Using AI to advance the Digital Forensics Process
By Jared McCann on December 8, 2023
Introduction
As AI increases its prevalence for public and general usage, novel use cases are becoming more accessible. General model tools such as ChatGPT and Stable Diffusion have forced AI into the minds of the general populace but only highlight narrow use cases. More specialized and advanced models will likely see more significant use. One of these use cases is likely to be digital forensics. While still in its infancy, the growth of AI over the next few years is expected to explode. With it, there will be a need to integrate it with forensics tools to streamline and advance the current process.
Normal 0 false false false EN-US X-NONE X-NONE /* Style Definitions */ table.MsoNormalTable {mso-style-name:"Table Normal"; mso-tstyle-rowband-size:0; mso-tstyle-colband-size:0; mso-style-noshow:yes; mso-style-priority:99; mso-style-parent:""; mso-padding-alt:0in 5.4pt 0in 5.4pt; mso-para-margin:0in; line-height:115%; mso-pagination:widow-orphan; font-size:11.0pt; font-family:"Arial",sans-serif; mso-ansi-language:EN;}
The state of AI today
AI technology is now maturing at a rapid pace. With increased consumer appeal, companies are all vying for a stake in the growing consumer AI marketplace. As of September 2023, OpenAI, the creator of ChatGPT, has sought a new valuation of up to $90 billion. [1] This is just a tiny portion of the overall current market value. In 2021, the overall AI market was valued at over $100 billion and is estimated to grow to $1.75 trillion by 2030. [2] Now, it’s likely that that value is significantly higher. The US Government Accountability Office (GAO) notes that with the rapid growth spurred heavily by the popularity of OpenAI’s products, it will be essential to implement proper oversight and ethical standards. [3]
Normal 0 false false false EN-US X-NONE X-NONE /* Style Definitions */ table.MsoNormalTable {mso-style-name:"Table Normal"; mso-tstyle-rowband-size:0; mso-tstyle-colband-size:0; mso-style-noshow:yes; mso-style-priority:99; mso-style-parent:""; mso-padding-alt:0in 5.4pt 0in 5.4pt; mso-para-margin:0in; line-height:115%; mso-pagination:widow-orphan; font-size:11.0pt; font-family:"Arial",sans-serif; mso-ansi-language:EN;}
Current Uses in Digital Forensics
Within the field of digital forensics, it is easy to see the many uses AI can have. We can streamline the analysis process, find artifacts, and even generate reports faster than usual by utilizing AI. The impact of this is clear as well. By generating a report and finishing a case study rapidly, security teams can better decide how to protect their systems on a shorter timeline than ever before. AI can also rapidly analyze enormous amounts of data in a significantly shorter time than any human investigator. [4] This means that larger cases that typically take months or even years to complete can be compressed into possibly only a few weeks to complete. This doesn’t mean that AI isn’t without its flaws, either. AI inherently relies on having data first to understand data. Without a large enough dataset for training, an AI model can give false positives and inaccurate predictions. Even with a large dataset, these issues can still be present, so conclusions still need to be verified by a human investigator. [5]
Normal 0 false false false EN-US X-NONE X-NONE /* Style Definitions */ table.MsoNormalTable {mso-style-name:"Table Normal"; mso-tstyle-rowband-size:0; mso-tstyle-colband-size:0; mso-style-noshow:yes; mso-style-priority:99; mso-style-parent:""; mso-padding-alt:0in 5.4pt 0in 5.4pt; mso-para-margin:0in; line-height:115%; mso-pagination:widow-orphan; font-size:11.0pt; font-family:"Arial",sans-serif; mso-ansi-language:EN;}
The Future
All AI and machine learning (ML) models are, at their heart, pattern analysis models. They look for key features and use those features to make decisions. In the future, these tools could be used to analyze images and detect patterns human investigators missed. [6] More advanced models, such as deep learning models, could better predict what might have happened without a human first having to feed it the data needed. In the future as well, we could see so-called “small data” models, which require significantly less data to train to create an accurate prediction. [7] All of these could be here tomorrow or in 50 years. Only time will tell where we will go with AI next.
Normal 0 false false false EN-US X-NONE X-NONE /* Style Definitions */ table.MsoNormalTable {mso-style-name:"Table Normal"; mso-tstyle-rowband-size:0; mso-tstyle-colband-size:0; mso-style-noshow:yes; mso-style-priority:99; mso-style-parent:""; mso-padding-alt:0in 5.4pt 0in 5.4pt; mso-para-margin:0in; line-height:115%; mso-pagination:widow-orphan; font-size:11.0pt; font-family:"Arial",sans-serif; mso-ansi-language:EN;}
Conclusion
AI has sparked many debates on how to regulate and implement it ethically and safely. However, regardless of what many think, it is a key piece of our daily lives. It touches almost everything we interact with online and impacts us in ways we don’t even see. For the digital forensics field, harnessing the power of current and future AI models will allow teams to better respond to threats and attacks. It will be important for companies and teams to embrace the usage of AI to stay ahead of threats and to respond.
References
[1] Deepa Seetharaman, “OpenAI Seeks New Valuation of Up to $90 Billion in Sale of Existing Shares,” Sept. 26, 2023 OpenAI Seeks New Valuation of Up to $90 Billion in Sale of Existing Shares – WSJ
[2] Burger Thormundsson, “Artificial intelligence (AI) market size worldwide in 2021 with a forecast until 2030,” Oct. 6, 2023 Artificial Intelligence market size 2030 | Statista
[3] Government Accountability Office, “Artificial Intelligence’s Use and Rapid Growth Highlight Its Possibilities and Perils,” Sept. 6, 2023 Artificial Intelligence’s Use and Rapid Growth Highlight Its Possibilities and Perils | U.S. GAO
[4] Tim Rollins, “The Use Of Artificial Intelligence In Digital Forensics,” Oct. 17, 2023The Use Of Artificial Intelligence In Digital Forensics – Exterro
[5] Jason Silva, “THE FUTURE IS NOW — THE IMPACT OF AI ON DIGITAL FORENSICS” The Future Is Now — The Impact of AI on Digital Forensics – Cornerstone Discovery
[6] Christopher Massimine, “The Role of AI in Forensic Science,” Apr. 10, 2023 The Role of AI in Forensic Science | by Christopher Massimine | Predict | Medium
[7] IBM, “The new AI innovation equation” What’s next for AI – Innovation equation (ibm.com)