diff --git a/services/admin/admin_ai_service.py b/services/admin/admin_ai_service.py
index f484b264..6ddc8373 100644
--- a/services/admin/admin_ai_service.py
+++ b/services/admin/admin_ai_service.py
@@ -3,6 +3,7 @@
 import tempfile
 import numpy
 import re
+import gc
 from typing import List
 
 from config import Config
@@ -48,6 +49,37 @@ def get_substance(drugs: List[Drug]):
     return drugs_array
 
 
+def get_substance_by_drug_name(drug_names: list[str]):
+    drugs_dict = {}
+    drug_names_words = []
+
+    if not drug_names:
+        return drugs_dict
+
+    for n in drug_names:
+        drug_names_words.append(" ".join(re.findall("\w{4,}", n)))
+
+    model_subst = _get_model("models/noharm-ml-subst.gz")
+    token_subst = _get_model("models/noharm-tk-subst.gz")
+
+    vector = token_subst.transform(drug_names_words)
+    prediction = model_subst.predict(vector)
+    probabilities = model_subst.predict_proba(vector).round(2)
+    MIN_PROB = 0.5
+
+    for idx, p in enumerate(prediction):
+        subst_idx = numpy.where(model_subst.classes_ == p)[0][0]
+        subst_prob = probabilities[idx][subst_idx]
+
+        if subst_prob >= MIN_PROB:
+            drugs_dict[drug_names[idx]] = p
+
+    del model_subst, token_subst, vector, prediction, probabilities
+    gc.collect()
+
+    return drugs_dict
+
+
 def get_factors(conversions):
     if len(conversions) == 0:
         return conversions
diff --git a/services/prescription_view_service.py b/services/prescription_view_service.py
index f2fe5d34..8222749a 100644
--- a/services/prescription_view_service.py
+++ b/services/prescription_view_service.py
@@ -102,6 +102,7 @@ def _internal_get_prescription(
         alerts_data=alerts_data,
         exams_data=exam_data,
         config_data=config_data,
+        is_complete=is_complete,
     )
 
     review_data = _get_review_data(prescription=prescription, is_complete=is_complete)
@@ -522,6 +523,7 @@ def _get_drug_data(
     alerts_data: dict,
     exams_data: dict,
     config_data: dict,
+    is_complete: bool,
 ):
     drug_list = DrugList(
         drugList=drugs,
@@ -578,6 +580,9 @@ def _get_drug_data(
             )
             concilia_list = drug_list.conciliaList(concilia_drugs, [])
 
+        if is_complete and concilia_list:
+            p_drugs = drug_list.infer_substance(pDrugs=p_drugs)
+
     return {
         "drug_list": drug_list,
         "headers": headers,
diff --git a/utils/drug_list.py b/utils/drug_list.py
index 141407bd..aa81fb9a 100644
--- a/utils/drug_list.py
+++ b/utils/drug_list.py
@@ -3,6 +3,7 @@
 
 from utils.dateutils import to_iso
 from services import drug_service
+from services.admin import admin_ai_service
 from utils import stringutils, numberutils, prescriptionutils
 
 
@@ -398,6 +399,27 @@ def changeDrugName(pDrugs):
 
         return result
 
+    @staticmethod
+    def infer_substance(pDrugs):
+        names = []
+        for p in pDrugs:
+            if p["idDrug"] == 0:
+                names.append(p["drug"])
+
+        substances = admin_ai_service.get_substance_by_drug_name(drug_names=names)
+
+        result = []
+        for p in pDrugs:
+            if p["idDrug"] == 0:
+                if p["drug"] in substances:
+                    p["sctid_infer"] = substances[p["drug"]]
+
+                result.append(p)
+            else:
+                result.append(p)
+
+        return result
+
     @staticmethod
     def conciliaList(pDrugs, result=[]):
         for pd in pDrugs:
@@ -434,6 +456,7 @@ def conciliaList(pDrugs, result=[]):
                         ),
                         "time": prescriptionutils.timeValue(pd[0].interval),
                         "recommendation": pd[0].notes,
+                        "sctid": str(pd.Substance.id) if pd.Substance else None,
                     }
                 )