new requirements added to Project CC and Data CC

compliance_analysis.py

@@ -1,5 +1,5 @@
 import yaml
-from utils import set_type, set_operator_role_and_location, set_eu_market_status, check_within_scope, check_prohibited
+from utils import set_type, set_operator_role_and_location, set_eu_market_status, check_within_scope
 
 # Create some variables we will use throughout our analysis
 
@@ -25,6 +25,8 @@ project_variables = {
     }
 }
 
+project_intended_purpose = None 
+
 def run_compliance_analysis_on_project(project_cc_yaml):
 
     # Determine project type (AI system vs. GPAI model) as well as operator type. We will use these for different things.
@@ -33,16 +35,17 @@ def run_compliance_analysis_on_project(project_cc_yaml):
     set_eu_market_status(project_variables, project_cc_yaml)
 
     # Check if the project is within scope of the Act. If it's not, the analysis is over.
-    if check_within_scope(project_variables, project_cc_yaml):
+    if check_within_scope(project_cc_yaml):
         msg = ("Project is within the scope of Act. Let's continue...") 
     else: 
         msg = ("Project is not within the scope of what is regulated by the Act.")
-        return msg
-    
-    # Check for prohibited practices. If any exist, the analysis is over.
-    if check_prohibited(project_variables, project_cc_yaml) == True:
-        msg = ("Project is non-compliant due to a prohibited practice.")
-        return msg
+
+    # # Check for prohibited practices. If any exist, the analysis is over.
+    # if check_prohibited(project_cc_yaml) == True: 
+    #     print("Project contains prohibited practices and is therefore non-compliant.")
+    #     msg = ("Project is non-compliant due to a prohibited practice.")
+    # else: 
+    #     print("Project does not contain prohibited practies. Let's continue...")
 
     # If project is high-risk AI system, check that is has met all the requirements for such systems: 
 
@@ -50,6 +53,8 @@ def run_compliance_analysis_on_project(project_cc_yaml):
 
     # Do this by examining the Project CC
 
+        # WE HAVE TO ADD A CONDITION THAT APPLIES THESE RULES BELOW ONLY IF operator_role == provider
+
         for key, value in project_cc_yaml['risk_management_system']:
             if not value:
                 msg = ("Because of project-level characteristics, this high-risk AI system fails the risk management requirements under Article 9.")
@@ -111,6 +116,107 @@ def run_compliance_analysis_on_model(model_cc_yaml):
    
     return msg
 
+    def_check_intended_use_aligned():  # We want to run this function for everything classified as a high_risk_ai_system
+                                        # We also need to run it for all 
+
+        # Add any of the intended purposes of the overall project to a set of intended purposes 
+
+        if project_cc_yaml['high_risk_ai_system']['safety_component'] == True:
+            intended_purpose = "safety_component"
+        if project_cc_yaml['high_risk_ai_system']['product_regulated_machinery'] == True:
+            intended_purpose = "product_regulated_machinery" 
+        if project_cc_yaml['high_risk_ai_system']['product_regulated_toy'] == True:
+            intended_purpose = "product_regulated_toy" 
+        if project_cc_yaml['high_risk_ai_system']['product_regulated_watercraft'] == True:
+            intended_purpose = "product_regulated_watercraft" 
+        if project_cc_yaml['high_risk_ai_system']['biometric_categorization'] == True:
+            intended_purpose = "biometric_categorization" 
+        if project_cc_yaml['high_risk_ai_system']['emotion_recognition'] == True:
+            intended_purpose = "emotion_recognition" 
+        if project_cc_yaml['high_risk_ai_system']['critical_infrastructure'] == True:
+            intended_purpose = "critical_infrastructure" 
+        if project_cc_yaml['high_risk_ai_system']['admission'] == True:
+            intended_purpose = "admission" 
+        if project_cc_yaml['high_risk_ai_system']['recruitment'] == True:
+            intended_purpose = "recruitment" 
+        if project_cc_yaml['high_risk_ai_system']['public_assistance'] == True:
+            intended_purpose = "public_assistance"
+        if project_cc_yaml['high_risk_ai_system']['victim_assessment'] == True:
+            intended_purpose = "victim_assessment" 
+        if project_cc_yaml['high_risk_ai_system']['polygraph'] == True:
+            intended_purpose = "polygraph" 
+        if project_cc_yaml['high_risk_ai_system']['judicial'] == True:
+            intended_purpose = "judicial" 
+        
+        # For each Data CC, put the intended uses in a set and then make sure the Project's intended use is in the set
+
+        dataset_intended_purpose = () 
+
+        if data_cc_yaml['intended_purpose']['safety_component'] == True:
+            dataset_intended_purpose.add("safety_component")
+        if data_cc_yaml['intended_purpose']['product_regulated_machinery'] == True:
+            dataset_intended_purpose.add("product_regulated_machinery")
+        if data_cc_yaml['intended_purpose']['product_regulated_toy'] == True:
+            dataset_intended_purpose.add("product_regulated_toy")
+        if data_cc_yaml['intended_purpose']['product_regulated_watercraft'] == True:
+            dataset_intended_purpose.add("product_regulated_watercraft")
+        if data_cc_yaml['intended_purpose']['biometric_categorization'] == True:
+            dataset_intended_purpose.add("biometric_categorization")
+        if data_cc_yaml['intended_purpose']['emotion_recognition'] == True:
+            dataset_intended_purpose.add("emotion_recognition")
+        if data_cc_yaml['intended_purpose']['critical_infrastructure'] == True:
+            dataset_intended_purpose.add("critical_infrastructure")
+        if data_cc_yaml['intended_purpose']['admission'] == True:
+            dataset_intended_purpose.add("admission")
+        if data_cc_yaml['intended_purpose']['recruitment'] == True:
+            dataset_intended_purpose.add("recruitment")
+        if data_cc_yaml['intended_purpose']['public_assistance'] == True:
+            dataset_intended_purpose.add("public_assistance")       
+        if data_cc_yaml['intended_purpose']['victim_assessment'] == True:
+            dataset_intended_purpose.add("victim_assessment")
+        if data_cc_yaml['intended_purpose']['polygraph'] == True:
+            dataset_intended_purpose.add("polygraph")
+        if data_cc_yaml['intended_purpose']['judicial'] == True:
+            dataset_intended_purpose.add("judicial")
+
+        if project_intended_purpose not in dataset_intended_purpose:
+            print("You are not compliant")
+
+        # Now do the exact same thing for all models
+
+        model_intended_purpose = () 
+
+        if model_cc_yaml['intended_purpose']['safety_component'] == True:
+            dataset_intended_purpose.add("safety_component")
+        if model_cc_yaml['intended_purpose']['product_regulated_machinery'] == True:
+            dataset_intended_purpose.add("product_regulated_machinery")
+        if model_cc_yaml['intended_purpose']['product_regulated_toy'] == True:
+            dataset_intended_purpose.add("product_regulated_toy")
+        if model_cc_yaml['intended_purpose']['product_regulated_watercraft'] == True:
+            dataset_intended_purpose.add("product_regulated_watercraft")
+        if model_cc_yaml['intended_purpose']['biometric_categorization'] == True:
+            dataset_intended_purpose.add("biometric_categorization")
+        if model_cc_yaml['intended_purpose']['emotion_recognition'] == True:
+            dataset_intended_purpose.add("emotion_recognition")
+        if model_cc_yaml['intended_purpose']['critical_infrastructure'] == True:
+            dataset_intended_purpose.add("critical_infrastructure")
+        if model_cc_yaml['intended_purpose']['admission'] == True:
+            dataset_intended_purpose.add("admission")
+        if model_cc_yaml['intended_purpose']['recruitment'] == True:
+            dataset_intended_purpose.add("recruitment")
+        if model_cc_yaml['intended_purpose']['public_assistance'] == True:
+            dataset_intended_purpose.add("public_assistance")       
+        if model_cc_yaml['intended_purpose']['victim_assessment'] == True:
+            dataset_intended_purpose.add("victim_assessment")
+        if model_cc_yaml['intended_purpose']['polygraph'] == True:
+            dataset_intended_purpose.add("polygraph")
+        if model_cc_yaml['intended_purpose']['judicial'] == True:
+            dataset_intended_purpose.add("judicial")
+
+        if project_intended_purpose not in dataset_intended_purpose:
+            print("You are not compliant")
+
+
 
     # # If the project is a GPAI model, check that is has met all the requirements for such systems: 
 

data_cc.yaml

@@ -1,6 +1,47 @@
+intended_purpose:
+  safety_component: # Art. 6(1)(a)
+    verbose: 'This dataset is appropriate to use for AI projects involving product safety components'
+    value: !!bool false 
+  product_regulated_machinery: # Art. 6(1)(b); Annex I
+    verbose: 'This dataset is appropriate to use for AI projects involving products covered by Directive 2006/42/EC of the European Parliament and of the Council of 17 May 2006 on machinery, and amending Directive 95/16/EC (OJ L 157, 9.6.2006, p. 24) [as repealed by the Machinery Regulation]'
+    value: !!bool false 
+  product_regulated_toy: # Art. 6(1)(b); Annex I
+    verbose: 'This dataset is appropriate to use for AI projects involving products covered by Directive 2009/48/EC of the European Parliament and of the Council of 18 June 2009 on the safety of toys (OJ L 170, 30.6.2009, p. 1)'
+    value: !!bool false 
+  product_regulated_watercraft: # Art. 6(1)(b); Annex I
+    verbose: 'This dataset is appropriate to use for AI projects involving products covered by Directive 2013/53/EU of the European Parliament and of the Council of 20 November 2013 on recreational craft and personal watercraft and repealing Directive 94/25/EC (OJ L 354, 28.12.2013, p. 90)'
+    value: !!bool false 
+  biometric_categorization: # Art. 6(2); Annex III(1)(b)
+    verbose: 'This dataset is appropriate to use for AI projects involving biometric categorisation, according to sensitive or protected attributes or characteristics based on the inference of those attributes or characteristics'
+    value: !!bool false 
+  emotion_recognition: # Art. 6(2); Annex III(1)(c)
+    verbose: 'This dataset is appropriate to use for AI projects involving emotion recognition'
+    value: !!bool false 
+  critical_infrastructure: # Art. 6(2); Annex III(2)
+    verbose: 'This dataset is appropriate to use for AI projects involving safety components in the management and operation of critical digital infrastructure, road traffic, or in the supply of water, gas, heating or electricity'
+    value: !!bool false 
+  admission: # Art. 6(2); Annex III(3)(a)
+    verbose: 'This dataset is appropriate to use for AI projects involving the determination of access or admission or to assigning natural persons to educational and vocational training institutions at all levels'
+    value: !!bool false 
+  recruitment: # Art. 6(2); Annex III(4)(a)
+    verbose: 'This dataset is appropriate to use for AI projects involving the recruitment or selection of natural persons, in particular to place targeted job advertisements, to analyse and filter job applications, and to evaluate candidates'
+    value: !!bool false 
+  public_assistance: # Art. 6(2); Annex III(5)(a)
+    verbose: 'This dataset is appropriate to use for AI projects intended to be used by public authorities or on behalf of public authorities to evaluate the eligibility of natural persons for essential public assistance benefits and services, including healthcare services, as well as to grant, reduce, revoke, or reclaim such benefits and services'
+    value: !!bool false 
+  victim_assessment: # Art. 6(2); Annex III(6)(a)
+    verbose: 'This dataset is appropriate to use for AI projects intended to be used by or on behalf of law enforcement authorities, or by Union institutions, bodies, offices or agencies in support of law enforcement authorities or on their behalf to assess the risk of a natural person becoming the victim of criminal offences'
+    value: !!bool false 
+  polygraph: # Art. 6(2); Annex III(7)(a)
+    verbose: 'This dataset is appropriate to use for AI projects intended to be used by or on behalf of competent public authorities or by Union institutions, bodies, offices or agencies as polygraphs or similar tools'
+    value: !!bool false 
+  judicial: # Art. 6(2); Annex III(8)(a)
+    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'
+    value: !!bool false 
+
 data_and_data_governance:
   data_governance: # Art. 10(1)-(2)
-    verbose: 'The dataset has been subject to data governance and management practices'
+    verbose: 'The dataset was subject to data governance and management practices appropriate to the intended use case'
     value: !!bool false
   design_choices:  # Art. 10(2)(a)
     verbose: 'The dataset has been subject to data governance and management practices as regards its relevant design choices'
@@ -106,22 +147,21 @@ quality_management_system: # Art. 17(1)(f)
     value: !!bool false
 
 gpai_requirements: # Art. 53(1); Annex XI(2)(c)
-  gpai_requirements: 
-    data_type: 
-      verbose: 'Documentation for the dataset is available that contains the type of data'
-      value: !!bool false
-    data_provenance: 
-      verbose: 'Documentation for the dataset is available that contains the provenance of data'
-      value: !!bool false 
-    data_curation:
-      verbose: 'Documentation for the dataset is available that contains the curation methodologies (e.g. cleaning, filtering, etc.)'
-      value: !!bool false 
-    data_number:
-      verbose: 'Documentation for the dataset is available that contains the number of data points'
-      value: !!bool false 
-    data_scope:
-      verbose: 'Documentation for the dataset is available that contains the number of data scope and main characteristics'
-      value: !!bool false 
-    data_origin:
-      verbose: 'Documentation for the dataset is available that contains information on how the data was obtained and selected as well as all other measures to detect the unsuitability of data sources and methods to detect identifiable biases'
-      value: !!bool false 
+  data_type: 
+    verbose: 'Documentation for the dataset is available that contains the type of data'
+    value: !!bool false
+  data_provenance: 
+    verbose: 'Documentation for the dataset is available that contains the provenance of data'
+    value: !!bool false 
+  data_curation:
+    verbose: 'Documentation for the dataset is available that contains the curation methodologies (e.g. cleaning, filtering, etc.)'
+    value: !!bool false 
+  data_number:
+    verbose: 'Documentation for the dataset is available that contains the number of data points'
+    value: !!bool false 
+  data_scope:
+    verbose: 'Documentation for the dataset is available that contains the number of data scope and main characteristics'
+    value: !!bool false 
+  data_origin:
+    verbose: 'Documentation for the dataset is available that contains information on how the data was obtained and selected as well as all other measures to detect the unsuitability of data sources and methods to detect identifiable biases'
+    value: !!bool false 

model_cc.yaml

@@ -1,3 +1,44 @@
+intended_purpose:
+  safety_component: # Art. 6(1)(a)
+    verbose: 'This model is appropriate to use for AI projects involving product safety components'
+    value: !!bool false 
+  product_regulated_machinery: # Art. 6(1)(b); Annex I
+    verbose: 'This model is appropriate to use for AI projects involving products covered by Directive 2006/42/EC of the European Parliament and of the Council of 17 May 2006 on machinery, and amending Directive 95/16/EC (OJ L 157, 9.6.2006, p. 24) [as repealed by the Machinery Regulation]'
+    value: !!bool false 
+  product_regulated_toy: # Art. 6(1)(b); Annex I
+    verbose: 'This model is appropriate to use for AI projects involving products covered by Directive 2009/48/EC of the European Parliament and of the Council of 18 June 2009 on the safety of toys (OJ L 170, 30.6.2009, p. 1)'
+    value: !!bool false 
+  product_regulated_watercraft: # Art. 6(1)(b); Annex I
+    verbose: 'This model is appropriate to use for AI projects involving products covered by Directive 2013/53/EU of the European Parliament and of the Council of 20 November 2013 on recreational craft and personal watercraft and repealing Directive 94/25/EC (OJ L 354, 28.12.2013, p. 90)'
+    value: !!bool false 
+  biometric_categorization: # Art. 6(2); Annex III(1)(b)
+    verbose: 'This model is appropriate to use for AI projects involving biometric categorisation, according to sensitive or protected attributes or characteristics based on the inference of those attributes or characteristics'
+    value: !!bool false 
+  emotion_recognition: # Art. 6(2); Annex III(1)(c)
+    verbose: 'This model is appropriate to use for AI projects involving emotion recognition'
+    value: !!bool false 
+  critical_infrastructure: # Art. 6(2); Annex III(2)
+    verbose: 'This model is appropriate to use for AI projects involving safety components in the management and operation of critical digital infrastructure, road traffic, or in the supply of water, gas, heating or electricity'
+    value: !!bool false 
+  admission: # Art. 6(2); Annex III(3)(a)
+    verbose: 'This model is appropriate to use for AI projects involving the determination of access or admission or to assigning natural persons to educational and vocational training institutions at all levels'
+    value: !!bool false 
+  recruitment: # Art. 6(2); Annex III(4)(a)
+    verbose: 'This model is appropriate to use for AI projects involving the recruitment or selection of natural persons, in particular to place targeted job advertisements, to analyse and filter job applications, and to evaluate candidates'
+    value: !!bool false 
+  public_assistance: # Art. 6(2); Annex III(5)(a)
+    verbose: 'This model is appropriate to use for AI projects intended to be used by public authorities or on behalf of public authorities to evaluate the eligibility of natural persons for essential public assistance benefits and services, including healthcare services, as well as to grant, reduce, revoke, or reclaim such benefits and services'
+    value: !!bool false 
+  victim_assessment: # Art. 6(2); Annex III(6)(a)
+    verbose: 'This model is appropriate to use for AI projects intended to be used by or on behalf of law enforcement authorities, or by Union institutions, bodies, offices or agencies in support of law enforcement authorities or on their behalf to assess the risk of a natural person becoming the victim of criminal offences'
+    value: !!bool false 
+  polygraph: # Art. 6(2); Annex III(7)(a)
+    verbose: 'This model is appropriate to use for AI projects intended to be used by or on behalf of competent public authorities or by Union institutions, bodies, offices or agencies as polygraphs or similar tools'
+    value: !!bool false 
+  judicial: # Art. 6(2); Annex III(8)(a)
+    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'
+    value: !!bool false 
+
 risk_management_system:
   foreseeable_risks: # Art. 9(2)(a)
     verbose: 'Known or reasonably foreseeable risks the model can pose to health or safety when used for intended purpose'
@@ -140,7 +181,7 @@ obligations_for_providers_of_gpai_models:
     verbose: 'The architecture and number of parameters'
     value: !!bool false
   input_output_modality: # Art. 53; Annex XI(1)(1)(e)
-    verbose: 'Modality (e.g. text, image) and format of inputs and outputs'
+    verbos: 'Modality (e.g. text, image) and format of inputs and outputs'
     value: !!bool false
   license: # Art. 53; Annex XI(1)(1)(f)
     verbose: 'The license'

project_cc.yaml

@@ -157,7 +157,7 @@ high_risk_ai_system:
   critical_infrastructure: # Art. 6(2); Annex III(2)
     verbose: 'AI project is intended to be used as a safety component in the management and operation of critical digital infrastructure, road traffic, or in the supply of water, gas, heating or electricity'
     value: !!bool false 
-  educational: # Art. 6(2); Annex III(3)(a)
+  admission: # Art. 6(2); Annex III(3)(a)
     verbose: 'AI project is intended to be used to determine access or admission or to assign natural persons to educational and vocational training institutions at all levels'
     value: !!bool false 
   recruitment: # Art. 6(2); Annex III(4)(a)
@@ -222,7 +222,10 @@ risk_management_system:
 
 technical_documentation:
   drawn_up: # Art. 11(1)
-    verbose: 'Technical documentation for the high-risk AI project has been drawn up before the system has been placed on the market or put into service and will be kept up-to date'
+    verbose: 'Technical documentation for the AI project has been drawn up before the system has been placed on the market or put into service and will be kept up-to date'
+    value: !!bool false 
+  kept: # Art. 16(d), Art. 18
+    verbose: 'The technical documentation for the AI project that has been drawn up before the system has been placed on the market or put into service and will be kept up-to date will be kept by the provider for a period of ten years'
     value: !!bool false 
   intended_purpose: # Art. 11(1); Annex IV(1)(a)
     verbose: 'The Technical Documentation includes a general description of the AI project that covers its intended purpose, the name of the provider and the version of the system reflecting its relation to previous versions'
@@ -265,6 +268,9 @@ record_keeping:
   logging_generally: # Article 12(1)
     verbose: 'The AI project technically allows for the automatic recording of events (logs) over the lifetime of the system'
     value: !!bool false 
+  logging_kept: # Article 12(1), Article 19(1)
+    verbose: 'The automatic recording of events (logs) that are being generated over the lifetime of the system are being kept by the provider for at least sixth months'
+    value: !!bool false 
   logging_risk: # Art. 12(1)(a)
     verbose: 'The AI project technically allows for the automatic recording of events (logs) over the lifetime of the system and these logging capabilities enable the recording of events relevant for identifying situations that may result in the AI projectpresenting a risk within the meaning of Article 79(1) or in a substantial modification'
     value: !!bool false 
@@ -478,4 +484,7 @@ obligations_for_gpai_models_with_systemic_risk:
     value: !!bool false 
   cybersecurity: # Art. 55(1)(d)
     verbose: 'Adequate cybersecurity protection for any models and infrastructures in the AI project was ensured'
-    value: !!bool false 
+    value: !!bool false 
+
+
+