices'
105
  value: !!bool false
106
+ technical_documentation_parameters:
107
  article: 'Art. 11; Annex IV(2)(b)'
108
+ verbose: 'Model has technical documentation that describes what model is designed to optimise for and relevance of its different parameters'
109
  value: !!bool false
110
+ technical_documentation_expected_output:
111
  article: 'Art. 11; Annex IV(2)(b)'
112
+ verbose: 'Model has technical documentation that the expected output and output quality of the system'
113
  value: !!bool false
114
+ technical_documentation_act_compliance:
115
  article: 'Art. 11; Annex IV(2)'
116
+ verbose: 'Model has technical documentation that describes decisions about any possible trade-off made regarding the technical solutions adopted to comply with the requirements set out in Title III, Chapter 2'
117
  value: !!bool false
118
+ technical_documentation_human_oversight:
119
  article: 'Art. 11; Annex IV(2)(e)'
120
+ verbose: 'Model has technical documentation that describes an assessment of the human oversight measures needed in accordance with Article 14, including an assessment of the technical measures needed to facilitate the interpretation of the outputs of AI systems by the deployers, in accordance with Articles 13(3)(d)'
121
  value: !!bool false
122
+ technical_documentation_validation:
123
  article: 'Art. 11; Annex IV(2)(g)'
124
+ verbose: 'Model has technical documentation that describes validation and testing procedures used, including information about the validation and testing data used and their main characteristics; metrics used to measure accuracy, robustness and compliance with other relevant requirements set out in Title III, Chapter 2 as well as potentially discriminatory impacts; test logs and all test reports dated and signed by the responsible persons, including with regard to predetermined changes as referred to under point (f)'
125
  value: !!bool false
126
+ technical_documentation_cybersecurity:
127
  article: 'Art. 11; Annex IV(2)(h)'
128
+ verbose: 'Model has technical documentation that describes cybersecurity measures put in place'
129
  value: !!bool false
130
+ transparency_to_deployers_intended_purpose:
 
 
131
  article: 'Art. 13(3)(b)(i)'
132
+ verbose: 'Model is accompanied by instructions for use that include the characteristics, capabilities, performance limitations, and intended purpose of the model'
133
  value: !!bool false
134
+ transparency_to_deployers_metrics:
135
  article: 'Art. 13(3)(b)(ii)'
136
+ verbose: 'Model is accompanied by instructions for use that include the level of accuracy, including its metrics, robustness and cybersecurity against which the model has been tested and validated and which can be expected, and any known and foreseeable circumstances that may have an impact on that expected level of accuracy, robustness and cybersecurity'
137
  value: !!bool false
138
+ transparency_to_deployers_foreseeable_misuse:
139
  article: 'Art. 13(3)(b)(iii)'
140
+ verbose: 'Model is accompanied by instructions for use that include any known or foreseeable circumstance, related to the use of the model in accordance with its intended purpose or under conditions of reasonably foreseeable misuse, which may lead to risks to the health and safety or fundamental rights referred to in Article 9(2)'
141
  value: !!bool false
142
+ transparency_to_deployers_explainability:
143
  article: 'Art. 13(3)(b)(iv)'
144
+ verbose: 'Model is accompanied by instructions for use that include technical capabilities and characteristics of the model to provide information that is relevant to explain its output'
145
  value: !!bool false
146
+ transparency_to_deployers_specific_groups:
147
  article: 'Art. 13(3)(b)(v)'
148
+ verbose: 'Model is accompanied by instructions for use that include performance regarding specific persons or groups of persons on which the model is intended to be used'
149
  value: !!bool false
150
+ transparency_to_deployers_data:
151
  article: 'Art. 13(3)(b)(vi)'
152
+ verbose: 'Model is accompanied by instructions for use that include specifications for the input data, or any other relevant information in terms of the training, validation and testing data sets used, taking into account the intended purpose of the model'
153
  value: !!bool false
154
+ transparency_to_deployers_interpretability:
155
  article: 'Art. 13(3)(b)(vii)'
156
+ verbose: 'Model is accompanied by instructions for use that include information to enable deployers to interpret the output of the model and use it appropriately'
157
  value: !!bool false
158
+ transparency_to_deployers_human_oversight:
159
  article: 'Art. 13(3)(d)'
160
+ verbose: 'Model is accompanied by instructions for use that include human oversight measures, including the technical measures put in place to facilitate the interpretation of the outputs of model by the deployers'
161
  value: !!bool false
162
+ transparency_to_deployers_hardware:
163
  article: 'Art. 13(3)(e)'
164
+ verbose: 'Model is accompanied by instructions for use that include computational and hardware resources needed, the expected lifetime of the model and any necessary maintenance and care measures, including their frequency, to ensure the proper functioning of that model, including as regards software updates'
165
  value: !!bool false
166
+ accuracy_robustness_cybersecurity_accuracy: # These need to be cleaned up and to match/compliment project cc
 
 
167
  article: 'Art. 15(1)'
168
  verbose: 'Model is designed and developed to achieve appropriate level of accuracy'
169
  value: !!bool false
170
+ accuracy_robustness_cybersecurity_robustiness:
171
  article: 'Art. 15(1)'
172
  verbose: 'Model is designed and developed to achieve appropriate level of robustness'
173
  value: !!bool false
174
+ accuracy_robustness_cybersecurity_cybersecurity:
175
  article: 'Art. 15(1)'
176
  verbose: 'Model is designed and developed to achieve appropriate level of cybersecurity'
177
  value: !!bool false
178
+ accuracy_robustness_cybersecurity_accuracy_metrics:
179
  article: 'Art. 15(2)'
180
  verbose: 'Use of relevant accuracy metrics'
181
  value: !!bool false
182
+ accuracy_robustness_cybersecurity_fault_resilience:
183
  article: 'Art. 15(4)'
184
  verbose: 'Maximum possible resilience regarding errors, faults or inconsistencies that may occur within the system or the environment in which the system operates, in particular due to their interaction with natural persons or other systems. Technical and organisational measures shall be taken towards this regard'
185
  value: !!bool false
186
+ accuracy_robustness_cybersecurity_attacks:
187
  article: 'Art. 15(5)'
188
  verbose: 'Measures were taken to prevent, detect, respond to, resolve and control for model poisoning attacks, adversarial examples or model evasion attacks (attacks using inputs designed to cause the model to make a mistake), and confidentiality attacks or model flaws'
189
  value: !!bool false
190
+ quality_management_system:
 
 
191
  article: 'Art. 17(1)(d)'
192
  verbose: 'Examination, test and validation procedures to be carried out before, during and after the development of the high-risk AI system, and the frequency with which they have to be carried out'
193
  value: !!bool false
194
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
195
  classification_of_gpai_models:
196
  high_impact_capabilities:
197
  article: 'Art. 51(1)(a)'
 
300
  cybersecurity:
301
  article: 'Art. 55(1)(d)'
302
  verbose: 'Ensure an adequate level of cybersecurity protection for the GPAI model with systemic risk and the physical infrastructure of the mode'
303
+ value: !!bool false
304
+
305
+ transparency_obligations:
306
+ generates_media: # clean this up
307
+ article: 'Art. 50(2)'
308
+ verbose: 'If the AI project generates synthetic audio, image, video or text content'
309
+ value: !!bool false
310
+ marked_as_generated:
311
+ article: 'Art. 50(2)'
312
+ verbose: 'If outputs are marked in a machine-readable format and detectable as artificially generated or manipulated'
313
+ value: !!bool false
314
+ interoperability:
315
+ article: 'Art. 50(2)'
316
+ verbose: 'Providers shall ensure their technical solutions are effective, interoperable, robust and reliable as far as this is technically feasible, taking into account specificities and limitations of different types of content, costs of implementation and the generally acknowledged state-of-the-art, as may be reflected in relevant technical standards'
317
  value: !!bool false