AI Research

Introducing Mobility AI: Advancing urban transportation

Published

5 months ago

April 23, 2025

1. Measurement: Understanding mobility patterns

Accurately evaluating the current state of the transportation network and mobility patterns is the first step to improving mobility. This involves gathering and analyzing real-time and historical data from various sources to understand both current and historical conditions and trends. We need to track the effects of changes as we implement them in the network. ML powers estimations and metric computations, while statistical approaches measure impact. Key areas include:

Congestion functions

Similar to well-known fundamental diagrams of traffic flow, congestion functions mathematically describe how rising vehicle volume increases congestion and reduces travel speeds, providing crucial insights into traffic behavior. Unlike fundamental diagrams, congestion functions are built based on a portion of vehicles (e.g., floating car data) rather than all traveling vehicles. We have advanced the understanding of congestion formation and propagation using an ML approach that created city-wide models, which enable robust inference on roads with limited data and, through analytical formulation, reveal how traffic signal adjustments influence flow distribution and congestion patterns in urban areas.

Foundational geospatial understanding

We develop novel frameworks, leveraging techniques like self-supervised learning on geospatial data and movement patterns, to learn embeddings that capture both local characteristics and broader spatial relationships. These representations improve the understanding of mobility patterns and can aid downstream tasks, especially where data might be sparse or when complementing other data modalities. Collaboration with related Google Research efforts in Geospatial Reasoning using generative AI and foundation models is crucial for advancing these capabilities.

Parking insights

Understanding urban intricacies includes parking. Building on our work using ML to predict parking difficulty, Mobility AI aims to provide better insights for managing parking availability, crucial for various people, including commuters, ride-sharing drivers, commercial delivery vehicles, and the emerging needs of self-driving vehicles.

Origin–destination travel demand estimation

Origin–destination (OD) travel demand, which describes where trips — like daily commutes, goods deliveries, or shopping journeys — start and end, is fundamental to understanding and optimizing mobility. Knowing these patterns is crucial because it reveals exactly where the transportation network is stressed and where services or infrastructure improvements are most needed. We calibrate OD matrices — tables quantifying these trips between locations — to accurately replicate observed traffic patterns, providing a spatially complete understanding essential for planning and optimization of transportation networks.

Performance metrics: Safety, emissions and congestion impact

We use aggregated and anonymized Google Maps traffic trends to assess impact of transportation interventions on congestion, and we build models to assess safety and emissions impact. To build safety metrics scalably, we go beyond reactive crash data by utilizing hard braking events (HBEs). HBEs are shown to be strongly correlated with crashes and can be used for road safety services to pinpoint high-risk locations and predict future collision risks.

To measure environmental impact, we’ve developed AI models in partnership with the National Renewable Energy Laboratory (NREL) that predict vehicle energy consumption (whether gas, diesel, hybrid, or electric). This powers fuel-efficient routing in Google Maps, estimated to have helped avoid 2.9M metric tons of GHG emissions in the US alone, which is equivalent to taking ~650,000 cars off the road for a year. This capability is fundamental for monitoring climate and health impacts related to transportation choices.

Impact evaluation

Randomized trials are often infeasible for evaluating transportation policy changes. To assess the impact of a change, we need to estimate outcomes in its absence. This can be done by finding cities or regions with similar mobility patterns to serve as a “control group”. Our analysis of NYC’s congestion pricing demonstrates this method through use of sophisticated statistical techniques like synthetic controls to rigorously estimate the policy’s impact and by providing valuable insights for agencies evaluating interventions.

Source link

AI Research

Study shakes Silicon Valley: Researchers break AI

Published

2 hours ago

September 14, 2025

The Editors

Study shakes Silicon Valley: Researchers break AI | The Jerusalem Post

Study shows researchers can manipulate chatbots with simple psychology, raising serious concerns about AI’s vulnerability and potential dangers.

ChatGPT encouraged a teenager toward suicide

(photo credit: OpenAI)

ByDR. ITAY GAL

SEPTEMBER 14, 2025 09:13

Source link

AI Research

Password1: how scammers exploit variations of your logins | Money

Published

2 hours ago

September 14, 2025

Shane Hickey

The first you know about it is when you find out someone has accessed one of your accounts. You’ve been careful with your details so you can’t work out what has gone wrong, but you have made one mistake – recycling part of your password.

Reusing the same word in a password – even if it is altered to include numbers or symbols – gives criminals a way in to your accounts.

Brandyn Murtagh, an ethical “white hat” hacker, says information obtained through data breaches on sites such as DropBox and Tumblr and through cyber-attacks has been circulating on the internet for some time.

Hackers obtain passwords and test them out on other websites – a practice known as credential stuffing – to see whether they can break into accounts.

But in some cases they do not just try the exact passwords from the hacked data: as well as credential stuffing, the fraudsters also attempt to access accounts with derivations of the hacked password.

Research from Virgin Media O2 suggests four out of every five people use the same or nearly identical passwords on online accounts.

Using a slightly altered passwords – such as Guardian1 instead of Guardian – is almost an open door for hackers to compromise online accounts, Murtagh says.

Working with Virgin Media O2, he has shown volunteers how easy it is to trace their password when they supply their email address, often getting a result within minutes.

A spokesperson for Virgin Media O2 says: “Human behaviour is quite easy to model. [Criminals] know, for example, you might use one password and then add a full stop or an exclamation mark to the end.”

What the scam looks like

The criminals use scripts – automated sets of instructions for the computer – to go through variations of the passwords in an attempt to access other accounts. This can happen on an industrial scale, says Murtagh.

“It’s very rare that you are targeted as an individual – you are [usually] in a group of thousands of people that are getting targeted. These processes scale just like they would in business,” he says.

You might be alerted by messages saying that you have been trying to change your email address or other details connected to an account.

What to do

Change any passwords that are variations on the same word – Murtagh advises starting with the most important four sets of accounts: banks, email, work accounts and mobile.

Use a password managers – these are often integrated into web browsers. Apple has iCloud Keychain while Androids have Google Password Manager, both of which can suggest and save complicated passwords.

Put in place two-factor authentication or multi-factor authentication (2FA or MFA), which mean means you have two steps to log into a site.

Source link