Volume 8, Issue 12, December – 2023 International Journal of Innovative Science and Research Technology

ISSN No:-2456-2165

Forensic Trial Investigation Using AI and Blockchain

Abhishek Kumar
Law College Dehradun
Uttaranchal University
Dehradun, India

Abstract:- Forensic science is a science of how evidences are learning, deep learning and blockchain. The goal adopted by
collected, which is a very important step in criminal UN namely 17 sustainable development goal (SDG) which is
investigation process and plays a vital role in preparing a aimed to be achieved by 2030[1] does not specifically mention
hypothesis. Collaboration of technology and an inter- AI but goal number16 which talks about peace, justice and
disciplinary approach can work wonders in this field. AI strong intuitions can be achieved by the use of AI and
and Blockchain have a great potential in revolutionizing the Blockchain in the field of forensic science. The; SDG’s aim is
way evidences are collected and analyzed and comes under to provide sustainable development in various field which also
the part of IoT (Internet of things). The AI can analyze very include advancement in science and technology, the
complex and vast amount of data by Artificial Neural improvement in the overall life of people and improvement in
Network (ANN) or Convolutional Neural Network (CNN) judicial system through various innovation in science which
and generate a logical output which one can use to arrive at will surely improve the transparency, accountability and
a certain conclusion. The blockchain technology on the accuracy of the judicial system and help to maintain rule of law
other hand protects the integrity and security of the in the nation. The use of AI can contribute towards this goal of
database and make it tamper-proof. AI and Blockchain can the UN SDG’s by improving the accuracy and efficacy of the
go hand in hand and make this process of forensic trial trials and investigation and by providing fair and just legal
investigation an efficient and more accurate process. This proceeding [2]. Therefore, the use of AI and Blockchain can be
technology, does however also comes with number of seen as promoting multiple goals including goal 9: industry,
difficulties and ethical issues that needed to be addressed. innovation and infrastructure of the UN SDG’s. According to
This paper focuses on the applicability of AI and the most recent UN report on the SDGs, which was released in
Blockchain in field of forensic science and also highlights 2022, the 4G network coverage area doubled between the years
various limitation to the use of this technology in this field. 2015 and 2022, encompassing 88 percent of the global
Present article aims to analyze the ever-expanding role of population. Due to the massive transition from the 2G network
AI and blockchain in the criminal investigative techniques to the 3G and finally the 4G networks and the numerous fraud
and its ethical implication in its usage. The papers also and mobile crimes that followed, AI has become indispensable
intend to highlight the various ways in which AI has been in forensic science and trial investigation [3].
contributing towards the Criminal Administration of
Justice. Artificial intelligence and the blockchain technology have
the potential to change the latest trends and scenario in the
Keywords:- Artificial Neural Network (ANN), Blockchain, forensic trial investigation by providing faster, more precise and
Convolutional Neural Network (CNN), IoT (Internet of Things), accurate analyses of the evidence whereas blockchain have
Forensic Science. proved to play a significant role in forensic trial by providing
the security, integrity and transparency of the evidence gathered
I. INTRODUCTION in a database. AI can be used for pattern recognition, facial
recognition, speech and audio analysis, predictive modeling,
Forensic science is multi-disciplinary in nature and serves as sentiment analyses, document analyses etc. similarly block
a solution to one of the key aspects of legal trial, that is, the chain plays a vital role in chain of custody, smart contracts,
accountability of the evidences and finding out the most digital forensic, data management, verification etc. When these
relevant evidence and the link of that evidence to the criminal technologies are used together, they can significantly aid in
activity. The science becomes forensic if it is related to the upholding and defending the victim's dignity as well as in
judicial procedure. The accuracy of the forensic science is assessing the evidence more precisely, accelerating the legal
directly proportional to the efficiency in identifying the most process, and upholding the rule of law.
relevant suspects and provides the answer to 4W’s that are who,
what, where and when, related to a crime. There are various There are various forms of AI used in forensic trial
steps in forensic trial investigation crime scene investigation, investigation such as symbolism, Bayesian inference, nearest
crime detection, analyzing the evidences, collecting trace neighbor, artificial neural network, machine learning and deep
evidences etc. these steps can be performed in much more learning. This study deals with aid of artificial neural network
effective manner by the use of artificial intelligence, machine in forensic science as well as aid of blockchain in ensuring the

IJISRT23DEC886 www.ijisrt.com 971

Volume 8, Issue 12, December – 2023 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
integrity and dignity of the victim and securing the evidence  Uses of Artificial Intelligence
confidentiality. This study will also highlight the advantages
and disadvantages with the use of AI and Blockchain in A. Crime Scene Investigation with AI:
forensic science. Crime scene investigation is the most crucial part of
forensic trial as there are many crimes in which circumstantial
II. ENABLING TECHNOLOGY evidences plays a major role such as dowry death and abetment
to suicide under section 306 and 498A of IPC. This makes
Artificial intelligence in forensic science refers to the use crime scene investigation the most important part of crime
of advance computational techniques, using machine learning investigation and aided by AI can revolutionize the way how
algorithm, computer vision, and natural language processing, to evidences are collected. The police officer's photograph from
analyze and interpret forensic evidence, automate forensic the crime scene is stored in the database, and the AI then marks
processes, and aid forensic investigations [4]. Artificial and remembers the objects in the photograph. If the AI
intelligence basically involves taking an input and processing it discovers that the object is connected to another crime as well,
according to predetermined algorithms. These algorithms may it can be inferred from the output that the crimes are connected
be based on logic, image processing, statistics, operation in some way.
research, psychology, linguistic philosophy, or neurobiology,
among others. The result is a large amount of output, and this is AI based image and video analysis techniques can be used
where human intelligence comes into play. Now the database of to analyze C.C.T.V footage to draw the important information
the evidences and various other outputs generated by AI which about the suspect and for facial recognition, license plate
can easily be tampered are stored online and this issue is solved recognition and object detection. This technique can help
by blockchain. Blockchain, also known as distributed ledger identify the criminal and suspects related to crime and produce
technology refers to a decentralized and transparent digital and create leads for further investigation.
ledger that is used to record and verify transactions across a
network of computers. It uses cryptography to secure data and Spanish National Cyber Security institute is also working
ensure its integrity, make it tamper proof and immutable. Now on developing evidence recognizing tool which will use AI to
this technology if combined together can help and aid a lot in recognize objects uses in crime.
forensic trial investigation.
It is crucial to determine the victim's age and sex if the
III. AI & BLOCKCHAIN IN FORENSIC TRIAL body was discovered much later than the crime was committed.
INVESTIGATION The femur, patella, mandible, and calcaneus bones are
frequently used in predicting sex [7]. There are many crimes in
Artificial intelligence uses various approaches to enable which age becomes a deciding factor, and the court considers
machine learning and process data. Symbolism, which utilizes age very crucially. As a result, it is crucial to identify the
symbols to represent knowledge and reasoning is used in victim's age and sex. Now that AI has advanced to the point
knowledge-based system. Bayesian inference a probabilistic where it is possible to identify the age and sex of the victim by
approach is prevalent in AI application such as Spam detection Artificial Neural Network (ANN) and deep learning network
and medical diagnosis for statistical reasoning and decision [8].
making. For example, if there are three suspects of a crime and
maximum number of evidence are related to A then according B. AI in Pre- Detection of Crime
to probability A becomes the most potential suspect. This In smart cities, a method of criminal detection known as
approach is also known as forensic statistic. In this, the shot spotter is quite innovative. The shot spotter applies
outcome is based on the statistics provided by the input. Nearest artificial neural networks to the data gathered by sensors and
neighbor (considering all the possibility), it is a form of C.C.T.V cameras in order to locate the location of the gunshot
instance-based learning [5], that is utilized in pattern and identify it. Chinese business Hikvision also created
recognition. Artificial neural network (ANN). They are artificial intelligence (AI) specifically for C.C.T.V cameras that
extensively employed in various application such as image can recognize faces and detect any potential illegal activity.
recognition, natural language processing and speech recognition
[6]. Artificial neural network is a web of neurons just like our There is deep neural network system that uses voice
mind which are called perceptron these neurons work together activity detection with hybrid speech feature for deceptive
and forms a neural network which can function according to the speech detection and detect any suspicious and spam calls that
algorithm predetermined by the authorized person and generate can potentially commit fraud [9].
an output.

IJISRT23DEC886 www.ijisrt.com 972

Volume 8, Issue 12, December – 2023 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
C. Prevention of Child Abuse, Pornography, Human platform for a safer world). This AI mobile forensic tool used
Trafficking and Others Crime Related to Women. by law enforcement agency and digital forensic investigators to
extract and analyze the data collected from various
 Content Monitoring technological devices such as mobile phones, GPS, laptop,
Pornographic and content related to child abuse is tablet etc. the software tool such as Cellebrite physical analyzer
continuously monitored over the internet and the suspicious is used to analyze and interpret the extracted data [11].
data can easily be analyzed and removed from the internet. AI
algorithm can also track the user IP address so that in near E. Leveraging AI for Criminal Intelligence Analysis:
future if such person again posts such content the content can This involves gathering and analyzing data on criminal
easily be taken down from the internet. behavior in order to develop an algorithm that will detect and
correlate offenders. VALCRI (visual analytics for sense making
 Natural Language Processing in criminal intelligence analysis) is one such advancement. This
The methods can identify the suspicious language in chat system uses a complex database and analyze the data to co-
logs, messages and social media post and if it finds anything relate various suspects with the crime and creates a hypothesis
that can promote child abuse and pornography, the content is that can potentially lead the law enforcement agency to a
automatically taken down from the internet. All the chats, probable suspect and ultimately justice is delivered effectively
messages and post are analyzed and remembered by the AI and and efficiently.
crime can be prevented. For example- boys locker room
incident on Instagram [10]. F. AI in Trace Evidences and DNA Separation:
The AI evidence database is packed with pictures and
 Pattern Identification images of numerous crime scenes and trace evidences. AI is
AI can analyze the ups and downs of the internet and any utilized in a variety of approaches such as fiber and hair
abrupt changes in a person's activities. From there, they can analysis, gunshot analysis, glass analysis, paint analysis,
follow that individual and any suspicious sudden changes, and chemical and drug material analysis, and it may even aid in the
they may be able to stop them from engaging in unlawful diagnosis of minute particles. All of these evidences and trails
behavior. are examined, and the algorithm generates a logical output from
the vast database on its own. AI-powered next-generation
D. Empower Mobile Forensic with AI driven tool sequencing in DNA samples is being developed [12]. In the
Mobile forensic is the branch of digital forensics which event of mass devastation or other occurrences when forensic
aims at investigating the digital evidence recovered from a cell specimens and samples are difficult to get and analyze, this
phone that can provide a wealth of information in a forensically technique can quickly separate the numerous DNA samples at
sound manner. There are lot of operating system in the market the same time and generate a logical output.
which make it difficult for latest techniques and trick to gain
full insight of the data of the device and collect the relevant Blockchain offers a better solution to the security and
evidences needed for criminal investigation. Latest confidentiality of the database accumulated, as all the databases
advancement in AI has helped and revolutionized the mobile and complex windows of AI generated outputs are very crucial
forensic tool thus making it more convenient and an effective to the legal proceeding and their tampering can cause unfair
way to collect relevant evidences from the mobile devices. One judgement and it will directly impact the credibility of the
such technology is THE CELLIBRITE (digital intelligence criminal judicial system.

Table 1 Use of Ai and Blockchain in Various Steps of Forensic Trial Investigation

Stages of trial investigations Use of AI Use of blockchain
Crime scene investigation Identifying Objects and Evidences Maintaining a Database
Evidence Analyses Analyzing Fingerprints, Facial Recognition, Voice Encrypting the Database to avoid any
Recognition, DNA sampling tampering with the evidences
Creating Hypothesis Checking all the probability No Information Leakage
Data Management Analyzing the Complex data Securing the integrity of data and
producing it to the person who is
authorized to access it

In blockchain all the communication in the internet or the A blockchain that consists of four chains of custody or a
IoT (internet of things) are stored in the blockchain as series of transactions can be used to store all circumstantial
transaction, and this creates the existing chain of custody evidence. However, altering the blockchain necessitates the
process pretty handy [13]. creation of new blocks or transactions, which will eventually
have an impact on the current series of blocks and alert the
authorized person of any tampering with the evidence and
database [14].

IJISRT23DEC886 www.ijisrt.com 973

Volume 8, Issue 12, December – 2023 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165

 Principals to be followed to design secure distributed ledger

of blockchain:

 Flexibility in the structure of ledger:

The need of the clients is changing because of the
dynamic nature of the society, the ledger should be design in
such a way that it can be modified according to the need and
increase the adaptability of the ledger.

 Easy fault detection:

The ledger must be incorporated with smart AI that can
easily identify and detect any fault in the ledger the moment it
arises. This will not cause any misappropriation in the result of
the output created by the system. Fig. 1. Chain of Custody
 Linear performance: There are various steps involve in collection of evidences:
This is a very big challenge for an AI and Blockchain  Samples received
driven system that it must handle the complex large scale
 Case is assigned to the investigator
distributed ledgers and integrate them to work in a linear
 Finalizing of report
progression. This defines the performance of the system
generated.  Dispatched (sec 173 of Indian Evidence Act) to the
magistrate
 Reliability:
This is one of the most important principals and ranked at Because of 4 chain of custody records cannot be altered at
the top priority in the system generated for forensic trial any stage and if there is any alteration then authorized person
investigation. The outcomes generated should be reliable as will get to know immediately. Blockchain protect the security
they will be used in the court of law as evidence before the and integrity of the victim as well as the accused and it also
judges. So, there must be a sense of reliability on the output ensures the better enforceability of the section 165 of the Indian
created by the system so that there can exist proper and fair Evidence Act, which prohibits tampering and alteration of the
delivery of justice. evidences. The blockchain system uses hash encryption for the
security and provide a unique access to only the authorized
 Security in the ledger: person so as to avoid tampering. The admissibility of the
The ledger must be secure and the data must be tamper- document in the court requires chronological documentation so
proof. Security should be everywhere in the blockchain as to achieve this a broader chain of custody is needed to
mechanism. The sole purpose of using an advance technology facilitate modern forensic and incident response.
of blockchain is to have the sense of security that the data will
be secured and there will be no manipulation. Now there are various stages of how blockchain
technology helps in the efficient and effective functionality of
 Expandable: forensic trial investigation process. These stages are:
The ledger designed should be in a way that it can Preparation, media acquirement, gathering evidences,
incorporate new services and new tools. This will make the transportation of evidences, analysis and report.
ledger a system that will be future-proof. This will reduce the
cost in the future and the survival rate of the system in Now in the preparation stage all the preliminary work is
advancement of technology will be higher. done and the crime scene investigation is done with collecting
all the relevant evidences. After the preparation stage there
The following principles must be present in a blockchain comes the response stage in which all the relevant media of
that deals with large-scale distributed ledgers. This will crime scene is acquired and the crime scene is safeguarded and
guarantee the blockchain system's effectiveness and efficiency. preserved.
Making it an even more cutting-edge technology. Forensic trial
investigation is a process which require a system that can After this the stage of evidence acquisition stage comes in
generate logical output and at the same time stores it in a place this stage evidence medium and storage medium is checked and
which is secure and tamper-proof. The blockchain technology the evidence is sealed. After all the necessary proceedings the
fulfills all the requirements of the process. evidences which are gathered are carried to the investigating
agency. The gathered data arrives at the investigating agency
then the agency confirms the data and this stage is called the
confirmation stage. Then the copy is generated and the original
report of evidence is preserved. Then at the very last stage the

IJISRT23DEC886 www.ijisrt.com 974

Volume 8, Issue 12, December – 2023 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
analysed report is testified in the court and this stage is known IV. LIMITATION OF AI AND SUGGESTIONS:
as report [15]
 Even if AI is very capable of producing logical output from
At every stage of the investigation through blockchain a complicated database, selecting the most advantageous
new chain of command is used which is authorized and any sort output from the most likely ones still need the human mind
of tampering is checked at the affirmation stage. and intellect. For this reason, a human-in-the-loop method
that will support the researcher rather than replace them
There are various types of blockchain technology on a must be used.
very root level, the technologies are public blockchain, private,  AI is built on algorithms that are applied to data, and if the
blockchain and the consortium blockchain.[16] data input has any bias, the outcome will also have bias,
which will affect how fair the AI result is. To prevent this,
Now, if everyone has access to the ledger, everyone can AI must be updated frequently, and the new algorithm must
monitor the ledger, everyone can verify the chain of be regularly validated.
blockchains, and everyone has control over the chain, then that  Because large neural network AI models perform several
blockchain is a public blockchain. Members of the public operations simultaneously, it is highly difficult to
domain can share and record the ledger [17]. Private ledgers are understand how the AI system arrived at a given result and
those that can only be supervised by the participants and are what strategy it employed, which reduces its explainability
only accessible to them. Only one central authority has the and justification for usage. To solve this Explainable AI
power to monitor the blockchain. Only if the central authority should be used which will allow the investigator on how the
permits the addition of the participant to the ledger can it be AI reached to that conclusion. For ex: XAI by IBM
done. The main difference between the public and private  The accuracy and dependability of AI output depends on the
ledgers is that the private ledger has an exclusive supervisor, correctness and clarity of the data, yet the data acquired at
making it effective and relatively simple to change and amend crime scenes is frequently ambiguous, compromising the
the ledger, whereas the public ledger does not have a central AI's accuracy.
authority overseeing it. The public and private ledgers are  As technology has developed, several more issues with AI
combined to create the consortium blockchain, which is used have arisen, including hacking, a lack of privacy, and
for transactions between parties who have previously been scalability, which renders AI less trustworthy in ethical and
granted authorization. These kinds of blockchains can be legal concerns. In order to resolve this, a more secure
utilized for a wide range of tasks and can be employed in encryption method—such as the Rabin cryptosystem,
accordance with the requirements of the tasks. Merkle tree, Blake2 algorithm, etc.—must be employed to
encrypt the database.
 The cost of integrating AI in forensic science is quite high,
and smaller forensic organizations are unable to integrate AI
into their forensic trial system, which has an impact on the
usability and cost-effectiveness of AI.
 In DNA sampling if there is a very complex data and
sample to analyse then the interpretability may be
compromised so the deep learning can help in getting more
accurate result.
 Technology and tools that are accessible to the common
population are extremely far apart. Because of this
mismatch between tools and technology, fraudsters benefit,
and regular people who are unaware of the technology risk
being conned and becoming a target of scammers.
 When processing data, an AI-driven forensic tool moves
quite slowly. This could reduce the trail's effectiveness
because time is such an important consideration.

V. CONCLUSION

Blockchain and AI integration have the potential to

completely transform the way that legal evidence is gathered
today. Investigating and solving crimes may be made easier and
more effective by integrating AI into the forensic system. In the
legal industry, this technology can provide excellent accuracy
and efficiency. The enormous and complicated database can be
Fig. 2. Stages of Evidence analyses through Blockchain handled and analyzed quickly by AI, while blockchain ensures

IJISRT23DEC886 www.ijisrt.com 975

Volume 8, Issue 12, December – 2023 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
the data's security and integrity. The blockchain technology [9]. S. Mihalache and D. Burileanu, “Using voice activity
guarantees the data's secrecy and makes it tamper-proof. detection and deep neural networks with hybrid speech
feature extraction for deceptive speech detection,”
The technology does, however, also present a number of Sensors, vol. 22, no. 3, pp. 1228-1239, Feb. 2022.
difficulties and ethical issues that need to be addressed, such as doi:10.3390/s22031228.
hacking into the system database that puts privacy at risk and [10]. "Bois locker room: Indian teens’ lewd Instagram group
bias in the AI algorithm that can significantly alter the output of causes outrage," BBC News, Apr. 8, 2023. [Online].
AI and lead to unfair results. The development of the AI system Available: https://www.bbc.com/news/world-asia-india-
must make it more interactive and explicable as to how the 52541298. (Accessed Apr. 8, 2023).
system arrived at that conclusion as technology may assist [11]. O. Afonin and Vladimir Katalov, "Mobile Forensics –
human labour but cannot totally replace human intelligence. Advanced Investigative Strategies," in Google Books,
With further updation, research, and development, the [Online]. Available:
technology can be improved to a much higher level of accuracy https://books.google.com/books/about/Mobile_Forensics_
so as to get a more just outcome and one can ensure that justice Advanced_Investigative.html?id=9oVcDgAAQBAJ.
is served more accurately and efficiently, benefiting both the (Accessed Apr. 9, 2023).
victim and the accused. The integration of AI and blockchain is [12]. M. J. Alvarez-Cubero et al., "Next generation sequencing:
an exciting and cutting-edge technology that offers immense An application in forensic sciences?," Annals of Human
benefits. Biology, vol. 44, no. 7, pp. 581-592, Nov. 2017.
[13]. J. H. Ryu, P. K. Sharma, J. H. Jo, and J. H. Park, “A
REFERENCES blockchain-based decentralized efficient investigation
framework for IOT Digital Forensics,” The Journal of
[1]. H. S. Sætra, “AI in context and the Sustainable Supercomputing, vol. 75, no. 8, pp. 4372–4387, 2019.
Development Goals: Factoring in the unsustainability of doi:10.1007/s11227-019-02779-9.
the sociotechnical system” in Sustainability, vol. 13, no. 4, [14]. H. Al-Khateeb, G. Epiphaniou, and H. Daly, “Blockchain
p. 1738, 2021 [doi:10.3390/su13041738]. for Modern Digital Forensics: The chain-of-custody as a
[2]. "The Sustainable Development Goals (sdgs) and Global distributed ledger,” in Blockchain and Clinical Trial, pp.
Health," ISGlobal. [Online]. Available: 149–168, 2019. doi:10.1007/978-3-030-11289-9_7.
https://www.isglobal.org/en/-/global-health-inequities. [15]. M. Harbawi and A. Varol, "An improved digital evidence
(Accessed: May 11, 2023). acquisition model for the Internet of Things forensic I: A
[3]. "Sustainable development goals: United Nations theoretical framework," in 2017 5th International
Development Programme," UNDP. [Online]. Available: Symposium on Digital Forensic and Security (ISDFS),
https://www.undp.org/sustainable-development-goals. Tirgu Mures, Romania, 2017, pp. 1-6, doi:
(Accessed: April 08, 2023). 10.1109/ISDFS.2017.7916508.
[4]. C. Watson, Q. Shen, and C.M. Bowers (Eds.), "Forensic [16]. P. K. Sharma, S. Singh, Y. -S. Jeong, and J. H. Park,
Applications of Artificial Intelligence and Computer "DistBlockNet: A Distributed Blockchains-Based Secure
Vision," Springer, 2020. SDN Architecture for IoT Networks," IEEE
[5]. N. S. Altman, "An introduction to kernel and nearest- Communications Magazine, vol. 55, no. 9, pp. 78-85, Sep.
neighbor nonparametric regression," in The American 2017, doi: 10.1109/MCOM.2017.1700041.
Statistician, vol. 46, no. 3, pp. 175, Aug. 1992, doi: [17]. M. Harbawi and A. Varol, "The role of digital forensics in
10.2307/2685209. combating cybercrimes," in 2016 4th International
[6]. J. Heaton, "Ian Goodfellow, Yoshua Bengio, and Aaron Symposium on Digital Forensic and Security (ISDFS),
Courville: Deep learning," in Genetic Programming and April 2016, pp. 138-142.
Evolvable Machines, vol. 19, no. 1–2, pp. 305–307, 2017,
doi: 10.1007/s10710-017-9314-z.
[7]. M. K. Turan, Z. Oner, Y. Secgin, and S. Oner, “A trial on
artificial neural networks in predicting sex through bone
length measurements on the first and fifth phalanges and
metatarsals,” Computers in Biology and Medicine, vol.
115, pp. 103490, Dec. 2019.
doi:10.1016/j.compbiomed.2019.103490.
[8]. N. Vila-Blanco, M. J. Carreira, P. Varas-Quintana, C.
Balsa-Castro, and I. Tomas, "Deep Neural Networks for
chronological age estimation from OPG Images," in IEEE
Transactions on Medical Imaging, [Online]. Available:
https://pubmed.ncbi.nlm.nih.gov/32012002/ (accessed
April 08, 2023).

IJISRT23DEC886 www.ijisrt.com 976