Conducting Trustworthy Research with ChatGPT

Codecademy Team
Use ChatGPT to help conduct trustworthy research

Introduction

Are you new to your career field, or job role, or trying to become a subject matter expert? Maybe you’re already an advanced researcher trying to identify solid resources in your field of choice. Regardless of the circumstance, ChatGPT, an AI chatbot developed by OpenAI, can help you conduct research or learn more about a topic you’re interested in. In this tutorial, we will show you how to use ChatGPT as a tool to conduct research in the cyberspace domain.

Getting Started

To get started, we will assume the role of a junior cybersecurity analyst interested in learning more about cybersecurity risks and how often they occur. We will begin by setting the context for ChatGPT to learn more about cyberspace. By setting the context, we ask ChatGPT to draw upon relevant cybersecurity-related data in its training set when it responds to our queries.

Prompt:

You are an expert in the Cyberspace domain.

Response:

I can certainly help you with information and advice related to the cyberspace domain. Whether you have questions about cybersecurity, internet technologies, online privacy, digital communication, or any other related topics, feel free to ask, and I'll do my best to provide accurate and relevant information. Just keep in mind that my knowledge is based on information available up until September 2021, so there might be more recent developments that I'm not aware of.

Prompt:

I'd like to conduct research in the Cyberspace domain.

Response:

That's great! Conducting research in the cyberspace domain can be a fascinating and important endeavor. The field of cyberspace encompasses various aspects of digital technology, computer networks, information security, and the interactions between humans and digital systems. To get started with your research, here are some steps you can follow:

Identify your research area: Cyberspace is a vast domain, so it's essential to narrow down your focus to a specific area of interest. For example, you could explore topics like cybersecurity, data privacy, network architecture, digital forensics, social implications of technology, or emerging technologies like artificial intelligence and blockchain.

Review existing literature: Familiarize yourself with the current state of knowledge in your chosen research area. Conduct a thorough literature review to understand the existing theories, research findings, and gaps in knowledge. This will help you refine your research questions and develop a strong foundation for your work.

Define your research questions: Based on your literature review, identify specific research questions or hypotheses that you want to address. Ensure that your research questions are clear, specific, and aligned with your research objectives. This will provide a focus for your research and guide your investigation.

Develop a research methodology: Determine the most appropriate research methodology to address your research questions. Depending on your area of study, this could involve quantitative analysis, qualitative research, experimental studies, case studies, or a combination of methods. Plan your data collection methods, such as surveys, interviews, observations, or data analysis techniques.

Gather and analyze data: Implement your chosen research methodology to collect relevant data. This could involve conducting experiments, analyzing existing datasets, or gathering primary data through surveys or interviews. Apply appropriate data analysis techniques to derive meaningful insights and draw conclusions from your findings.

Interpret and discuss your results: Analyze and interpret your research findings in the context of your research questions and objectives. Discuss the implications of your results, highlighting their significance, limitations, and potential applications. Compare your findings with existing literature and contribute to the existing knowledge in the field.

Communicate your research: Share your research findings with the academic and professional communities. Consider publishing your work in conferences or reputable journals in the field of cyberspace research. Present your research at conferences or seminars to receive feedback and engage in scholarly discussions. You can also create educational content or contribute to policy discussions related to your research area.

Remember to maintain ethical considerations throughout your research process, especially when dealing with sensitive data or potential security risks. It's also essential to stay updated with the latest developments and emerging trends in cyberspace to ensure your research's relevance and impact.

Good luck with your research in the cyberspace domain!

Conducting Research

As cybersecurity analysts, we might want to learn more about the risks, vulnerabilities, and threats in the cyberspace domain. As a junior cybersecurity analyst, we may not have a thorough understanding to know about specific risks, vulnerabilities, or threats. Let’s ask ChatGPT about the top five cybersecurity risks.

Prompt:

What are the top 5 cybersecurity risks?

Response:

Cybersecurity risks evolve constantly, but here are five common and significant risks that organizations and individuals face:

Phishing Attacks: Phishing is a prevalent form of social engineering where attackers masquerade as trustworthy entities to trick individuals into revealing sensitive information like passwords, credit card details, or personal data. These attacks are often carried out through deceptive emails, text messages, or malicious websites.

Malware Infections: Malware refers to malicious software designed to harm systems or steal information. It includes viruses, ransomware, spyware, and trojans. Malware can infect computers and networks through various means, such as downloading malicious files, visiting compromised websites, or clicking on malicious links.

Data Breaches: Data breaches involve unauthorized access to sensitive information, resulting in its exposure or theft. Cybercriminals target databases, network systems, or cloud services to gain access to personal, financial, or proprietary data. Breaches can lead to financial loss, reputational damage, and regulatory non-compliance.

Insider Threats: Insider threats arise from individuals within an organization who intentionally or accidentally misuse their access privileges. This can involve employees leaking confidential information, deliberately causing damage, or falling victim to social engineering attacks. Insider threats can be challenging to detect and mitigate.

Advanced Persistent Threats (APTs): APTs are sophisticated, targeted attacks carried out by skilled threat actors over an extended period. APTs often involve multiple stages, including reconnaissance, infiltration, and data exfiltration. They can target government entities, organizations, or critical infrastructure, aiming to steal sensitive information or disrupt operations.

It's important to note that these are just a few examples of cybersecurity risks, and the landscape is continuously evolving. Other notable risks include distributed denial-of-service (DDoS) attacks, zero-day vulnerabilities, supply chain attacks, and emerging technologies such as Internet of Things (IoT) vulnerabilities. Organizations and individuals should adopt comprehensive cybersecurity measures, including education, robust security practices, and up-to-date defenses, to mitigate these risks effectively.

Although ChatGPT can provide information for us, it is not capable of critical thinking. As we consider these risks, let’s think about how many attacks occur from the provided categories. We will prompt ChatGPT to identify the percentage of attacks relative to the categories above.

Prompt:

Are there statistics showing the percentage of attacks relative to the cybersecurity risks above?

Response:

As an AI language model, I don't have real-time access to current statistics. However, I can provide you with some general information on the prevalence of cybersecurity risks based on historical data and industry reports. It's important to note that the landscape of cyber threats is dynamic, and the prominence of specific risks may vary over time. Here is a broad overview of the percentage of attacks relative to the cybersecurity risks mentioned earlier:

Phishing Attacks: Phishing attacks have remained a significant threat over the years. According to the 2021 Verizon Data Breach Investigations Report, phishing attacks accounted for approximately 36% of data breaches. The Anti-Phishing Working Group (APWG) reported a significant increase in phishing attacks in recent years.

Malware Infections: Malware infections continue to be a prevalent cybersecurity risk. The 2020 Internet Security Threat Report by Symantec reported that malware attacks accounted for 27% of observed cybersecurity incidents. Ransomware attacks, a specific type of malware, have seen a notable rise in recent years, targeting both individuals and organizations.

Data Breaches: Data breaches have affected numerous organizations and individuals globally. The number of reported data breaches varies each year, with some high-profile breaches impacting millions of individuals. The Identity Theft Resource Center's Data Breach Report stated that there were 1,108 reported data breaches in the United States alone in 2020.

Insider Threats: Insider threats can be challenging to quantify accurately due to their diverse nature. The 2020 Insider Threat Report by Cybersecurity Insiders revealed that 68% of organizations surveyed were concerned about insider attacks. However, specific statistics on the percentage of attacks attributed solely to insider threats may be harder to find.

Advanced Persistent Threats (APTs): APTs are highly targeted and sophisticated attacks, often associated with nation-state actors or advanced cybercriminal groups. It is difficult to provide precise statistics on the percentage of APT attacks compared to other threats due to their covert nature. APT attacks are typically focused on specific industries, government entities, or high-value targets.

It's worth noting that the prominence of different types of attacks and risks can vary based on factors such as industry sector, geographical location, and evolving threat landscapes. Organizations and individuals should stay updated with the latest cybersecurity reports and research to gain a better understanding of the current trends and risks in the field.

Validating ChatGPT’s Research

ChatGPT can provide an invalid response, also known as a hallucination. A hallucination in AI refers to the generation of outputs that are either factually incorrect or unrelated to the given context. It is your responsibility to ensure that ChatGPT responses are valid.

We will prompt ChatGPT to provide reliable sources that outline the threats outlined above, ensuring that the answers relate to a junior analyst.

Prompt:

Can you provide a reliable resource outlining the cybersecurity threats and mitigations outlined above for a beginner?

Response:

Certainly! Here are some beginner-friendly resources that provide information on cybersecurity threats and mitigation strategies:

"Cybersecurity 101" by the National Cyber Security Alliance (NCSA):

This resource provides an overview of essential cybersecurity concepts, common threats, and practical tips for individuals and small businesses. It covers topics such as phishing, malware, data protection, and incident response.
Resource link: https://staysafeonline.org/cybersecurity-101/
"Beginner's Guide to Cybersecurity" by the United States Federal Trade Commission (FTC):

This guide offers easy-to-understand explanations of various cybersecurity threats, including identity theft, ransomware, and phishing. It also provides practical steps and resources to help individuals and businesses protect themselves online.
Resource link: https://www.consumer.ftc.gov/articles/0009-beginners-guide-cybersecurity
"Securing Your Digital Life" by the Electronic Frontier Foundation (EFF):

This guide provides practical advice for protecting your digital privacy and security. It covers topics such as secure messaging, password management, encryption, and protecting your devices from malware.
Resource link: https://ssd.eff.org/en/module-categories/digital-security-basics
"StaySafeOnline" by the National Cyber Security Centre (UK):

This website offers easy-to-understand guidance on a wide range of cybersecurity topics. It covers threats like phishing, malware, and social engineering, along with practical tips for securing your devices and protecting your online identity.
Resource link: https://www.ncsc.gov.uk/collection/staysafeonline
"Cybersecurity Essentials" by Cisco Networking Academy:

This online course provides an introduction to cybersecurity for beginners. It covers fundamental concepts, threats, and mitigation strategies. The course includes interactive modules, quizzes, and hands-on labs to enhance your learning experience.
Resource link: https://www.netacad.com/courses/cybersecurity/cybersecurity-essentials

These resources should give you a good starting point in understanding cybersecurity threats and implementing basic mitigation strategies. Remember to continue exploring and learning from reputable sources as you progress in your cybersecurity journey.

We can click through these references and validate the information ChatGPT is providing. Furthermore, we could also pair this with other sources of information or Generative AI tools. Some tools include ResearchRabbit, Google Scholar, and Scite.ai.

Conclusion

Great work! let’s summarize what we just learned.

  1. We used ChatGPT to conduct research in the cyberspace domain.
  2. We narrowed the scope of research by telling ChatGPT to focus on cyberspace.
  3. We discussed some of the top risks associated with cyberspace and ensured they were valid by requesting sources of information from ChatGPT to avoid hallucinations.

As you conduct research in your areas of interest, be sure to confirm the information ChatGPT generates.

To see what else you can do with ChatGPT (or generative AI in general), checkout some of the articles located here: AI Articles.