Our digital evidence framework must evolve with AI

In Bangladesh, a person may be criminally liable based on digital evidence in the form of CCTV footage, WhatsApp conversations, Facebook posts, and so on. These are all admissible in legal proceedings. The recent investigation into the death of HSC candidate Nuzshat Jahan in Rangpur illustrates this clearly, wherein police relied on call records, CCTV footage, and witness statements to identify and arrest the accused in an abetment to suicide case.

The admissibility of electronic records before courts was recognised by the Evidence (Amendment) Act, 2022, which established that justice cannot be denied in the digital age. However, our legal system faces a real challenge when digital evidence enters the courtroom. The question arises as to whether or not judges, lawyers, and litigants can be confident that a digital record is authentic, complete, and free from manipulation. These questions require more attention than they currently receive.

Tap Here to Add The Daily Star As A Trusted Source

Unlike physical evidence, digital evidence is easier to alter without leaving traces of original sources. A photograph can be edited in seconds, metadata can be altered, and screenshots can be manipulated. Artificial intelligence (AI) tools now allow the generation of convincing audio and visuals without a person ever speaking the recorded words. Such technological advancement poses a significant challenge to the court system. Courts exist to determine the truth based on the presented evidence and deliver a verdict just as an umpire would. If the evidence itself can no longer be determined to be authentic, the legal mechanisms must develop to be able to differentiate authentic digital evidence from fabricated content.

The existing law requires provisions to address critical factors such as forensic authentication, preservation of digital records, chain of custody, and establishment of evidentiary standards essential for dealing with AI-generated elements. In the absence of such provisions, courts have to largely rely on individual investigators, experts, and judges, giving way to uncertainty that can affect the administration of justice.

The rapid advances of artificial intelligence make the challenge even more critical. Governments globally are debating how AI-influenced evidence should be treated in court. Bangladesh has already started implementing AI-assisted traffic law enforcement, although our statutory framework is still silent on the legal implementations of AI evidence, automated decision-making, and algorithmic accountability.

Some international jurisdictions have implemented comprehensive digital forensic standards, certification requirements for forensic labs, detailed protocols for preserving digital materials, and judicial training programmes focused on advanced technologies. These steps do not eliminate the risk of manipulation; but they do significantly reduce uncertainty and improve public confidence in the court system.

Firstly, Bangladesh’s legislative body should establish clear and technology-based statutory standards regulating the collection, preservation, authentication, and presentation of digital evidence. This would reduce inconsistencies between investigation and trial. The government could introduce supplementary rules for the Evidence (Amendment) Act, 2022 or amend the law by incorporating the relevant provisions.

Second, there should be continued investment in technology and practice of internationally recognised standards for a specialised digital forensic lab. The credibility of digital evidence largely depends on the quality of the digital forensic lab which evaluated it.

Third, judges, prosecutors, investigators, and defence lawyers should receive regular training on how to work with increasingly advanced technologies. Basic knowledge about metadata, encryption, cloud storage, blockchain records, and AI-generated content should not be seen as specialised technical knowledge that is reserved for forensic experts.

Bangladesh should also develop a comprehensive legal framework that can regulate AI in judicial and law enforcement contexts. Standards for transparency, explainability, accountability, and human oversight should be incorporated in this framework.

The aforementioned recommendations can strengthen public confidence in the justice system in an era where AI is starting to dominate almost every sector. It will certainly be a challenge to ensure that every piece of digital evidence presented before the court is subjected to clear, transparent, and scientifically reliable standards. But at a time when a fabricated image or manipulated recording can damage one’s reputation, influence investigations, and change verdicts, justice demands nothing less.

Md Nazmul Hossain Sefat is legal researcher and former lecturer in Department of Law and Human Rights at the University of Asia Pacific.

Views expressed in this article are the author's own. 

Follow The Daily Star Opinion on Facebook for the latest opinions, commentaries, and analyses by experts and professionals. To contribute your article or letter to The Daily Star Opinion, see our guidelines for submission.