Comprehensive Guide to IoMT Cybersecurity – Risks, Safeguards, and What We Protect

The Internet of Medical Things, also known as the IoMT, is one of the most revolutionary developments in healthcare today.  It empowers physicians to monitor patients remotely by providing the patient with network-enabled devices. These devices can track a wide variety of processes, from medication compliance to blood glucose level. 

IoMT has become an extremely profitable industry. Let's look at the statistics:  

• 87 percent of healthcare companies are using or preparing to use IoT technologies 

• 63 percent of IoT technologies in use today are used for patient monitoring 

• 71 percent of researchers believe that IoMT technologies are developing faster than the clinical world is ready to use them 

By 2020, as many as 30 billion IoMT devices will be in use worldwide. By 2022, the market for IoMT will hit $158 billion.

The time is come to get very serious about IoMT cybersecurity.

Why IoMT Cybersecurity Is Worth the Trouble 

IoMT devices offer benefits in every value metric of health care - patient outcomes, patient satisfaction, scientific advancement, and financial viability.

Lower Cost of Care

With the percentage of elderly persons rapidly increasing worldwide, experts are anticipating a potentially overwhelming rise in health care costs. The IoMT can make it possible for doctors to monitor patients remotely and send automatic recommendations when data exceeds the normal range. With this resource, doctors can lower costs by scheduling fewer in-person visits.   

Additionally, because IoMT devices report data automatically, they reduce the need for nurses and support staff to verify data. This makes for more efficient use of staffing resources and focuses more attention on patient care.

Better Outcomes

Today, IoMT is most helpful in ensuring patient compliance with doctors' instructions. An enabled device communicates data regarding outcome-related behaviors such as medication adherence or exercise habits. The same devices can track whether a patient's condition improves or declines, thus providing key information that physicians can use to adjust care recommendations.

Customized Treatment

The pharmaceutical and medical device industries are already using IoMT to innovate around patient-centered treatment. Patients with Parkinson's disease, for example, now have access to wearable sensors that monitor and optimize medication use to improve patients' quality of life. 
 
Additionally, there are numerous possibilities to add IoMT technology to pre-existing devices. For example, for patients with diabetes who use insulin pens, the IoMT could gather blood sugar data or record self-treatment schedules. The resulting data could help doctors better refine patients' treatment plans.

Access to Research

IoMT devices allow researchers to gather vast amounts of data conveniently, thus allowing them to more easily recruit and monitor study participants. Optimizing this process benefits not only the researcher, who can spend more time with participants and evaluating results, but also the many patients who can benefit from experimental treatments.

Innovations in IoMT Technology

IoMT is going places - down a patient's esophagus, to the grocery store, and even in a diabetic's eye, all thanks to technology's ability to get smaller, faster, and smarter.

Wearable medical device examples. Image source: https://www.researchgate.net/figure/Three-types-of-wearable-sensor-nodes-powered-by-thermoelectric-energy-harvesters-The_fig1_279634036

Wearable Medical Devices

As sensors become more advanced, device manufacturers have found ways to integrate them into wearable fabrics. This has led to advancements such as:

• “smart” wound dressing that automates the delivery of painkillers, antibiotics, and tissue regenerating medicines    

• clothing and bedding that can alert caregivers to the needs of bedridden patients   

• bluetooth-enabled gloves that help stroke patients recover function faster 

• EEG-enhanced earbuds and headbands that improve pain management

All of these advancements help patients to get better care and take control of their health outcomes.

Ingestible Sensors

In order to increase the number of patients who take their medication according to doctor's orders – a number calculated by the World Health Organization at just 50 percent – Proteus Digital Health has developed ingestible sensors that track when a patient has taken his or her prescribed pill. This information travels to a smartphone app so that patients and their doctors can monitor adherence.

Glucose Tracking Contact Lens. Image source: https://www.engadget.com/2014/01/17/google-health-smart-contact-lenses-diabetes/

Glucose Monitoring

The Eversense XL system, distributed by Roche, uses an implanted sensor and rechargeable transmitter to send information about blood glucose levels to patients' smartphones. Patients receive vibration alerts when glucose levels drop or rise to unhealthy levels, as well as phone notifications when levels appear to be approaching those levels. Users can also track their daily habits to understand what affects their blood sugar.

Glucose-Tracking Contact Lenses

Yes, you read that right. Google and Novartis have teamed up to create a network-enabled contact lens that measures the blood sugar in patients' tears. The patient can view his or her data through a smartphone app and adjust insulin delivery based on results.

Ingestible sensor - camera. Image source: https://www.marsdd.com/news-and-insights/ingestibles-smart-pills-revolutionize-healthcare/

Swallowable Cameras

Doctors can now view the interior of a patient's digestive system without uncomfortable endoscopy procedures.  It happens thanks to PillCam™, a swallowable camera that provides internal views of the small bowel. As the camera travels through the patient's system, advanced visualization technologies enable the physician to track speed of movement through the bowel and view images of potential abnormalities.           

IoMT Out in the World

Not all IoMT devices are found in hospitals and laboratories. Many, like the consumer health wearables that have pervaded the consumer market, allow patients and their doctors to track health information on a day-to-day basis.

Fitness Trackers

Fitness trackers, like the well-known FitBit, are convenient wearable devices that track the body systems associated with exercise. They use:

• bioimpedance sensors to track heart rate, respiration rate, and blood flow

• thermometers to track increases in skin temperature 

• skin response sensors to detect sweat 

• optical heart rate monitor to measure cardiac exertion 

• accelerometers to count steps

These devices include software that can this information into health advice that the patient can use, but that's as far as it goes. They don't send data to doctors or recommend medications, but they do collect and store user information.

If not properly encrypted, that information could fall into the hands of hackers who might sell or alter it. There is even the possibility that a hacker could attempt to ransom a patient's own health data back to him or her.

Smartwatch Apps 

Many smartwatches now have health tracking systems that monitor the wearer's health data. The recently-released Apple Watch Series 4, for example, tracks the wearer's cardiac rhythm and notifies him or her if it detects any irregular heartbeats. 

The watch's Health app, meanwhile, tracks your stress levels, physical activity, sleep habits, and nutrition. These trackers can connect to a number of other third-party apps, which increases the number of companies that have access to your data.

Remote Patient Monitoring

Remote patient monitoring, also known as RPM, is one of the highest-profile trends in the health care industry.The current shift toward value-based care is making a distinct space for RPM technologies, which have the potential to reduce health care costs while improving patient outcomes and the overall patient experience.

Steps to Hack a Pacemaker

The Downside – IoMT Cybersecurity Concerns

Medical information is very valuable to hackers – up to 10 times more valuable than a credit card number. But it isn't just data that cybersecurity experts fear may be at risk. 

In 2012, the Showtime television series Homeland aired an episode in which a character hacked into the software that controlled the pacemaker of the US vice president. The hacker then used that access to take control of the pacemaker's programming and caused a fatal heart attack. Shortly thereafter, then-vice president Dick Cheney's medical team disabled the wireless feature on his pacemaker, fearing that the television program would inspire a terrorist to take similar action.   

Pacemakers and cybersecurity soon dropped from the news, but they came back again just four years later.

The Pacemaker Recall

In August of 2017, the Food and Drug Administration publicized a recall of 465,000 radiofrequency-enabled implantable pacemakers after a review identified security vulnerabilities. The FDA found that a hacker could use equipment at distributed on the commercial market to gain access to the device. With this access, a hacker could change the device's programming and harm the patient by depleting the battery or order dangerous pacing.

Fortunately, the FDA recall and development of firmware update occurred before any malicious access could occur.

Hacking Owlet infant heart rate monitors

Infant Health at Risk

In 2016 in the UK, the media reported a security vulnerability in the Owlet infant heart monitor sensor. Cybersecurity researchers found that while data sent between parent smartphones and base servers was secure, the networking between the sensor and the base server had no encryption whatsoever and could be accessed without so much as a login. 

 That means that anyone within range could monitor the infant's data, interfere with alert systems, or otherwise interfere with monitoring. The media deemed the situation to be the year's worst IoT security risk.

Hacking insulin pumps. Image source: https://www.extremetech.com/extreme/92054-black-hat-hacker-details-wireless-attack-on-insulin-pumps

Insulin Delivery Made Vulnerable

All of those IoMT enabled glucose monitors? Those are vulnerable to hackers too. 

In 2011, a security researcher who is also a diabetic conducted an independent investigation. He found that a hacker could intercept the signals from a glucose monitor and change the readout, which would likely cause the wearer to adjust the dosage. If the attacker repeated the process, it could prove fatal. More frighteningly, this could happen from up to half a mile's distance.

Hackable Medication Pumps 

Research has also identified vulnerabilities in a particular brand of infusion pump, a type of technology that delivers medication into patients' bloodstreams. In 2014, an independent professional reported that he had written a program that could instruct pumps to deliver lethal doses of medicine.  The researcher found these vulnerabilities in at least five pump models.

Whole Systems at Risk

IoMT devices also put the hospital at risk of what is known as a backdoor attack. In one well-publicized example, hackers used malware to infect blood gas analyzers, which are important for the monitoring of critical care and surgical patients. The infected devices allowed attackers to infiltrate hospital networks and extract confidential data, which was then sent to an undisclosed European location. 

A simulation of the attack revealed that the blood gas analyzers were routinely sending unencrypted data. That meant that the hackers could make any changes that they wished within the hospital system - to patient data, to treatment and diagnostic requests, and even to key administrative information.

Identifying Key Vulnerabilities

The IoMT would be a vulnerable situation even with adequate safeguards in place, simply because it sends and receives extremely personal information. Unfortunately, most of today's systems are not set up to protect such private data.

Outdated Software

Some working medical devices, such as CT scanners and even pacemakers, are upwards of 20 years old. These devices are capable of connecting to hospital systems but feature software systems that have not been conscientiously updated. Such systems are not set up to defend against today's advanced threats, which can disable an entire hospital network through a single device.   

Departments Working in Isolation

In order for medical devices to be truly secure, all stakeholders must participate in screening for and responding to threats. This can be a difficult task in a complex healthcare organization where the primary focus is typically the immediate health of the patient. 

The clinicians who use these devices often have so many pressing patient care tasks that they can easily overlook cybersecurity updates Meanwhile, facility IT staff and device manufacturers become involved with their own team's tasks.   

This isolated way of working does not stand up to the complex and interlocking demands of medical device cybersecurity. In order to defend systems and devices from intruders, stakeholders must come together to create a unified security strategy. 

Unclear Regulations and Poor Compliance

Manufacturers and healthcare delivery organizations (HDOs) have traditionally lacked clear mandates regarding the security of medical devices. A 2017 survey revealed that just 44 percent of HDOs and 51 percent of device manufacturers followed FDA guidelines to make devices more secure, despite the fact that approximately 33 percent of both groups were aware of potential harm to patients if a security breach did occur.

Improving Security and Protecting Patients

In one of the first major steps toward improved monitoring of cybersecurity, the FDA recognized the UL 2900 standard for medical device cybersecurity in August of 2017. Officially published as UL 2900-1, the standard offers a series of checks designed to identify vulnerabilities.

By recognizing the standard, the FDA enables manufacturers to indicate UL 2900-1 compliance as a response to cybersecurity concerns.

Regulatory Updates

In late 2018, the FDA published draft guidelines that bring medical device cybersecurity requirements up to date with today's risk set. Changes include:

1. A two-tier risk assessment system that prioritizes devices with the ability to connect with other products, a network, or the Web

2. Mandated documentation specific to a device's or system's risk tier         

3. A Cybersecurity Bill of Materials (CBOM) that lists software and hardware with potential vulnerability to hackers 

4. Greater alignment with the National Institute of Standards Technology's globally acknowledged Cybersecurity Framework

The FDA hopes that aligning with the Framework will improve UL 2009-1 compliance.

Recommendations, Not Mandates

The released documentation is geared toward premarket submissions and “contains nonbinding recommendations.” It is yet to be seen whether these recommendations are compelling enough to increase the percentage of manufacturers and HDOs that are actively trying to prevent attacks, a percentage most recently measured at 17 percent and 15 percent respectively.   

Recommendations for HDOs

In 2018, the FDA collaborated with government research contractor MITRE Corp. to develop a cybersecurity guide for medical devices. The guide provides recommendations for healthcare providers and other involved organizations to prevent and respond to cybersecurity incidents involving medical devices. It features such recommendations as:

• Creating a threat detection standard for all of an enterprise's devices and networks 

• Investment in firewalls and other prevention technologies

• Inventorying of an enterprise's full range of medical devices, with attention to those with connectivity capabilities

The latter plays a particularly important role in uncovering security vulnerabilities. Only by identifying areas of potential risk can a provider screen for and correct vulnerabilities.

Next Steps for Attack Prevention

Technology never stays still and neither do hackers. The new developments in protecting patient data and patients themselves will not be the end of healthcare cybersecurity, nor will they guard against every possible way of hacking IoMT devices.   

It is vitally important for all HDO's to develop and implement medical device security and IoMT security strategies. These strategies need to include not only a screening and threat mitigation standard for current devices but also a plan for maintaining security on a continuing basis.

Simplify Clinician Processes

Clinicians, both doctors and nurses, have their hands full with patient care. Any cybersecurity tasks, such as authentication checks, must be minimal on their part and automated as much as possible. Software updating and other processes must not disrupt workflow.

Threat Mitigation Training

The concept of keeping the clinician involved but uninterrupted also applies to threat intervention strategies. All personnel involved with a device or network must know what to do if a data compromise occurs.

Network Segmentation

Many security experts are now advocating for increased segmentation of enterprise systems. By segmenting their networks, organizations can more effectively implement security checks and isolate any threats before they pose a widespread risk. 

Encryption of Patient Information

Because IoMT devices expose more patient information to enterprise networks, it is important for electronic health records (EHRs) to be as secure as possible. Technology such as homomorphic encryption, which secures data even as it is being used, can keep patient information safe from unwanted access. Protection can even be applied to personally identifiable information (PII), which is particularly valuable to cybercriminals. 

Investing in Prevention

 An ounce of prevention is still worth a pound of cure, both in patient care and in IoMT cybersecurity. As much as possible, security analytics and software updating should be automatic. The most critical and complex tasks can be passed on to a third-party provider with expertise in medical device security and the hacking of IoMT devices. 

The FDA has placed the burden of IoMT cybersecurity on the people who develop and use them. This adds a set of high-urgency tasks for companies that are already tasked with the safety and well-being of patients. It is unsurprising that experts urge companies to outsource high-level tasks to a company with the expertise to handle them. 

Alpine Security provides penetration testing services with a focus on IoMT devices. Through extensive experience in ethical hacking and identification of cybersecurity risk, we can relieve HDOs and manufacturers of the necessary burden to review and test all enabled devices, freeing those organizations to focus on patient outcomes.

Alpine Security Demonstrates Commitment to Medical Device Security by Sponsoring the Archimedes Medical Device Security Conference

O’Fallon, IL, January 2, 2019 - Alpine Security is proud to announce its support for the 2019 Archimedes Medical Device Security 101 Conference. As one of the conference's top sponsors, Alpine Security hopes to boost awareness of the cybersecurity vulnerabilities currently present within the medical community — and the many opportunities for improving healthcare cybersecurity.

Hosted by the University of Michigan’s Archimedes Center for Medical Device Security, this year's conference will take place in Orlando on January 21st and 22nd. A variety of key industry players are expected to attend, including product engineering directors, clinical facilities engineers, IT security managers, and more. This year's speakers include Norton Healthcare's AJ Aspinwall, Royal Philips Healthcare's Jonathan Bagnall, and Edward Brennan from the Health Information Sharing and Analysis Center (H-ISAC). Support from Alpine Security and other sponsors has been instrumental in allowing these speakers and a variety of other panelists to share their expertise.

The team at Alpine Security is well aware of the threats that hacked medical devices pose, both in the medical community and in a variety of other industries and settings. Patients, in particular, are at risk — each United States hospital bed currently possesses between 10 and 15 devices, many of which are connected to the facility's network. Just one malfunction can prove deadly, especially for high-risk patients for whom every second is of grave importance.

As Alpine Security CEO, Christian Espinosa, points out, hospitals are uniquely vulnerable because they are open to the public.

"Because of the nature of hospitals and their openness...it's easy for an attacker to walk into a hospital and perform nefarious activities. Nobody is going to question them." - Christian Espinosa, Alpine Security CEO

Espinosa warns that hackers can easily find exposed ethernet connections or even hack into the facility's wireless network while hanging out in the hospital cafeteria.

Unfortunately, medical device security remains shockingly minimal at most healthcare facilities. This is where Alpine Security comes into play.

In addition to supporting the Archimedes Medical Device Security 101 Conference, Alpine Security does its part to improve healthcare security by providing risk assessment and penetration testing services tailored for medical devices and healthcare environments. Using Alpine Security's arsenal of penetration testing services, Healthcare Delivery Organizations (HDOs) and medical device manufacturers can more effectively address the risks brought about by modern medical device connectivity.

About Alpine Security

Alpine Security appreciates that every person and organization face an increasing risk of cyber attacks. They aim to use their resources to do something about it. Alpine Security offers a variety of cybersecurity services, including risk assessments, penetration testing, training, audits, and incident response. The company's team is committed to always remaining a step ahead of today's tech-savvy hackers.

About Archimedes Center for Medical Device Security

The Archimedes Center for Medical Device Security was established to help manufacturers and industry experts navigate the operational hazards of cybersecurity implementation and prepare them for future challenges of FDA requirements. Archimedes is an independent, pioneering center focused on the education and advancement of medical device security where key industry players come together for learning in a safe place.

CIS Controls: A Cybersecurity Blueprint to Prevent Cyber Attacks

Data breaches arguably constitute the gravest risk modern businesses face. According to an alarming report from the insurance company Hiscox, 70 percent of organizations are inadequately prepared for cyberattacks. This is particularly concerning in light of the multi-million, even billion dollar damages that can result from breaches. Already, far too many businesses have suffered devastating attacks; 45 percent of Hiscox's poll respondents admitted to suffering at least one cyberattack in the past year, with many suffering two or more attacks.

At the small to midsize business level, cyberattacks aren't merely annoying — they can spell certain doom for those already struggling to get by. Hence the need for robust security protocol. That's exactly what the Center for Internet Security provides with its Top 20 list of Critical Security Controls. While these controls have been in the making for well over a decade, they've recently gained greater prominence at the federal and state level — and among private entities. Below, we offer an in-depth overview of this critical security tool, as well as suggestions for implementation:

What Are the CIS Top 20 Critical Security Controls?

The Center for Internet Security maintains detailed guidelines outlining prioritized actions known as critical security controls (CSCs). The goal is to proactively and effectively address security threats, thereby minimizing the potential for future data breaches. While any organization can benefit from implementing these controls, they are particularly valuable for those that currently lack robust security protocol. The CSC provide an accessible means of implementing security features while paving the path to a fully-fleshed compliance framework.

A Brief History of the CIS Top 20 CSC

Before diving in to explore specific controls, it helps to gain a broader understanding of how they came to be. They were initially developed in response to a 2008 request from the Office of the Secretary of Defense, in which assistance was sought from the National Security Agency. At the time, the NSA best understood the nature of cyber attacks — and how to combat them. Then, the White House maintained the following cybersecurity mantra: "Offense must inform defense."

By the time the aforementioned request was made of the NSA, the agency had already compiled and refined a list of effective security controls dating back to the early 2000s. These controls were originally prompted by military requests. While the initial list was limited to official use, the NSA eventually agreed to share its compiled cybersecurity information in hopes of helping other government agencies improve their security protocol. 

Upon validation by the U.S. State Department, the CIS Top 20 was found to align closely with thousands of documented attacks suffered at the federal level. In an effort to address significant security weaknesses, the State Department made integration of the CIS standards a clear priority. The effort was an undeniable success; the State Department achieved an impressive 88 percent reduction in risk across tens of thousands of systems. As a result, the CIS standards quickly became the blueprint of choice not only for other federal organizations, but also in the private sector.

Today, the CIS Top 20 controls are maintained and updated by a vast team of volunteers, including experts from every segment of the cyber ecosystem. Expert volunteers include auditors, threat analysts, policy-makers, users, and more. They come from a wide array of sectors, including everything from transportation to defense. Their feedback ensures that the controls remain not only effective for protecting against a range of security threats, but also accessible, scalable, and easy to implement for a broad spectrum of businesses and organizations.

Defining the Basic Controls (Top 6)

Basic CIS Controls (click to enlarge)

The first six controls outlined by the CIS are often referred to as the 'basic controls.' While all of the outlined controls in the Top 20 are valuable, the basic CIS controls would ideally be implemented by all organizations seeking to ready themselves for future cyber attacks. Basic controls include the following:

#1 Inventory of Authorized and Unauthorized Hardware

It is imperative that organizations track all network devices — without a detailed inventory, it is virtually impossible to provide adequate protection. Successful application of the CIS first control is the foundation on which all other CIS controls rely. Putting this control into practice means identifying all relevant devices and maintaining a current inventory. This can be a huge undertaking; experts at CIS recommend using active scanning tools and other automated procedures.

#2 Inventory of Authorized and Unauthorized Software 

This control references the need for active management of all network software to ensure that only authorized software is installed — and to quickly detect and deal with any unauthorized software that finds its way into managed networks. Cybercriminals regard organizations with vulnerable software as easy targets. Organizations that identify all existing software and develop approved whitelists improve visibility and may even see considerable savings as they discard unnecessary programs.

#3 Continuous Vulnerability Management

From software updates and patches to threat bulletins, ample opportunities exist for attackers to take advantage of new vulnerabilities — even among systems initially designed to be secure. As new vulnerabilities are reported, would-be attackers race to exploit these gaps. Without continuous scanning, organizations risk falling behind — a huge threat given the common knowledge that time is of the essence in the event of an attack.

#4 Controlled Use of Administrative Privileges

Attackers commonly rely on administrative privileges to carry out harmful actions within targeted enterprises. While most employees are aware of such efforts, even notoriously vigilant individuals can fall victim to scams. Common examples include opening files from malicious sites, downloading problematic attachments, or even merely visiting websites capable of exploiting users. One of the easiest to implement — and yet most effective — controls available, this step involves separating administrative accounts from personal internet activity. Additionally, systems should be configured for log entries and alerts.

#5 Secure Configurations for Hardware and Software on Mobile Devices, Laptops, Workstations, and Servers 

Accessibility often plays a key role in modern software adoption — sometimes to the detriment of security. Default accounts and open services can leave organizations and their key players vulnerable. Hence, the need for this basic control, which is especially important in an increasingly BYOD (bring your own device) oriented workplace. Rigorous configuration management is essential if organizations are to keep attackers at bay.

#6 Maintenance, Monitoring, and Analysis of Audit Logs

While the cliche 'an ounce of prevention is worth a pound of cure' certainly applies to modern cybersecurity efforts, it's foolish to assume that breaches will never occur. Unfortunately, even organizations with exceptional security protocol sometimes fall victim to cyberattacks. When the worst-case scenario arrives, it's important to be prepared — not only in the interest of responding quickly and effectively, but also because a detailed understanding of one attack could potentially lead to new protocol to prevent future issues.

Foundational and Organizational CIS Controls (click to enlarge)

Foundational and Organizational Controls

The other control categories include foundational controls and organizational controls. These areas can help enterprises shore up their security after they have mastered the basic controls. Without mastery of the basics, however, these additional controls are not likely to serve as a viable security solution. Controls referred to as foundational or organizational include:

Foundational CIS Controls

7. Email and Web Browser Protections

8. Malware Defenses

9. Limitation and Control of Network Ports, Protocols, and Services

10. Data Recovery Capabilities

11. Secure Configuration for Network Devices, such as Firewalls, Routers, and Switches

12. Boundary Defenses

13. Data Protection

14. Controlled Access Based on the Need to Know

15. Wireless Access Control

16. Account Monitoring and Control

Organizational CIS Controls

17. Implement a Security Awareness and Training Program

18. Application Security Software

19. Incident Response and Management

20. Penetration Tests and Red Team Exercises

While there is clear value in all of the controls included on the Top 20 list, not all controls will prove accessible or applicable in all situations. Nearly all organizations, however, can benefit from implementing the controls identified as 'basic' by the CIS. 

Comparing CIS to Other Frameworks

The CIS Top 20 is just one of several security frameworks relied upon by government and private enterprises alike. Other top framework providers include National Institute of Standards and Technology (NIST) and Payment Card Industry (PCI) Security Standards Council. While these hold clear value, they can be difficult for small businesses to adopt. The CIS Top 20 serves as an essential stepping stone for those who may feel intimidated by other frameworks or otherwise incapable of implementing them. CIS controls can also be used alongside other frameworks; some companies even use CIS standards to prioritize alternate frameworks.

Mandating Controls

Presently, the CIS Top 20 serves as a standard of care for the state of California, where the Attorney General has partnered with CIS to help local businesses follow the protocol outlined in the state's Data Breach Report. 

While CIS became a key California framework in early 2016, recent changes in the state's privacy laws have further underscored the importance of adopting CIS controls. Signed into law in 2018 and expected to go into effect in 2020, the California Consumer Privacy Act now supplements the state's data breach legislation, granting consumers the right to sue if breaches of data occur and businesses do not provide appropriate protection via 'reasonable' procedures. While the state fails to adequately define 'reasonable' in the context of the Consumer Privacy Act, it is clear based on previous efforts with the CIS that implementation of, at minimum, the basic six CSCs is expected.

While California has clearly set the bar high in terms of mandated internet security protocol, other state and local government entities are beginning to follow suit with similar legislation. Michigan, in particular, has made great strides in terms of encouraging implementation of prominent security frameworks. The 2018 Performance Audit Report from the Michigan Office of the Auditor General referenced NIST Special Publication 800-53 as the bare minimum regarding security controls. 

In Ohio, the CIS controls are highlighted as one of a few available frameworks with which business cybersecurity practices would ideally conform. While the Ohio Attorney General maintains that the state's recent adoption of the Data Protection Act does not serve as a minimum cybersecurity standard, it does create a valuable incentive for the local business community to amp up security efforts.

In some states, CIS controls are not yet mandated — but they are used heavily by government entities. Success at this level may eventually lead to broader mandates for government organizations and private businesses alike. As CIS experts point out, top users of the CSCs include the states of Arizona, Colorado, and Idaho. In addition to making its mark at the state level, the CIS controls are increasingly utilized in major metropolitan areas, including Oklahoma City, San Diego, and Portland, among others. 

Success Stories

The Center for Internet Security offers numerous case studies that demonstrate the ability of various controls to protect against a broad range of security threats. For example, CIS highlights a top banking institution as relying on these controls for gap analysis. The banking institution's Chief Information Security Officer explains, "If we are not quite meeting the intent of a particular CIS Control, we can identify areas to focus on and improve." This valuable baseline has allowed the bank to maintain its status as one of the nation's most secure financial institutions.

While the value of the CIS controls is clear in banking and finance, many retailers have yet to adopt appropriate security measures to protect their customers and suppliers. One outdoor retailer, however, is leading the way by engaging in CIS analysis not just once, but twice every year. While the retailer also uses such familiar frameworks as NIST and PCI, a top security analyst for the company explains that the CIS controls provide, by far, the biggest bang for the buck by assisting in prioritization of other frameworks. "We see the real value is in the CIS controls because they are more user-friendly and are a practical, prioritized framework."

In another case study, CIS reveals that Corden Pharma has adopted the CSCs in hopes of meeting diverse security requirements for the company's range of customers. Manager of IT and Business Systems John Nord explains, "[Corden Pharma] needed a more standardized security program for our company to be able to provide to our customers. The CIS controls fit that need." Nord adds that, while initially daunting, implementation for the CSCs is not nearly as arduous as NIST framework adoption can be.

What Lies Ahead for the CIS Controls?

The CIS Top 20 are constantly evolving as new threats arise — and as new technologies become available for keeping these threats at bay. In early 2018, CIS released Version 7, which, as CIS executive Tony Sager explains, "sets the stage for future improvements in measurement, implementation, and alignment with other security frameworks." 

Although pleased thus far with the rollout of Version 7, companies are already looking ahead to Version 8. With new editions of the CSC released approximately once every three years, Version 8 can most likely be expected in 2021 or 2022. Until then, experts anticipate that states and municipalities will continue to follow California's lead in prioritizing these security measures and working closely with CIS to ensure the adoption of robust cybersecurity controls.

Next Steps for Adopting CIS Controls

Regardless of how the CIS controls evolve in the future, it is important to get a handle on security protocol today. The widely trusted security controls from the CIS can provide a valuable blueprint. Not sure where to start? Alpine Security's Enterprise Security Audit (ESA) service provides valuable insight into your organization's current security protocol. Closely aligned with the CIS Top 20, this comprehensive assessment offers a holistic approach involving effective controls for cyber defense. Reach out today to discover how Alpine Security can assist your organization in adopting and abiding by the Center for Internet Security's Top 20 Controls.

The Greatness of Being Uncomfortable

There is no denying that most people in this world want to avoid feeling uncomfortable as much as possible. The problem with this mindset is that those who fear discomfort and uncertainty are always going to be stuck in the same place. The best way to fail to succeed is to stay in one place for relative comfort.

Fear of failure ruins our lives

If you think about it, you will come to the conclusion that our biggest fear is not failure itself. What we truly fear is that other people will see us fail. This is a very common problem and it has no logic behind it. When you fail, you feel like others are going to ridicule you or laugh at you, but no one ever laughs at failure, they may feel bad for you, but they fear failure more than you do, they are completely frozen by fear and they never even try anything new because they would rather stay in their comfort zone.

Just imagine how ridiculous it would be for you to give up your goals and dreams because you are afraid of what others will say if you fail. The simple thought of this makes you cringe and it should be more than enough to motivate you to take action.

Good things will never happen to you

People are usually expecting god things to happen to them in life. They say “I’m a good person, I deserve good things, I know things will change” but the problem is that they are just standing on the road of life, while others are standing there with them waiting for good things to come to them. This is a huge problem for many people because they feel entitled to happiness and to success just because they are nice, or because they are honest and decent. Those are not qualities and virtues that give you a free pass to success. The sad truth is that no one has ever achieved success in life by thinking this way.

Imagine that you are standing on the road and you see a pot of gold a few miles away, but there are traps and pits all the way through. You are going to be waiting forever if you expect someone else to carry that gold to your location. Anyone that reaches that gold is going to keep it for themselves. They would be willing to share if you also walked towards that gold and took your chances with those traps.

The exact same thing happens with everything we want in life. If we don’t start heading in the direction of what we want, there is always going to be someone else willing to take the risk. The point here is that anything you want in like is going to require that you move forward and step out of your comfort zone.

Adding value to your journey

The greatest thing a person can do in life is to add value in their journey to the goals they set for themselves. If your goals include being a competitive and highly valuable asset to any modern business, you will find that cybersecurity training is going to be essential for that purpose. We have classes with official certifications in cybersecurity that will add thousands of dollars of value to your paychecks.

Step out of your comfort zone and invest in your future. Take action and move forward on the road to your dreams. Find out how to enroll in our awesome courses to add massive value to your journey to success!

Cyber Extortion: Ransomware vs Extortionware

Cyber crime is not just the plotline of bad TV movies. It's a $600 billion global business that accounts for 0.8% of the world's GDP. Anyone on the internet - and there are 3.2 billion of us - is a potential victim.

While phishing scams and identity theft are old news, the more serious threats to organizations are ransomware and extortionware. In fact at Europol's 2018 Internet Organized Crime Threat Assessment, the European crime prevention agency stated, "In a few short years, ransomware has become a staple attack tool for cybercriminals, rapidly accommodating aspects common to other successful malware such as affiliate programs and as-a-service business models." 

Ransomware's sister threats are a different form of cyber crime called cyber blackmail or cyber extortion. Blackmail doesn't necessarily involve sophisticated technology. But ransomware and cyber extortion typically do. While these two types of malware share common themes, they also differ in key respects. What's the difference between ransomware and extortionware? And what can you do to prevent your company from becoming a victim of cyber crime?

What is ransomware?

Ransomware - Bad Rabbit Example

Ransomware is a type of malware that locks a computer system down until the victim pays the extortioner for the key code to unlock the device. It's an obvious form of cyber crime. While some internet criminals try to fly below the surface, stealing data without you realizing it, ransomware is overt. The hacker is telling you in plain language, "We've encrypted your system. The key is in our lockbox. If you want your data or computer to work again, you'll pay us." Usually, the victim has to pay in Bitcoin or through another hard-to-trace method. 

It's the most common form of cybercrime and perhaps the most effective. Who wouldn't pay a few hundred dollars to a criminal instead of a few thousand to a break-fix IT service, which might not be able to unlock the system anyway? The trouble is, there's nothing to stop the criminal from coming back again ... and again. Ransomware can cost a lot more than the initial payment; it also costs time, productivity, and reputation.

Examples of ransomware

The first ransomware appeared in 1989. Hackers mailed floppy disks (remember those?) to unsuspecting victims who inadvertently installed the malware on their computers. To get the key, the victim had to mail either $189 or $378 to Panama. Today's cyber criminals operate a similar racket but with greater technical prowess. The WannaCry ransomware attack in 2017, which was linked to the North Korean government, infected about 200,000 computers in 150 countries. 

Bad Rabbit was a ransomware that affected European and Russian users. Unlike WannaCry, which spread randomly, Bad Rabbit targeted its victims, which included the Odessa airport and Kiev's mass transit system.

A relatively new and irritating form of ransomware is called "police-themed demands." In this, the criminals claim to be the police who are shutting down your computer due to its use in a terrorist act or a child pornography ring. By paying a few hundred dollars in fines, the "police" claim, you can unlock your computer. Of course, the whole thing is bogus, and you're the victim of ransomware.

“FBI” Extortionware Example (click to enlarge)

What is cyber extortion?

Extortion is simply demanding a good, service, or payment to prevent violence or destruction of property. Some cyber criminals get ahead of the competition by demanding money before they hurt you. Much like the mobsters of yore who wanted payment for "security services," these cyber extortionists basically tell you to pay them or they will lock up your system. According to Insureon, "Cyber extortionists may threaten to harm you, your reputation, or your property if you do not comply with their demands. Cyber extortion can take many forms." For instance, a hacker may send malicious traffic using a bot, overwhelm the site with more requests than it can handle, or simply shut it down. 

What is cyber blackmail?

Blackmailers employ the psychology of fear and shame to motivate victims. In these cases, the cyber criminal will claim that he's been inside your computer, snapped screenshots of embarrassing photos or searches, and recorded sketchy video using your camera. He will threaten to send the screenshots and pics to everyone in your address book or post them on social media unless you pay up. It's basically cyber blackmail.

"Cyber-blackmail is the act of threatening to share information about a person to the public, their friends or family, unless a demand is met or money is paid," according to the BBC. It is a particularly insidious kind of crime. In fact, some hackers are even targeting children.

Cyber blackmail doesn't only affect individuals, but it also harms companies. Hackers claim they have stolen sensitive data from a company's storage and threaten to expose it unless paid off. Many times, the hacker has no evidence at all and is simply bluffing. The really dangerous part is that emails from hackers often reveal they have secured your password, and while that may be all they have, it's often enough to do some serious damage.

Cyber Extortion Example (click to enlarge)

Am I a victim of ransomware, extortionware, or cyber blackmail?

Because these are forthright kinds of crime, you'll likely know if you're a victim. The hackers will tell you when they demand the ransom, threaten you for money, or want blackmail payoff. One handy way to assess the credibility of an extortion scheme or blackmail threat is at HaveIBeenPwned. This site lets you see what breaches your email address has been associated with, so you can determine if the password that is supposedly compromised and used for the cyber extortion is simply from a previous breach or if the hacker really has broken through your security system. 

In no case should you pay off a cybercriminal even if you think they have something on you. A qualified managed services provider or IT professional can help protect you against ransomware and can reverse most of it easily. Cyber blackmailers almost never have the goods on you they say they have, and extortionists are probably no match for your IT managed service provider or cybersecurity squad. Report cybercriminals to the police. Your company doesn't need to be a victim.

How penetration testing and training can prevent cyber crime

At Alpine Security, our penetration testing services and training can help prevent both ransomware and extortionware. We can also help you know what to do if a hacker uses ransomware to lock down your system. Penetration testing can help identify gaps in systems and networks, and our user awareness training, including our phishing services, can identify gaps in user training. To learn more about protecting your organization from ransomware and cyber extortion, contact us today. 

The Ultimate Guide to Cyber Threat Maps

A cyber threat map, also known as a cyber attack map, is a real-time map of the computer security attacks that are going on at any given time. One of the most famous was released by the company Norse and went so viral, even among non-hackers, that it got its own story in Newsweek in 2015.    

The map itself looks like a global game of laser tag. Beams of light, represented by different colors, shoot across a darkened screen showing where an attack comes from and where it is going. When it first caught the public eye, captivated audiences watched hackers wage cyber-war across hundreds and thousands of miles. 

How Does A Cyber Threat Map Work?

If cyber attacks are sneaky mice – or, more appropriately, giant rats – then cyber attack maps work like the mousetrap. 

Norse, for example, maintained a global threat intelligence network of more than 8 million sensors and “honeypots” in 47 countries across the world. These tools impersonated thousands of applications and devices that are common targets of hackers. 

When a hacker hits a Norse sensor, the hacker believed that it had breached a system. Instead, Norse collected information about the hacker's toolkit, including his or her IP address. This information then manifested as data on the cyber attack map.

This model has continued past the demise of Norse to power live maps like Cyberthreat, ThreatCloud, and Fortinet. Some companies claim that these are real-time cyber attacks, but most are more like selections of recent attacks.

What Are They Good For?

First of all, anyone can look at a cyber threat map and understand something about current attacks. For laypeople, it might just be an understanding of how prevalent these attacks are. 

For pros, a cyber threat map can provide data that can lead to the discovery of a potential new attack. Even though these maps don't really show cyber attacks in real-time, they give you information, and a cybersecurity pro can never have too much of that.

The Best of the Best in Cyber Threat Maps

Not all cyber threat maps are created equal. Some are great eye candy for pen test companies, but others do offer good functionality. A few even let you manipulate the map to focus on its most useful information.

#1 Cyberthreat by Kaspersky Lab

Kaspersky's cyber threat map may be the best in the business in terms of interactivity and visual appeal. It's also mobile optimized so you can show it to clients or colleagues from anywhere. And if you want particular data, the site offers you the option to find it.

Narrowing It Down

Kaspersky's Cyberthreat also offers useful statistics about current cyber attacks.  Through its website, you can learn about:

• Vulnerabilities identified 

• Most widespread threats 

• Botnet activity trends 

• Prevalence of infected mail and spam

You can find this data for a specific country or look at it on a worldwide level. 

The Widget

Kaspersky Lab has figured out that cybersecurity professionals often use threat maps to visually represent their work to clients.  To make it easier, they have created a Cyberthreat map widget, which you can embed in your website, as shown below, so that clients can see a representation of the latest threats.

Fortinet Threat Map (click to enlarge)

#2 Fortinet by Fortiguard

This one isn't quite as customizable as Cyberthreat, but it's informative enough to warrant runner-up honors. The main map color codes its attacks by severity to make it easier to focus on the most critical situations, coded in red. The details of these attacks scroll quickly along the bottom of the map, where the viewer can identify the threat type and location.

Fortiguard clients have a bit more freedom in terms of map design. By placing a tool known as the ThreatGate, users can monitor the threats approaching a particular location.

Check Point Threatcloud (click to enlarge)

The Best of the Rest

In addition to Fortinet and Cyberthreat, there are a few other cyber attack maps that repeatedly appear on “Best of” lists. They are:

• #3 ThreatCloud by Check Point      

• #4 FireEye    

• #5 Trend Micro    

• #6 Deteque     

• #7 Arbor Networks     

• #8 Akamai

Each of these maps presents its data in a slightly different format and with a slightly different emphasis. Take a look, try them out, and find out which one will work best for you.

The Take-Away 

If you find these maps intriguing,  know that they're not enough to make you an expert. You'll need to learn more about cybersecurity and take some training, like the penetration testing courses available through Alpine Security.  There are tests available for all levels of pen testers – check it out today.  

Hacking Microdrones for Lethal Gain

An assassin's most powerful new weapon could be resting on the shelf at your neighborhood Best Buy. 

Warfare is no longer about dumping thousands of men in a field and shooting at each other. Today, non-governmental forces are packing explosives onto commercially available drones and flying them over crowded areas. This past August, a dissident organization called Soldiers in T-Shirts attempted to assassinate Venezuelan President Nicolás Maduro using a drone. While this attempt was unsuccessful, it marked the first time -- but almost certainly not the last -- that a paramilitary organization tried to assassinate a sitting head of state with a drone.

As early as 2012, Business Insider was sounding an alarm about hacking microdrones for lethal gain. The magazine posited the question: What if mosquito-sized drones could deliver deadly viruses to targeted humans? 

It sounds like bioengineered- and- virological science fiction with a macabre twist. But as President Maduro can attest, we're not far away from it being a reality. Swarms of microdrones armed with viruses, poisons, or other technological weaponry could pose a serious threat to national security and human life in conflict zones. In January, 2018, Forbes magazine interviewed Randall Nichols, a professor of unmanned systems and cybersecurity at Kansas State University. "It is clear to me that increased global investments and research in advanced UAS cyber-related technologies are yielding clever, advanced offensive cyberweapons as payloads," Nichols said. "Combine this with investments and growth in automation capabilities and add ability/threat to use swarming and team efforts with multiple drones against U.S. targets, we have a potential Black Swan event."

The key to global power may rest on a fleet of technological objects the size of an insect but with extraordinary power for good or evil, depending upon whose hands it rests in.

What is a microdrone?

Microdrones have no official specifications, but Wikipedia does a good job defining them. "A micro air vehicle (MAV), or micro aerial vehicle, is a class of miniature UAVs that has a size restriction and may be autonomous. Modern craft can be as small as 5 centimeters." That's smaller than a golf tee or about 3.5 times as long as an aspirin. Basically, these devices are just like the Tracker Jackers your kids read about in The Hunger Games. Only real.

Like the deadly jackers of fiction, microdrones actually emerged from the study of insects. Scientists at Stanford University and at the École Polytechnique Fédérale de Lausanne received funding from the Swiss government to make an electronic, human-controlled robotic wasp. Called the FlyCroTug, this microdrone was built to carry proportionally large and heavy items across long distances the way wasps carry home too-large prey in the wild. According to its write-up in Popular Mechanics, the FlyCroTug possesses three carrying features -- winches, gecko grippers, and microspines -- the same ones a wasp uses. These features give the FlyCroTug the ability to do more than fly around and observe. They also also the microdrone to walk, climb, grasp, and build.

How can microdrones be weaponized?

In 2008, the Air Force Research Laboratory began fashioning the world's first micro-sized killer robot. Larger drones may sometimes misidentify targets, cannot respond to a last-minute maneuver by a target, and can injure or kill nearby civilians so the Air Force decided to create something more focused. It came up with a wasp-sized microdrone that weighed less than one pound and could be controlled at a distance of up to three miles. It only took $1.75 million of American taxpayers' dollars to create. The Chinese military development technologists are working on similar technology.

Microdrones can carry far more weight than they appear able to, 40x as much. Plus, they can work together in swarms, which means a few dozen of them can carry the same virus, poison, or explosive to the same population, exponentially increasing the viability of their threat. Not only that, but they can also work together to accomplish tasks such as opening doors.

Perhaps most disturbing is the concept of hacking microdrones for lethal gain. In this scenario, a terrorist, rogue state actor, or wired-up teen techno-whiz in his mom's basement would not need to build his or her own insect-sized weapons. They could just turn ours against us. By hijacking our drones, outside forces gain control over powerful weaponry. As James O'Malley put it in his article in Engineering and Technology, "There are now thousands of devices out in the wild running insecure firmware. But instead of being used to attack cyberspace, they can be let loose on the physical world."

A microdrone's ability to target terrorist leaders and others who pose a threat to geopolitical security may prove worth the investment. The scary part, of course, is what happens when the other side can create the same technology -- and use it without concern for ethics or the safety of the larger population. 

Is the microdrone threat legit or make believe? 

Are deadly microdrones really something to fear or are they science fiction from the minds of popular techno-whizzes who've read too many Margaret Atwood novels? 

That depends on what you're talking about when you say "microdrone threat." If you are asking about swarms of self-controlled robots that make their own ethical decisions about who lives and dies in a dystopian society, then we're probably a long, long way from that. On the other hand, a microdrone in the hands of a non-state actor like Osama bin Laden or a powerful and unpredictable rogue state like North Korea could wage acts of terror and destabilization on society.

Of course, microdrones carry enormous power for good, too. They could provide significant first-response aid to people experiencing natural catastrophes, potentially saving lives and property due to their ability to mobilize, move fast, and go into tiny spaces where human personnel cannot. On the other hand, credible scientists, academics, and public policy makers also express concern about what microdrones can do and the magnitude at which they can do it.

Both Elon Musk and the late Stephen Hawking called for a ban on developing autonomous weapons, and they specifically mentioned "armed quadcopters." The Future of Life Institute, a Musk-backed think tank, published UC-Berkley professor Stuart Russel's movie Slaughterbots, which showed the power of an army of micro-sized robots could have against humans. The movie was shown at the United Nations Convention on Conventional Weapons in Geneva in 2017, too.

According to a March 2018 article from the National Academies of Sciences, Engineering, and Medicine, "The emergence of inexpensive small unmanned aircraft systems (sUASs) that operate without a human pilot, commonly known as drones, has led to adversarial groups threatening deployed U.S. forces, especially infantry units. Although the U.S. Army and the U.S. Department of Defense (DOD) are developing tactics and systems to counter single sUASs, a new report by the National Academies of Sciences, Engineering, and Medicine emphasizes the need for developing countermeasures against multiple sUASs -- organized in coordinated groups, swarms, and collaborative groups -- which could be used much sooner than the Army anticipates."

Clearly, serious people take the microdone threat as a genuine concern.

What is the magnitude of a lethal microdrone's potential damage?

In 2015, Israel was already selling a seven-pound drone that could carry a one-pound warhead. Tinier microdrones of the future could feasibly carry equally powerful explosives.

Much more threatening, perhaps, than bombs -- which amount to little more than ultra-sophisticated TNT, after all -- are diseases. Could a swarm of microdrones infect a city with dengue fever, which might not prove fatal but could certainly cause chaos and damage, or even something worse such as plague? Entire cities could potentially be at risk.

By combining the delivery power of a drone with the specificity of a virus, bad actors could expose select targets to a disease specifically designed to attack their genetic makeup. For instance, if a terrorist wanted to kill the U.S. president, he would need only the president's DNA map, a trained virologist, a microdrone, and a copy of the president's schedule to do it.

The stuff of futuristic political thriller shows on Netflix? Maybe. Or quite possibly a reality that nation-states will increasingly need to guard against. 

How much does a microdrone cost to buy or make?

Small size drones are typically cheaper than their older and larger counterparts. NewEgg says you can buy a Micro Drone 3.0 kit, the quadcopter, gimbal, transmitter, and Google Cardboard VR headset for $215. The replacement batteries cost $16, and the device is tough enough for a child to use. Many small drones go for between $20 and $150 on Amazon. Typically, these drones sell to kids and to beginning enthusiasts who want to fly them for fun around the neighborhood, not generally to notorious terrorists who want to assassinate Western politicians. Still, it was a very ordinary drone that Soldiers in T-Shirts used in its assassination attempt on Venezuelan President Maduro. 

And if you're a real DIY-er, you can even build your own microdrone if you want. YouTube and Google offer step-by-step tutorials for crafting a low-cost, fun-to-fly microdrone. For many hobbyists, building your own drone at home is a great way to save money. For terrorists, it's not yet a viable way to construct high-tech weaponry, but it soon could be. Remember that the 9/11 terrorists hijacked entire airliners with nothing more than boxcutters. Super technological skills are not required to build weapons of mass destruction.

In short, you can't buy a wasp-sized robot killer at Target or Best Buy. But could you buy what you need to make it happen? 

 Can you weaponize a microdrone yourself? 

Since microdrones are easy to come by, are they equally easy to weaponize? Is there a DIY way to make a killer robot at your kitchen table? How vulnerable are these devices to hacking? Can a teenage computer enthusiast take over one with his controller and direct it where he wants, hijacking it for his own use?

People have used drones for all kinds of nefarious purposes, from prison breaks to drug smuggling. Basically, you can weaponize anything. In 2013, Adam Piore described his drone weaponization journey in a humorous article for Popular Mechanics. "In this Wild West age of unregulated personal flight, even a rank amateur like myself can transform a toy into a hazard, an action that should be—and probably soon will be—illegal."

It is neither expensive nor complicated to build, weaponize, or hijack a microdrone. That isn't to say you can sit in your garage and craft a military-grade weapon with supplies from Target and a YouTube tutorial. You could, however, build a device that would do a lot of damage. And after all, does a terrorist need a military-grade weapon to assassinate a single target? Or to deliver a deadly, contagious virus to a few folks in a massive city?

While no one wants to sound alarmist at the threat microdrones may pose, the fact is that drone technology opens up the possibility of new kinds of warfare, terror, crime, and security. By blending cyber technology with physical activity, drones are a logical next step both for those who want to threaten others and those who want to protect them.

As part of our commitment to cybersecurity, Alpine Security offers extensive penetration testing services for companies and government agencies based on our deep experience testing medical devices, blockchain, aircraft, embedded systems, and complex systems. Our highly trained, certified, vetted, and experienced team uses a proven process and effective penetration testing methodology to provide clients with actionable, easy-to-understand reports. Contact us to learn more about how Alpine Security can help you protect the data, systems, and people you care about -- from data leaks to armed killer robots.