Novel Psychoactive Substances: Slaying the Dragon With Artificial Intelligence
Bibliographic record
Abstract
BACKGROUND: The emergence of novel psychoactive substances (NPSs) has overwhelmed forensic, health care, and regulatory systems. Conventional analytical techniques are ineffective for identifying known compounds but fail against newly synthesized NPSs lacking reference standards. This review explores the roles of artificial intelligence (AI) and machine learning in addressing growing challenges in NPS identification and characterization. METHODS: The authors reviewed the current forensic workflows and the integration of AI-based approaches, including deep learning models, chemical language models, and spectral prediction tools. Particular emphasis was placed on the DarkNPS framework, which uses Long Short-Term Memory networks and SMILES-based data augmentation to generate millions of plausible NPS structures, and on spectral prediction tools, such as Competitive Fragmentation Modeling for Metabolite Identification (CFM-ID) and novel psychoactive substances-mass spectrometry, for in silico MS/MS spectra generation. Additional emerging AI technologies, such as transformers, graph neural networks, and multimodal frameworks, were also examined. RESULTS: AI-based systems significantly reduced the time and resources required for NPS identification by enabling structure generation, spectral prediction, and prioritization without physical standards. The DarkNPS model successfully predicted structures for >8.9 million plausible NPS compounds, with over 90% of the future market NPS accurately anticipated. In silico MS/MS spectral libraries built using AI tools demonstrated high cosine similarity scores (>0.7) with the experimental spectra, allowing top-hit identification in 75%-90% of the cases. This improved efficiency can facilitate more accurate diagnoses, guide timely treatment decisions, and support public health responses to emerging NPS threats. CONCLUSIONS: Integrating AI with traditional analytical chemistry significantly enhanced the speed, scope, precision, and utility of NPS identification, marking a promising shift in forensic toxicology and chemical surveillance.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".