Researcher · Data Scientist · Security Researcher

Ibrahim Maniku

Investigating how AI shapes human judgement, trust, and the fabric of society.

Human AI Society
Try the AI Judgement Lab Explore my research About me
Interactive Exhibit · v0.3

AI Judgement Lab

Experience firsthand how AI framing shifts human decisions — the core of my PhD research on LLMs as cognitive mediators.

Select a scenario
1 · Your call
2 · AI advises
3 · Final call
Scenario

What do you do?
AI Recommendation
Now what do you do?
Judgement Shift Analysis
Human oversight weight40%
Verdict
ESCALATE
Risk Score
62
Areas of work

Four pillars

Research and practice across AI safety, data, cybersecurity, and design.

Featured work

Projects & publications

Research, demos, and design work.

Research · Publication
AI for Cyber Situational Awareness
Examining how AI recommendations affect human trust, decision-making and situational awareness in cybersecurity contexts.
📄 CyberScience 2026 · Presentation · June 2026
Artificial Intelligence for Cyber Situation Awareness: Trust, Distortion and Human Judgement
Investigates how AI-generated recommendations influence analyst judgement in cyber defence scenarios — exploring trust calibration, cognitive distortion, and implications for human-in-the-loop security systems.
DOI: forthcoming — full paper in review
AI safetyCyber SAHuman judgementLLMs
Data Science · Demo
NLP Sentiment Analysis
Transformer-based sentiment scoring across AI ethics discourse.
▸ sentiment_model.run(texts)
Cybersecurity · Demo
Anomaly Detection Engine
Simulated time-series login activity with statistical outlier detection.
▸ anomaly_detector.scan(stream)
⚠ 3 anomalies detected · confidence 0.87
Design · Engineering
Design & Mechanical Engineering
Arkwright Scholarship, back rehabilitation device, and extensive CAD portfolios.
📐
Portfolio files coming soon
CAD designs and engineering documentation uploading shortly.
Mechanical designCADProduct thinking
Get in touch

Collaborate

Open to academic collaboration, research opportunities, speaking, and consulting.

🔬
Academic Collaboration
Joint research on AI safety, human–AI interaction, or cognitive mediation.
📄
Research Opportunities
PhD connections, postdoc openings, or funded projects in AI and human behaviour.
🎤
Speaking & Consulting
Keynotes, workshops, and advisory on responsible AI and cybersecurity.
📥
CV Download
Academic and industry CV available on request.
Request CV
GitHub LinkedIn ORCID i.maniku@qmul.ac.uk
Background & Journey
Ibrahim Maniku
From mechanical engineering to AI safety research — driven by one question: how do we make technology work for society?
🧑‍💻
🎓 PhD Researcher · QMUL
🏛 Arkwright Scholar
🔐 Red Teamer · OXIQA
📍 London, UK
← Back to portfolio
2015 – 2016 · A Levels
Foundation
Design Technology — 100%
Achieved full marks across all design coursework, earning an Arkwright Scholarship and widespread recognition for CAD work. This set the foundation for everything that followed — precision thinking, product reasoning, and a belief that engineering should serve people.
2017 – 2020 · BEng (Hons)
Degree
Mechanical Engineering
Trained as a mechanical engineer with a drive to build things that help society evolve. I quickly noticed most engineering equations had been encoded into digital tools and wanted to go deeper — applying predictive models to problems like bridge maintenance and environmental failure projection.
2020 – 2021 · MSc · Distinction
Postgraduate
Data Science & Artificial Intelligence
Graduated with Distinction. Studied NLP, machine learning, statistics and AI — but the work opened harder questions. ML models can fail catastrophically in critical systems. They demand enormous data. And that data is often ours, harvested without meaningful consent, routinely in breach of GDPR and the UK DPA.
2021 – Present · Head of Data Science
Industry · Promoted
Data Scientist → Head of Data Science · Travel Agency
Applied data science in the family business — ML for customer behaviour, web scraping for content population, statistical SEO optimisation. Promoted to Head of Data Science. Also designed an ongoing loyalty scheme connecting travellers to environmentally friendly local experiences, raising sustainability awareness through seasonal passes in collaboration with photographers.
2023 – Ongoing · Consultation
Security
Cybersecurity Red Teamer · OXIQA
Trained and worked as a red teamer during the ChatGPT-3.5 moment — the beginning of the AI boom. Everyone now had equal access to generate malicious code. OXIQA taught me how organisations handle threats, but raised a bigger question: what happens to society when dangerous capabilities become universally accessible?
September 2025 – Ongoing · PhD
Now
PhD Researcher · Psychology & AI · QMUL
That question led here. I now research the impact of artificial social agents (ASAs) on human behaviour and society — specifically how LLMs act as cognitive mediators, shaping trust, distorting judgement, and altering decisions. Current projects: LLM personality modelling and a published paper on AI in cyber situational awareness.