📌 Key facts

🔹 Mission: We, the AI Strategy Research Cluster TUM X BCG, explore how organisations can generate value of AI through organisational structure, processes, technology, people and governance. Our goal is to uncover the organisational factors that determine whether AI creates real impact or fails to deliver.

🔹 Starting date: Start anytime. Applications open now!

🔹 How to apply: Send an email including your CV and grade report, and research interest (Details at the end of this page).

• 📌 Key facts
• 💡 Background
• Master Thesis Opportunities – BCG Strategy AI Research Cluster
• 🧠 Thesis Proposals
• 1. Systematic Literature Review on Conversational AI, Agentic AI, Employee Voice & Sentiment Analysis
• 2. MIT: Conversational AI for Lecture Feedback
• 3. Irresponsible AI; The effectiveness of Deep Fake (Ravel8)
• 4. Persuasion Effectiveness Child-Ai Interactions (Steiff*)
• 5. Employee Voice with Conversational AI (BCG X)
• 6. Agentic AI Value Creation (n8n*)
• 7. Conversational AI for Employee Feedback (Sona8*)
• 8. Predicting CVR for Sales Agents (BMW*)
• 🦾Who We Are
• 🎓 Expectations
• 📄 Requirements to any Work
• 📬 How to apply

💡 Background

Our AI Strategy Research Cluster at TUM x BCG is on a mission to uncover how AI truly creates value beyond the hype. While many organisations have launched GenAI initiatives, few can demonstrate measurable top- or bottom-line impact. We believe that one potential missing link lies in how humans interact with AI: How they trust it, adapt to it, and integrate it into workflows and decision-making.

Our research blends behavioral science, organizational theory, and rigorous empirical methods with real world testings to understand how Conversational Agentic AI should be deployed as scale.

Master Thesis Opportunities – BCG Strategy AI Research Cluster

Hosted at the Chair of Strategy and Organization — Focus: Conversational & Agentic AI

We offer a portfolio of cutting-edge Master thesis projects exploring the intersection of Conversational AI, Agentic AI, organizational behavior, and human–AI interaction. The projects combine rigorous academic research with highly practical applications and collaborations with leading institutions such as MIT, Ravel8, BCGX, n8n, Steiff*, and various industry partners.

Students may apply for a specific topic and will be closely supervised throughout the research process.

🧠 Thesis Proposals

1. Systematic Literature Review on Conversational AI, Agentic AI, Employee Voice & Sentiment Analysis

Description

This thesis focuses on conducting a Systematic Literature Review (SLR) across four core domains:

• Conversational AI
• Agentic AI
• Employee Voice
• Sentiment Analysis

The goal is to map the academic landscape, identify theoretical foundations, synthesize current evidence, and highlight future research opportunities. Students develop a structured research framework and contribute to a foundational knowledge base for subsequent empirical work.

Ideal for

Students who enjoy theoretical, conceptual research, synthesizing complex fields, and working analytically — no programming required.

Type of work: Systematic Literature Review

2. MIT: Conversational AI for Lecture Feedback

Partner: MIT. Prof. Hartmann (Communication Class)

Description

This thesis explores how students use a voice-based Conversational AI to provide feedback on an MIT lecture. The core objective is to evaluate how AI-mediated feedback differs from traditional survey formats.

Key tasks include:

• Designing and managing the data collection pipeline
• Structuring and coding lecture feedback
• Creating randomized test groups
• Performing qualitative and quantitative analyses

This project provides a unique opportunity to participate in an active MIT research collaboration.

Type of work: Applied research & data analysis

3. Irresponsible AI; The effectiveness of Deep Fake (Ravel8)

Partner: REVEL-8 (Berlin)

Description

This project examines the effectiveness of voice-based deepfake phishing attacks, comparing three experimental conditions:

1. No contextual information
2. Context derived from social media
3. Insider-level contextual knowledge

Students design and run the experiment, support the creation of AI-generated voice agents, coordinate participant groups, and analyze attack success metrics.

Highlights:

• High degree of autonomy
• Cutting-edge intersection of AI, cybersecurity, and behavioral research
• Realistic deepfake generation (voice cloning provided)

Type of work: Experimental design & behavioral analysis

4. Persuasion Effectiveness Child-Ai Interactions (Steiff*)

Partner: Steiff (tbd, pilot planned for summer 2026, requires corporate alignment / pitching)

Description

Building on a fully functional prototype — a voice-interactive Teddy bear powered by Raspberry Pi — this thesis explores how children form relationships with AI-based toys. The central focus is the impact of context memory and personalized AI behavior on engagement and long-term usage.

Research areas include:

• How stored context and memory shape interaction quality
• Whether long-term relationship dynamics emerge
• Behavioral patterns that predict sustained engagement

The role includes supporting experiment logistics, prototype refinement, and comprehensive data analysis.

Type of work: Experimental field research & data evaluation

5. Employee Voice with Conversational AI (BCG X)

Partner: BCGX (expected launch: March)

Description

In this large-scale organizational study, approx. 3,000 BCG employees will participate in the Annual People Survey and additionally engage with a Conversational AI to share more nuanced insights.

The thesis analyzes how AI-facilitated dialogue augments traditional survey methods.

Responsibilities include:

• Supporting large-scale data collection
• Conducting sentiment and topic modeling
• Comparing structured survey results with conversational insights
• Identifying organizational improvement opportunities

Type of work: Organizational research & advanced sentiment analysis

6. Agentic AI Value Creation (n8n*)

Partner: n8n (tbd, requires corporate alignment / pitching)

Description

This thesis investigates how Agentic AI — autonomous AI agents — influence value creation in real organizational environments. Employees interact with AI agents through their daily tasks, and the thesis evaluates whether conversational engagement can enhance productivity and adoption.

Focus areas:

• Designing and implementing an organizational experiment
• Measuring productivity, workflow efficiency, and value contribution
• Identifying moderators such as task framing and AI-user dialogue
• Understanding adoption barriers and enablers

Type of work: Experimental organizational research

7. Conversational AI for Employee Feedback (Sona8*)

Partner: Sona8 (First test use case is a US VC)

Description

This thesis centers on the deployment and evaluation of Conversational AI solutions for employee feedback across multiple partner companies. Students gain hands-on experience in real-world testing, implementation, and iteration.

Core tasks:

• Designing and running pilot studies
• Collecting and analyzing employee interaction data
• Developing and validating new use cases
• Managing collaborations with corporate partners

Type of work: Applied AI research & project-based work

8. Predicting CVR for Sales Agents (BMW*)

Partner: BMW Group (tbd, topic under refinement)

Description

Developing virtual sales agents that assist end consumers in configuring vehicles, exploring options, checking availability, and receiving personalized recommendations. This thesis investigates how to predict and improve configuration completion rates, focusing on understanding when customers are likely to finalize a configuration or drop off during the process.

The research explores behavioral patterns in consumer–AI interactions, the role of configuration complexity, and opportunities to optimize agent guidance (e.g., follow-ups, emailed configurations, personalized suggestions).

Possible research areas:

• Modeling configuration completion likelihood based on user behavior and car configuration data
• Identifying friction points and dropout patterns in the configuration journey
• Optimizing agent prompting and recommendation strategies
• Understanding how conversational guidance influences consumer decisions

Ideal for

Students with strong analytical skills and interest in consumer behavior, AI interaction design, or digital sales experiences. Ideally previous/ongoing experience at BMW.

Type of work: Applied AI research & predictive modeling

🦾Who We Are

We are a research collaboration between the Technical University of Munich (TUM) and the Boston Consulting Group (BCG), focused on exploring the business value of Artificial Intelligence (AI), particularly Generative AI. Combining academic depth with industry relevance, we investigate how AI creates measurable impact across organizations spanning people, processes, platforms, and policies. Our work is grounded in real-world implementations and aims to produce actionable insights for both scholarly and managerial audiences. The Chair for Strategy and Organization is focused on research with impact. This means we do not want to repeat old ideas and base our research solely on the research people did 10 years ago. Instead, we currently research topics that will shape the future. Topics such as Agile Organisations and Digital Disruption, Blockchain Technology, Creativity and Innovation, Digital Transformation and Business Model Innovation, Diversity, Education: Education Technology and Performance Management, HRTech, Leadership, and Teams. We are always early in noticing trends, technologies, strategies, and organisations that shape the future, which has its ups and downs

🎓 Expectations

• Applicants should have the desire to co-author a publication in an academic journal
• Strong academic track record is a pre-requirement, the average TUM grade must be <1.X
• Applicants should have interest in hands-on field research
• Reliable, structured, precise and self-driven working - pragmatic solution oriented
• Applicants should be digital natives, ability to code is a plus, but not a requirement
• Applicants need motivation, optimism, and should write their thesis with fun

📄 Requirements to any Work

We do not want your research to gather dust in some corner of bookshelf but make it accessible to the world. Thus, we warmly encourage you to create some or all of the following:

• Infograph - visually represent some of your work (find examples here)
• Slide Deck - Summarise your research and possibly present it
• Optional: Medium Article - let people outside the university know about your research and start your personal brand

Please note that these deliverables are not officially required.

📬 How to apply

📌 Send us a brief application including:

✅ Your CV

✅ Your Transcript of Records/Grade Report(Transcript of Records)

✅ Your preferred starting date & your research interest/motivation

✅ Your motivation via mail, don’t use GPT to generate your application

📩 E-Mail an: thilo.tamme@tum.de or tamme.thilo@bcg.com // Thilo Tamme Get in touch today, before others may take your thesis spot….! 🚀