Projects - Generative AI

Pair Noms

Singapore

Create government-style minutes, in minutes

Pair Search

Singapore

Search complex policy and legal records with ease

Pair Chat

Singapore

Empowering officers with Pair Chat: higher quality work, faster decisions, and better citizen experiences.

Pair Intern

Singapore

Bring AI assistance to internet-separated devices via email

Prune

San Francisco

AI-powered takedowns. We weaponize LLMs against digital exploitation

Sensemaker is an evolving toolkit developed by Jigsaw, designed to demonstrate how Google AI can help make sense of large-scale conversations. Sensemaker empowers users to categorize, summarize, and identify areas of agreement and disagreement in any discussion – turning conversation into clear, actionable insights, quickly.

Productive Ways to Use GenAI for Civil Servants

2300 N Street Northwest, Washington, DC 20037, USA

This playbook is designed specifically for civil servants, offering practical guidance for using generative AI (GenAI) to help them perform their work more effectively and efficiently.

guides

Google Maps Platform

Google Cloud now offers Geospatial Analytics, Generative AI on Street View imagery, and more environmental data as part of the broader Google Maps Platform

Local Minutes

Vermont

A chatbot AI that has access to millions of pages of municipal documents from small towns throughout Vermont.

local

Using Collective Dialogues and AI to Find Common Ground Between Israeli and Palestinian Peacebuilders

United States of America (the)

Human Computer Interaction article by Andrew Konya, Luke Thorburn, Wasim Almasri, Oded Adomi Leshem, Ariel D. Procaccia, Lisa Schirch, Michiel A. Bakker

Bridging Dictionary

MIT Media Lab

[A] dynamic prototype from MIT’s Center for Constructive Communication that identifies how words common in American political discourse are used differently across the political divide

Red teaming large language models (LLMs) for resilience to scientific disinformation

United States of America (the)

The red teaming event brought together 40 health and climate postgraduate students with the objective to scrutinise and bring attention to potential vulnerabilities in large language models (LLMs1 ).

The Students' Assembly on Gen AI in Higher Education

The University of Edinburgh

In the context of rising concerns regarding GenAI use, a two-day deliberative process was organized and hosted at the University of Edinburgh’s Edinburgh Futures Institute. A total of 17 students from 6 disciplines, gathered and deliberated on the question “How can we reimagine teaching, learning, and research methods to align with the advancements in AI and ethics?”.

Spinoza

The Spinoza project initially aims to create a tool for journalists based on generative artificial intelligence (AI) to enhance media coverage of climate change and ecological transition, all while relying on reliable sources.

Consult (Gov.UK)

10 Downing Street

Consult is a machine learning and LLM-powered tool to automate the processing of public consultations.

United Kingdom (government)

What the Fed?

United States of America (the)

AI-powered insights to help you find out what the f(ed) the US government is doing

Dokopalsya!

Russian Federation (the)

Dokopalsya! (which playfully means "got to the bottom of it" in Russian), leverages large language models to serve as a rapid fact-checking assistant in Russian. When users share claims or news snippets, the bot identifies key assertions and checks them against reliable sources, providing evidence-based verification with links to authoritative references.

Telegram

Naskvoz

Russian Federation (the)

"Naskvoz" (meaning "seeing through" in Russian), analyzes news texts for emotional manipulation techniques and propaganda markers.

Telegram

UrbanistAI

Europe

The platform to reimagine the future of our cities.

Open Politics Project

Berlin

The mission is to create an open-source data science and AI toolkit to analyse, summarise, and visualise political information.

State of Nevada using Google Vertex AI to Determine Worker Benefits

Nevada

"Nevada plans to launch a generative AI system powered by Google that will analyze transcripts of unemployment appeals hearings and issue recommendations to human referees about whether or not claimants should receive benefits."

Google

Synthetic Memories

Barcelona, Spain

an initiative aimed at recovering lost or undocumented visual memories using AI image generation.

Seelaw

London

A tool that provides targeted legislative insights and AI-powered analysis, delivering updates and impact assessments on legislation.

MLCommons AI Safety

The MLCommons AI Safety working group is composed of a global consortium of industry leaders, practitioners, researchers, and civil society experts committed to building a harmonized approach to AI safety.

CoCounsel

United States of America (the)

The First AI Legal Assistant, Made for Lawyers

commercial

Chat2024

United States of America (the)

An AI startup's chatbots representing each of the 17 major US presidential candidates.

US Election 2024

LAPD to use AI to analyze body cam videos for officers' language use

Los Angeles, CA

One of multiple studies in the US using artificial intelligence to analyze the tone and word choice police officers use, to determine whether it leads to unnecessarily escalations with the public.

Civic AI Handbook

London

The Civic AI Handbook is for digital leads of civic organisations who want to develop generative AI strategies for their organisations / departments / teams.

guides

Dive

India (Bhārat)

Dive is not just an AI assistant; it's your partner in optimizing meetings for maximum efficiency.

commercial

Hugging Face

The HF Hub is the central place to explore, experiment, collaborate and build technology with Machine Learning.

Futurepedia

The Largest AI Tools Directory

Albus

"Albus is an AI powered brainstorming tool that allows you to explore topics in depth. Helps prompt ideas so you can expand your knowledge with a mind map style visuals. Great for education and research." -futureailab.com

Meta's Massively Multilingual Speech (MMS) model

Facebook, Hacker Way, Menlo Park, CA, USA

Our Massively Multilingual Speech AI research models can identify more than 4,000 spoken languages, 40 times more than any known previous technology. These models expand text-to-speech and speech-to-text technology from around 100 languages to more than 1,100.