Page 33 - Veritas

5. It is possible to design algorithms that will be able to identify objects in
the video such as accidents, violent acts, robbery, criminal acts etc. This will

reduce the painstaking work of video observation for hours for the purpose
of investigations.

6. Machine learning-based systems can be used to analyze complex DNA
profiles especially in cases of admixing.

7. Through AI, it is possible to find out the number of guns firing, the ca-
liber of the gun by just audio analysis.

8. It can also be used for detecting criminal recidivism and crime forecasting
by the analysis of behaviour and pattern on social media. It can be useful to

find out the people at risk. Both processes require large amounts of data.
9. After extracting the data and analyzing it, from the given inputs like

dead bodies or objects at crime scenes, it can be used to create multiple
animated videos to help reconstruction of crime scenes.

10. Artificial intelligence would be especially helpful in tackling current
problems of cybersecurity and data security by providing robust and intelli-

gent cyber defense systems.

In the current scenario, the evidence generated using artificial intelligence
are not considered as a substitute for human decision making but just a

helping hand for the investigation of the crimes. It is considered as a Deci-
sion Support System (DSS) i.e. can be used in a trial and be used by a judge

as supportive evidence while giving a verdict but not as a substitute to evi-
dence generated by human intervention. Though AI has an exceptional and

ice-breaking potential to advance the field of Forensic Science, there is a
need for more research and development to realize its full potential.

Myth - Criminals are all very intelligent

Fact - Most of them are of average
intelligence.

18 VERITAS