All posts by Kaustub Bhati

Law Enforcement v. End-to-End Encryption

In a post-Snowden world, there has been relatively more awareness and interest in the right to privacy regarding digital communications; and in knowing when the government can snoop-in on personal conversations. A majority of the communications taking place today are digital and involve two crucial processes i.e. encryption and decryption. Encryption (which is conversion of information into a code) happens when a message/call is initiated. At the same time, decryption (conversion of code back into useful information) happens when the message/call is received by the recipient. There are multiple nuances in this process; both in the technological aspect and the legal aspect.

For quite a while now, WhatsApp chat pages show the message – “Messages and chats are now protected with end-to-end encryption.”. The end-to-end encryption or E2EE (first used in program called Pretty Good Privacy, by Phil Zimmermann in 1991) is a form of encryption that makes it improbable if not impossible to intercept a private conversation. Traditionally, there are 3 instances when a conversation can be intercepted – firstly, from the device of the sender before encryption, secondly, when the information code in is transmission and thirdly from the device of the recipient after decryption.

The two ends i.e. the sender and the recipient, stay vulnerable to unwanted physical access or hacking but it’s the second instance where majority of the snooping takes place. It is here that E2EE becomes useful, for tech companies to bypass court orders and in extension protect user data. E2EE in the simplest of terms means that two people communicating are the only ones who have the specific keys to decrypt each other’s messages and any other person who intercepts such data will have nothing but an unintelligible code. While most communication apps or telecommunication providers have the decryption keys in their own servers; which grant them the ability to see or hear any conversation that passes through their servers, E2EE eliminates this obstacle by giving the keys to both individuals and not the service provider. Imagine the system as that of a letter-box, anyone can put in the messages and lock it (public key) but only the intended recipient has the key to unlock his messages.

This effective bypassing of the service providers has both pros and cons to it. On one hand, it allows for greater freedom for expression of opinions and beliefs without the fear of any sanctions, while on the other hand it stops the governments from carrying out intelligence activities vital for national security. The government, in ensuring the safety of its citizens, does covert operations such as surveillance which enables them to intercept vital communications between suspects; which may lead to stopping of terrorist threat. The importance of such can be gauged from the fact that the latest attack in London included secure-device (encryption) communications between the terrorists and also that ISIS issued instructions for its followers on how to communicate through encrypted apps to plan attacks.

Privacy however is not the only lens through which encryption can be seen in a global setting, for example, promotion and use of E2EE is seen as a human rights issue, as it furthers individual privacy and freedom of expression, which are two rights contained in the International Covenant on Civil and Political Rights (ICCPR). Yet UN reports like “The Right to Privacy in a Digital Age” and “The Promotion and Protection of the Right to Freedom of Opinion and Expression” expound on the idea that judicially ordered decryption is not violative of human rights and has laid down a three part test to limit when a government can restrict encryption.

There is an intense debate about the curbing of powers of law enforcement authorities to gather information through court notices from service provider companies. This debate gained public light after the incident in Brazil, where Facebook had its assets frozen subsequent to its non-compliance of a court order to provide WhatsApp conversational information of a bank robbery gang; which Facebook couldn’t have provided even if it wanted to, as it had no means to do so after enabling E2EE. This incident coupled by Apple’s refusal to the FBI to decrypt the iPhone of the San Bernardino shooter and install a backdoor in their operating system for use by law enforcement, have prompted the UK government to take the issue one step further by enacting a new legislation for surveillance through equipment interference.

The intelligence gathering aspect of E2EE is marred with internal conflictions in the state itself as the state needs such encryption tools to secure its own data but also resents it, as it makes public surveillance harder. Hence, while promoting stronger encryption programs for state use, they limit their citizen’s ability to do so. While certain states like Germany encourage public use of E2EE to avert the covert intelligence gathering abilities of the FIVE EYES countries.

Another facet of this issue is the restrictions in commercial and export area; profit driven tech companies in order to boost sales promote E2EE (more popularity, more sales) and oppose state-imposed rules as this would mean that they can import or build only those applications which allow third party access. Since, every state strives for stronger encryption tools to protect their own data and deal with upcoming security threats, there is state imposed regulations on selling such technology by tech companies to certain states for the purposes of national security and foreign policy goals.

A hypothetical solution to this problem of law enforcement and national security versus privacy can be the innovation of an internal system within the system of such service providing companies like WhatsApp. The internal system, when established would compare every number that tries to send a message with a blacklist (The numbers law enforcement wants to track with judicial approval). When a blacklisted number tries to send a message, the server can stop the E2EE services for the said number from that point onwards and the collected information can be stored in a separate database which can only be accessed by the company’s department handling judicial obligations.

Technology has and will continue to benefit us in ways that cannot be counted, but unaccountable use of such is also capable of great harm. The need for security more than privacy might result in a paradigm shift against rigid privacy laws as is prima facie seen in lawmakers of Florida after the Orlando attacks. It is the cooperation of law and technology together that will result in swifter disbursal of justice and achieve a balance between privacy security and public safety. The adoption of a system such as the one suggested above might be the first step to strike that balance between privacy and safety.

Reconstructing a Crime Scene: Virtual Reality in Courtrooms

Virtual Reality is the latest buzz in the technological sphere, especially with the arrival of VR devices like headsets from giants like Facebook (Oculus Rift), Sony (PlayStation VR) and HTC (Vive). It is a relatively old concept (Aspen Movie Map was the first example, created by MIT in 1978) but with advancement in contemporary technologies, virtual reality has progressed by leaps and bounds in its effectiveness i.e. from being a mere 3-D image to an immersive and interactive system. Apart from its use in gaming and other entertainment purposes, it has been proposed to use this technology in another rather unexpected facet that of judicial proceedings.

Prior to the use of the technology it is important to analyse the concerns regarding the technology in question in a courtroom setting. The concerns are multifold: the manipulation of facts regarding the actual scene of crime/accident, time delay in court proceedings, cost of the expensive procedure and the possibility of manipulation of the trier of facts (judge/jury) i.e. inducing bias. While all of these are relevant questions regarding the use of virtual reality in courtrooms, each of the above concerns can be mitigated.

The problem repeatedly being raised in regard to VR i.e. the manipulation of facts is actually a factual misconception of what VR is capable of. Virtual reality is a computer generated 3-D interactive simulation, in simpler terms, it is a computer-generated environment which simulates some of our senses like vision and hearing to make the artificial environment seem real. The virtual environment, in this case, the scene of the crime/ accident is made by the computer, using 2-D images of the real-world place with minor adjustments by a technician in giving the architectural design the same aesthetics as that of the real-world place. In the case of a shooting, the area of impact and the trajectory of the bullets, or in the case of stabbing, the angle of wound and weapon penetration, are all put inside the simulation after gathering factual information from standard forensic tests. Hence, nothing in the simulation is the manifestation of a recollection from memory. The layout of the virtual crime scene can be checked using the photographs of the real crime scene for any inconsistencies, removing the question of manipulation of facts.

The issue of time delay can be also dealt with. Software like crime scene virtual tour  and IC-CRIME can create a virtual simulation from the 2-D images of the crime scene significantly faster. The cost concerns are very pragmatic issues keeping in mind the novelty of the technology and hence its expensiveness, it can be argued that the relatively wealthy party in a case might use its resources to use this technology while the other party could not afford to fight fire with fire. But, the facts of the scene remaining same for both, it should be the prerogative of the state to provide a virtual simulation to be used by both the parties. Also, with further improvement in technology the cost of using such technologies will go down, hence making it more accessible.  Countries like UK, USA and Switzerland have already granted funds for further research and use of this technology in courtrooms.

The concern regarding the introduction of bias can be nullified as virtual reality instead of inducing bias has been proven to show a reduction in bias in both jury and judges. Firstly, no trial is free of bias. Bias is “inclination or prejudice for or against one person or group, especially in a way considered to be unfair.” It is of two types – explicit and implicit. Every human possesses both the types, formed through an individual’s life experiences. The notion of a right to fair trial rests on the notion of a judge/jury being unbiased and hence capable of looking at evidences objectively. To further this notion, a process called voir dire, is performed to eliminate prospective jury members on the criteria of certain signs of bias shown by them i.e. explicit bias.

Since explicit bias manifests itself in our actions and is easy to root out, the problem lies with implicit bias as it rests in our sub-conscious and is very hard to notice. An individual may/ may not be aware of his/her own implicit biases. Studies have shown that an average American judge/jury form an implicit relation between black person and weapon and between a white person and innocence. Contemporary studies have shown that a black judge is more likely to give a harsher sentence to a white person than a black person and vice versa. From the results of these studies it can be reasonably said that the strong preference of white judges for white people and black judges for black people is based on a very fundamental emotion – empathy i.e. the ability to feel from another’s perspective.

Virtual Reality comes into play in such a situation by providing the very platform to see another man’s perspective. Studies by Prof. Mel Slater from University of Barcelona have shown that when people with implicit racial biases (tested from above studies) were put in virtual simulation and made to see a mirror reflecting an individual, different in race than the person in simulation (technically called bodyswapping); people showed reduced implicit biases and more empathy. Since bias and empathy cannot be rooted out of an individual, the next best thing that can be, is have a diversity of biases so that no one starts severely disadvantaged.

The role of empathy in a judicial decision is not a quantifiable one but its presence is an unmistakable one even though not explicitly expressed. Judges hesitate to show empathy towards any party because of the stigma; that any empathy from a judge in his professional capacity is a taboo and should not be condoned. Also, the more practical reason is that any exhibition of empathy might invite a strict scrutiny from the appellate courts and along with it bring the possibility of mar on their performance reviews.

But in reality, we need judges to be more empathetic so as to understand why people do certain things and assign corresponding sentences since the purpose of law is not just retribution but also rehabilitation of the convict back in the society. As much as we would like to think that the law is absolute and clear, it is not always so and there is much room for a judge’s discretion.

Apart from the problems and their subsequent solutions, to the use of virtual reality in courtroom discussed above; there are a few more advantages. Stanford studies have shown that virtual reality simulations can help us check the reliability of a witness’s testimony by trying out the simulation from the witness’s perspective. For example, if a witness testifies to see a particular crime happen standing somewhere far-away from the crime scene, it could be checked in the simulation if the witness had the line of sight to see the crime from where he said he was standing. It can also help preserve a crime scene in the virtual world when preserving it in real world is not possible, such as homicide on a busy street or an accident on a bridge.

It is evident from the studies conducted till now that the use of VR in modern-day courtrooms is quite plausible given the fact that it can help incline the human factor i.e. the implicit emotions, in a way that is beneficial for all. Moreover, the practical concerns can very much be mitigated with further research in the technology as it will become more adaptive and hence more efficient for courtroom purposes. The pros of using this technology in courtrooms do appear to outweigh any significant impediments that can be foreseen and hence more research and enforcement of such technology should be promoted in courtrooms around the world for faster and more effective disbursal of justice.