Abstract—We conduct the first large-scale user study examining how users interact with an AI Code assistant to solve a
variety of security related tasks across different programming
languages. Overall, we find that participants who had access
to an AI assistant based on OpenAI’s codex-davinci-002
model wrote significantly less secure code than those without
access. Additionally, participants with access to an AI assistant
were more likely to believe they wrote secure code than those
without access to the AI assistant. Furthermore, we find that
participants who trusted the AI less and engaged more with
the language and format of their prompts (e.g. re-phrasing,
adjusting temperature) provided code with fewer security
vulnerabilities. Finally, in order to better inform the design
of future AI-based Code assistants, we provide an in-depth
analysis of participants’ language and interaction behavior, as
well as release our user interface as an instrument to conduct
similar studies in the future.
more »
« less
This content will become publicly available on February 27, 2025
Assessing the Effectiveness and Security Implications of AI Code Generators
Students, especially those outside the field of cybersecurity, are increasingly turning to Large Language Model (LLM)-based generative AI tools for coding assistance. These AI code generators provide valuable support to developers by generating code based on provided input and instructions. However, the quality and accuracy of the generated code can vary, depending on factors such as task complexity, the clarity of instructions, and the model’s familiarity with the programming language. Additionally, these generated codes may inadvertently utilize vulnerable built-in functions, potentially leading to source code vulnerabilities and exploits. This research undertakes an in-depth analysis and comparison of code generation, code completion, and security suggestions offered by prominent AI models, including OpenAI CodeX, CodeBert, and ChatGPT. The research aims to evaluate the effectiveness and security aspects of these tools in terms of their code generation, code completion capabilities, and their ability to enhance security. This analysis serves as a valuable resource for developers, enabling them to proactively avoid introducing security vulnerabilities in their projects. By doing so, developers can significantly reduce the need for extensive revisions and resource allocation, whether in the short or long term.
more » « less- Award ID(s):
- 2101161
- NSF-PAR ID:
- 10541804
- Publisher / Repository:
- CISSE
- Date Published:
- Journal Name:
- Journal of The Colloquium for Information Systems Security Education
- Volume:
- 11
- Issue:
- 1
- ISSN:
- 2641-4546
- Page Range / eLocation ID:
- 6
- Subject(s) / Keyword(s):
- Source Code Vulnerability Large Language Model LLM NLP Foundation Models CodeBert GPT
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Abstract—We conduct the first large-scale user study examining how users interact with an AI Code assistant to solve a variety of security related tasks across different programming languages. Overall, we find that participants who had access to an AI assistant based on OpenAI’s codex-davinci-002 model wrote significantly less secure code than those without access. Additionally, participants with access to an AI assistant were more likely to believe they wrote secure code than those without access to the AI assistant. Furthermore, we find that participants who trusted the AI less and engaged more with the language and format of their prompts (e.g. re-phrasing, adjusting temperature) provided code with fewer security vulnerabilities. Finally, in order to better inform the design of future AI-based Code assistants, we provide an in-depth analysis of participants’ language and interaction behavior, as well as release our user interface as an instrument to conduct similar studies in the future.more » « less
-
Despite the best efforts of the security community, security vulnerabilities in software are still prevalent, with new vulnerabilities reported daily and older ones stubbornly repeating themselves. One potential source of these vulnerabilities is shortcomings in the used language and library APIs. Developers tend to trust APIs, but can misunderstand or misuse them, introducing vulnerabilities. We call the causes of such misuse blindspots. In this paper, we study API blindspots from the developers' perspective to: (1) determine the extent to which developers can detect API blindspots in code and (2) examine the extent to which developer characteristics (i.e., perception of code correctness, familiarity with code, confidence, professional experience, cognitive function, and personality) affect this capability. We conducted a study with 109 developers from four countries solving programming puzzles that involve Java APIs known to contain blindspots. We find that (1) The presence of blindspots correlated negatively with the developers' accuracy in answering implicit security questions and the developers' ability to identify potential security concerns in the code. This effect was more pronounced for I/O-related APIs and for puzzles with higher cyclomatic complexity. (2) Higher cognitive functioning and more programming experience did not predict better ability to detect API blindspots. (3) Developers exhibiting greater openness as a personality trait were more likely to detect API blindspots. This study has the potential to advance API security in (1) design, implementation, and testing of new APIs; (2) addressing blindspots in legacy APIs; (3) development of novel methods for developer recruitment and training based on cognitive and personality assessments; and (4) improvement of software development processes (e.g., establishment of security and functionality teams).more » « less
-
Testing database-backed web applications is chal- lenging because their behaviors (e.g., control flow) are highly dependent on data returned from SQL queries. Without a database containing sufficient and realistic data, it is challenging to reach potentially vulnerable code snippets, limiting various existing dynamic-based security testing approaches. However, obtaining such a database for testing is difficult in practice as it often contains sensitive information. Sharing it can lead to data leaks and privacy issues. In this paper, we present SYNTHDB, a program analysis- based database generation technique for database-backed PHP applications. SYNTHDB leverages a concolic execution engine to identify interactions between PHP codebase and the SQL queries. It then collects and solves various constraints to reconstruct a database that can enable exploring uncovered program paths without violating database integrity. Our evaluation results show that the database generated by SYNTHDB outperforms state-of- the-arts database generation techniques in terms of code and query coverage in 17 real-world PHP applications. Specifically, SYNTHDB generated databases achieve 62.9% code and 77.1% query coverages, which are 14.0% and 24.2% more in code and query coverages than the state-of-the-art techniques. Fur- thermore, our security analysis results show that SYNTHDB effectively aids existing security testing tools: Burp Suite, Wfuzz, and webFuzz. Burp Suite aided by SYNTHDB detects 76.8% of vulnerabilities while other existing techniques cover 55.7% or fewer. Impressively, with SYNTHDB, Burp Suite discovers 33 pre- viously unknown vulnerabilities from 5 real-world applications.more » « less
-
We present an ethnographic study of secure software development processes in a software company using the anthropological research method of participant observation. Two PhD students in computer science trained in qualitative methods were embedded in a software company for 1.5 years of total research time. The researchers participated in everyday work activities such as coding and meetings, and observed software (in)security phenomena both through investigating historical data (code repositories and ticketing system records), and through pen-testing the developed software and observing developers’ and management’s reactions to the discovered vulnerabilities. Our study found that 1) security vulnerabilities are sometimes intentionally introduced and/or overlooked due to the difficulty in managing the various stakeholders’ responsibilities in an economic ecosystem, and cannot be simply blamed on developers’ lack of knowledge or skills; 2) accidental vulnerabilities discovered in the pen-testing process produce different reactions in the development team, often times contrary to what a security researcher would predict. These findings highlight the nuanced nature of the root causes of software vulnerabilities and indicate the need to take into account a significant amount of contextual information to understand how and why software vulnerabilities emerge during software development. Rather than simply addressing deficits in developer knowledge or practice, this research sheds light on at times forgotten human factors that significantly impact the security of software developed by actual companies. Our analysis also shows that improving software security in the development process can benefit from a co-creation model, where security experts work side by side with software developers to better identify security concerns and provide tools that are readily applicable within the specific context of the software development workflow.more » « less
