Generative AI for Software Project Management: Insights from a Review of Software Practitioner Literature

Basic Information

• Paper ID: 2510.10887
• Title: Generative AI for Software Project Management: Insights from a Review of Software Practitioner Literature
• Authors: Lakshana Iruni Assalaarachchi, Zainab Masood, Rashina Hoda, John Grundy
• Classification: cs.SE cs.AI
• Publication Date/Venue: IEEE Software (accepted on 7th Oct 2025)
• Paper Link: https://arxiv.org/abs/2510.10887

Abstract

Software practitioners are widely discussing the transformative impact of generative AI on software project management. To understand the current state, researchers conducted a grey literature review of 47 publicly accessible practitioner resources, including blogs, articles, and industry reports. The study found that software project managers primarily view GenAI as an "assistant," "co-pilot," or "colleague" rather than a "PM replacement." GenAI's support in automating routine tasks, predictive analytics, communication and collaboration, and agile practices contributes to project success. Practitioners emphasize responsible GenAI use, considering issues such as hallucinations, ethical privacy concerns, and the lack of emotional intelligence and human judgment. The research proposes skill enhancement requirements for software project managers in the GenAI era and maps these to the Project Management Institute's Talent Triangle model.

Research Background and Motivation

Problem Definition

The core problem this research addresses is: What changes are occurring in software project management practices in the context of rapid GenAI development? How are practitioners using GenAI tools, what challenges do they face, and what new skills are required?

Significance

1. Urgent Practical Demand: A 2024 PMI survey shows that 20% of respondents use GenAI in over half of their projects, with 54% using it in 16-50% of projects
2. Academic Research Lag: The academic community has conducted limited research on LLM applications in software project management, while practitioners are already widely discussing and implementing these tools
3. Critical Transformation Period: Software engineering is at a critical juncture of GenAI transformation, requiring redefinition of roles and skills

Limitations of Existing Approaches

• Existing research primarily focuses on GenAI applications in technical tasks such as code generation
• Lack of systematic research on GenAI applications in software project management practices
• Time gap between academic research and practical application

Research Motivation

By analyzing practitioners' real experiences and perspectives, the research aims to provide timely insights and recommendations for software project managers to maintain competitive advantage in the GenAI era.

Core Contributions

1. First Systematic Review: Conducted the first systematic review of practitioner literature on GenAI applications in software project management
2. Practical Insights Summary: Identified four major application domains of GenAI in software project management and associated benefits
3. Risk Identification and Response: Comprehensively analyzed major risks and response strategies identified by practitioners
4. Skill Framework Mapping: Mapped skill requirements in the GenAI era to the PMI Talent Triangle model
5. Practical Recommendations: Provided specific action recommendations for practitioners and researchers

Methodology Details

Task Definition

This research employs a grey literature review method to answer three research questions:

• RQ1: How do software practitioners use GenAI in SPM, and what benefits do they report?
• RQ2: What concerns do software practitioners have when using GenAI in SPM?
• RQ3: How can software project managers enhance their skills in the GenAI era?

Research Methodology Architecture

Data Collection Strategy

1. Search Tool: Google as the primary search engine
2. Search Scope: Includes keywords such as "Large Language Models," "LLMs," "ChatGPT," "Co-Pilot"
3. Time Range: From December 2022 (after ChatGPT's public release) to May 2025
4. Language Restriction: English-language resources only

Screening Process

• Initial Results: 618,000 search results
• First Round Screening: After applying time and language standards, 38,700 results remained
• Second Round Screening: Filtered for LinkedIn articles, blogs, and reports, yielding 15,400 results
• Manual Screening: Only reviewed the top 5 pages of most relevant results
• Final Sample: 47 valid practitioner resources

Quality Control

• Included only resources authored by software practitioners, software companies, or PM-related organizations
• Verified resource relevance through author affiliation or professional credentials
• Excluded promotional tool content

Data Analysis Method

Employed qualitative thematic analysis to extract key themes from practitioners' experiences and perspectives.

Experimental Setup

Dataset Characteristics

• Sample Size: 47 practitioner resources
• Resource Types: LinkedIn articles predominate
• Temporal Distribution: Primarily concentrated in 2023-2024
• Content Basis: Mostly based on authors' or organizations' practical experiences

Analysis Framework

Used the PMI Talent Triangle model as the theoretical framework for skills analysis:

• Ways of Working
• Power Skills
• Business Acumen

Coding System

Established a systematic coding system to categorize GenAI applications, benefits, and concerns (detailed in supplementary materials).

Experimental Results

Main Findings

RQ1: GenAI Usage and Benefits

Tools Used:

• General GenAI tools: ChatGPT, Microsoft Copilot, Gemini
• SPM tools with integrated GenAI: Jira, MS Project, Asana, Trello, Slack

Role Positioning:

• 17 resources positioned GenAI as an "assistant," "co-pilot," "team member," or "colleague"
• Emphasized supplementary rather than replacement roles

Four Major Application Domains:

RQ2: Major Concerns and Risks

Three Key Focus Areas:

1. Ethics and Privacy Issues (14 resources)
   • Risk of sharing sensitive project information
   • Concerns about customer and organizational information leakage
2. Data Quality and Accuracy (9 resources)
   • GenAI hallucinations leading to inaccurate insights
   • Impact of prompt quality on output quality
3. Lack of Emotional Intelligence and Human Judgment (4 resources)
   • Inability to provide human empathy and intuition
   • Limitations in scenarios such as conflict resolution

RQ3: Skill Enhancement Requirements

Skills Mapped to PMI Talent Triangle (22 resources):

• Ways of Working: Prompt engineering, ethics and data security, human-AI collaboration, learning adaptability
• Power Skills: Emotional intelligence (human interpersonal skills that GenAI cannot replace)
• Business Acumen: Analytical thinking, strategic decision-making, domain-specific ethics and data security knowledge

Enhancement Channels:

• Online blogs, videos, documentation
• Certification courses
• GenAI tools themselves as learning platforms

Case Analysis

One practitioner resource described using GenAI as a lightweight virtual assistant for most SPM tasks, indicating that GenAI can serve as a lightweight SPM assistant but cannot fully replace enterprise-level SPM tools' data storage, sharing, and collaboration capabilities.

Experimental Findings

1. Role Perception Shift: Practitioners widely view GenAI as an auxiliary tool rather than a replacement
2. Progressive Application: Expansion from simple document generation to predictive analytics and agile practices
3. Enhanced Risk Awareness: Practitioners have clear awareness of privacy, accuracy, and other issues
4. Diversified Skill Requirements: Need for not only technical skills but also ethical and emotional intelligence skills

Related Work

Major Research Directions in This Field

1. AI Applications in Software Engineering: Primarily focused on technical tasks such as code generation and testing
2. Digital Transformation of Project Management: Digitalization of traditional project management tools and methods
3. Human-AI Collaboration Research: Exploring collaboration patterns between AI and humans in work scenarios

Relationship of This Paper to Related Work

• Fills the research gap on LLM applications in software project management
• Bridges the gap between academic research and practical application
• Provides empirical support for the "augmented agile" concept

Advantages Over Related Work

• Based on real practitioner experiences rather than theoretical speculation
• Provides comprehensive risk assessment and response strategies
• Establishes a systematic skill development framework

Conclusions and Discussion

Main Conclusions

1. Auxiliary Role Established: Software project managers widely accept GenAI's role as an assistant
2. Application Value Evident: Demonstrates clear efficiency and quality improvements in four key domains
3. Risks Manageable: Major risks have been identified with corresponding response strategies
4. Skill Transformation Necessary: Comprehensive enhancement needed in technical, ethical, and interpersonal skills

Limitations

1. Data Source Limitations: Includes only text resources, excluding audio and video content
2. Sample Size: 47 resources may not represent all practitioner perspectives
3. Timeliness: Rapid GenAI development may render some findings outdated
4. Geographic Limitations: Primarily based on English-language resources, potentially subject to cultural bias

Future Directions

Recommendations for Researchers:

1. Research on GenAI's impact on software agile project management practices and project success
2. Research on barriers, challenges, and strategies for GenAI adoption in SPM
3. Research on ethical and responsible GenAI use in SPM
4. Research on the evolution of software PM roles in the GenAI era
5. Research on expectations, experiences, and impacts of human-AI collaboration in SPM

Technology Development Outlook: The paper proposes the concept of "agentic SPM," anticipating that AI agents will proactively enhance human decision-making through real-time data monitoring, improved risk management, and autonomous task management.

In-Depth Evaluation

Strengths

1. Research Timeliness: Captures the critical period of GenAI application in software project management
2. Methodological Appropriateness: Grey literature review method is suitable for exploring emerging practice areas
3. High Practical Value: Directly based on practitioner experiences with strong applicability
4. Systematic Framework: Provides structured skills analysis using the PMI Talent Triangle
5. Concrete Recommendations: Provides actionable recommendations for practitioners and researchers

Weaknesses

1. Sample Representativeness: The scale of 47 resources is relatively small with potential selection bias
2. Verification Gaps: Lacks empirical verification of practitioner perspectives
3. Limited Depth: As a review study, lacks quantitative analysis of specific application effects
4. Geographic Limitations: Primarily based on English-language resources with insufficient international representation
5. Temporal Snapshot: Reflects conditions at a specific point in time, lacking longitudinal analysis

Impact

1. Academic Contribution: Opens new research directions at the intersection of software engineering and project management
2. Practical Guidance: Provides a transformation guide for software project managers in the GenAI era
3. Policy Implications: Provides reference for organizations in formulating GenAI usage policies
4. Educational Value: Provides content framework for related professional education curriculum design

Applicable Scenarios

1. Software Project Management Practice: Applicable to software development projects of various scales
2. Organizational Digital Transformation: Provides reference for enterprise GenAI strategy formulation
3. Professional Skill Development: Provides guidance for project management professionals' career development
4. Academic Research Foundation: Provides theoretical and empirical foundation for subsequent in-depth research

References

The paper cites 15 important references covering recent research achievements in software engineering, artificial intelligence, project management, and other fields, with particular attention to PMI industry reports and cutting-edge concepts such as augmented agile.

Overall Assessment: This is a timely and practical research paper that fills the research gap on GenAI applications in software project management. While it has certain limitations in methodology and sample size, its insights based on real practitioner experiences have high practical value, providing valuable references for both academia and industry.