Social Good

Many organizations make their technology available to vulnerable populations. Where possible, we want to investigate how existing technology and/or telemetry could be used to address key issues pertaining to vulnerable populations. In addition, we want to develop new techniques to try and solve specific problems in the areas of abuse, scams, and child online safety. The Foster a Secure Tomorrow (FAST) initiative for providing technological support to foster children is an example of NortonLifeLock's commitment to protecting vulnerable populations. The free cyberbullying e-book released during the #RaiseOurVoicescampaign in 2017 is another one. There are many similar needs in the areas of abuse (e.g. Intimate Partner Violence, Human Sex Trafficking, Sextortion), scams, and child online safety.


Intimate Partner Violence affects one third of women, and one sixth of men in the United States. Abusers use mobile applications for surveillance, most of which are difficult to block because they have benign use cases, such as location-sharing and text message synching. Victims typically are not technically savvy enough to understand privacy threats and to know how to disentangle their digital life from that of their abuser. Human Sex Trafficking is rampant worldwide in both developed and undeveloped nations. A recent US-based survey highlights that human trafficking involves minors from vulnerable populations extensively within the US, and that most individuals involved in prostitution are trapped and unaware of resources to help them escape. This problem has received some attention in the past from technology companies (e.g. Project Intercept). Sextortion: Non-consensual sharing/theft of nude pictures has long been a serious problem, and there is a pandemic of cameras being placed in bathroom stalls and locker rooms to film victims without their consent (especially in South  Korea). These images and videos are monetized through sextortion and are posted on pornographic sites. The same signature-based technology being used by Thorn in cases of human and sex trafficking are applicable here. The development of nuanced privacy notifications, and context-based detection of dual-use applications when used for malicious purposes can be employed to counteract the abuse. Chatbots can also be leveraged to gather information about abusers.


The dominant problem facing NortonLifeLock customers are fake tech support scams. For instance, customers are contacted through a variety of means by fraudsters claiming to be Microsoft, NortonLifeLock, or even non-existent anti-virus providers, who claim that the user’s system is infected and must be remediated. Fake kidnapping scams have spread from third world countries with high rates of real kidnappings to the United States and other countries. Victims lose massive amounts of money. A combination of methods can be employed to combat these scams. These methods include using ML to detect scam callers, providing reputation metrics for source phone numbers, and leveraging chat bots to extract information about scammers.

Child Online Safety

It is often the case that child supervision applications lead to resentment, frustration, and many are privacy-invasive, even to an illegal extent.  Child online safety solutions should instead help parents and children work together to coach children, educate them, and teach them to self-regulate, while giving the parents a degree of comfort. In addition, careful thought needs to be given to help parents and children deal with technology addictions, online bullying, and mental health impacts of technology overuse.  Two methods to address this issues are the use of VPN solutions + sentiment analysis + text analytics (e.g. stereotype detection) and  analysis of URL categorization feedback data (e.g. RuleSpace) to detect new sources of problematic content.