Skip to content
alothPublic

About

Applied AI researcher & author. Building tools for information integrity (JudgeGPT, CRED-1, Origin Lens), developer productivity (olcli MCP, Trackless Links), and health tech (Mindful Coffee). Open-source, npm packages, MCP servers.

Topics

Resources

Stars

4 stars

Watchers

1 watching

Forks

Latest commit

 

History

45 Commits

Folders and files

NameName
Last commit message
Last commit date
 
 
 
 
 
 

Repository files navigation

Hi there, I'm Alexander Loth! 👋

Applied AI researcher and author combating misinformation through technology.

I build tools that strengthen information integrity - from cryptographic image verification to AI-driven fake news detection. My work sits at the intersection of Generative AI, Agentic AI, Human-Computer Interaction, and ethical systems, informed by a data science background from CERN.


📄 Journal Publication

Experts Disagree on How to Fight AI Disinformation, but Agree That Health and Politics Need Different Solutions
Harvard Kennedy School Misinformation Review 7(4), 2026 · doi:10.37016/mr-2020-205 · open data · survey instrument


🔬 Featured Research: Agentic AI & Information Integrity

Research hub: GitHub repo · landing pages
The umbrella entry point for projects, papers, datasets, and conference materials across this research program.

Publication Project Venue
Experts Disagree on How to Fight AI Disinformation, but Agree That Health and Politics Need Different Solutions verification-crisis HKS Misinformation Review 7(4)
The Mitigation Paradox: How AI-Disinformation Experts Converge on Threats but Polarize on Solutions (preprint) verification-crisis AIES 2026
Industrialized Deception: The Collateral Effects of LLM-Generated Misinformation on Digital Ecosystems JudgeGPT · RogueGPT WWW '26 Companion
Eroding the Truth-Default: A Causal Analysis of Human Susceptibility to Foundation Model Hallucinations and Disinformation in the Wild JudgeGPT · RogueGPT WWW '26 Companion
The Verification Crisis: Expert Perceptions of GenAI Disinformation and the Case for Reproducible Provenance verification-crisis WWW '26 Companion
Origin Lens: Reclaiming Trust on the AI-Mediated Web Through On-Device Image Provenance Verification Origin Lens WebSci '26
Human Perception of LLM-Generated News (Can Humans Tell? · Indistinguishability Threshold) JudgeGPT · RogueGPT WebSci '26 Companion
CRED-1: An Open Multi-Signal Domain Credibility Dataset for Automated Pre-Bunking of Online Misinformation CRED-1 SSRN
Blessing or Curse? A Survey on the Impact of Generative AI on Fake News JudgeGPT arXiv

📱 iOS Apps

Apps for iPhone and iPad, on the App Store.

App Description Tech Stack
Trackless Links · App Store Safari extension for iOS & macOS. Strips tracking parameters, blocks cookie banners, and warns about unreliable sources via CRED-1 - all on-device, zero data collection. Reviewed in Mac & i 4/2026 (heise). Also available as Trackless Links Pro. Swift JavaScript WebExtensions
Dreamwave Radio · App Store Sleep sound mixer. Layer up to three of 27 stations; most are built as they play, so they never develop the loop point that gives other noise apps away at 3am. Siri, a fading sleep timer, nine languages, no account and no network. SwiftUI AVAudioEngine DSP
Mindful Coffee · App Store Caffeine timing against your own cortisol rhythm - when a coffee helps, and when it costs you sleep. HealthKit & iCloud sync. SwiftUI SwiftData HealthKit
Mindful Body · App Store Body composition tracking with progress photos and trend analysis. HealthKit & iCloud sync. SwiftUI SwiftData CloudKit

🚀 Featured Open Source

Project Description Tech Stack
olcli npm GitHub stars Overleaf CLI & git remote - pull, push, sync, compile LaTeX projects from terminal or AI agents. MCP server with 14 tools for Claude, Cursor & Windsurf. Available on npm. Install as AgentSkill (overleaf-skill). TypeScript MCP Git CLI
awesome-ai-agents GitHub stars A curated list of AI agent frameworks, tools, platforms, research papers, and resources. Awesome List
PowerSkills GitHub stars Windows automation for AI agents - Microsoft 365 via Work IQ & Graph (mail, calendar, Teams, SharePoint, OneDrive), Outlook COM, Edge browser (CDP), desktop, and system control. Structured JSON output. PowerShell Microsoft 365 Graph API CDP Win32
CRED-1 npm GitHub stars Open, reproducible domain credibility dataset for automated pre-bunking of online misinformation. TypeScript library, CLI & MCP server - verify any news source's reliability from terminal, code, or AI agents. 5 MCP tools for Claude Desktop, Cursor & Windsurf. Presented at ACM WebSci 2026, CC BY 4.0 and DOI-citable. Powers Trackless Links. DOI TypeScript MCP Library CLI CC BY 4.0
Origin Lens GitHub stars On-device C2PA & EXIF image provenance verification - check whether an image carries valid provenance signals before you trust it. Open source, no tracking. Presented at WebSci '26. Dart Flutter C2PA
First Responder App GitHub stars An open source AI-powered web app providing real-time collaboration and guidance for emergency teams. React Azure AI PWA

📚 Authored Books

I have authored and co-authored several books on data and AI, aimed at making complex topics accessible to a broader audience.


Connect & Collaborate

My professional journey has taken me through Geneva, Frankfurt, Shanghai, Bangalore, and Redmond. I am always open to collaborating on projects related to AI ethics, applied health tech (especially chronobiology), and social good initiatives.

Website LinkedIn X Profile Bluesky Mastodon ORCID Google Scholar

About

Applied AI researcher & author. Building tools for information integrity (JudgeGPT, CRED-1, Origin Lens), developer productivity (olcli MCP, Trackless Links), and health tech (Mindful Coffee). Open-source, npm packages, MCP servers.

Topics

Resources

Stars

4 stars

Watchers

1 watching

Forks

Sponsor this project

Contributors