Bloomberg Businessweek

 TECHNOLOGY

23

FROM LEFT: ILLUSTRATION BY KATYA DOROKHINA; COURTESY REGULUS CYBER

Innovation GPS Spoof Defense

① A user plugs the Pyramid into a car

or drone between the GPS receiver

and control systems, or a manufacturer

builds it in.

② The device triangulates the

source of the signals it’s receiving

by combining five antennas and

onboard GPS receivers. Signals from

unexpected directions trigger an alert.

③ The Pyramid switches to navigation

from a route it downloaded before

starting the trip, then back to live GPS

data when it no longer detects any

suspicious signals.

Innovators

Yonatan Zur and Yoav

Zangvil

Ages: 41 and 39

Chief executive oicer

and chief technical oicer

of Regulus Cyber, an

11-employee startup in

Haifa, Israel

Origin

Zur and Zangvil,

co-workers at Israeli

defense contractor Elbit

Systems Ltd., founded

Regulus at the end of 2016

with an eye to improving

commercial drone security.

Funding

The company has raised

$6.3 million from Sierra

Ventures, Canaan Partners

Israel, the Technion-Israel

Institute of Technology, and

F2 Capital.

Early tests

NASA has been testing

prototype Pyramid

technology in drones in

North Dakota. Zur says

SwissDrones Operating AG

and AT&T Inc. are also in

the middle of trial runs.

Zur says Regulus is in talks with makers of cars and drones,

as well as operators of car and truck fleets, to conduct more

trials later this year, and it aims to bring the Pyramid to market

in 2019. As more autonomous vehicles hit the streets, there’ll

be a growing need for protection against GPS spoofing, says

Jonathan Petit, senior director of research at software maker

OnBoard Security Inc. Regulus says it’s refining its technology to

help defend the lidar, camera, and radar systems that help steer

self-driving cars. —Michael Belfiore

Next Steps

How It Works

Vehicles that rely on GPS navigation

are vulnerable to spoofing, the sending

of phony signals to lead them of

their intended course. The palm-size

Pyramid GPS SP from Regulus Cyber Ltd.

uses a bundle of antennas and receivers to

make sure the signals it’s reading are legit.

THE BOTTOM LINE Emergency managers in several cities and

counties are using AI services to help plan their disaster response,

but they remain largely unproven.

“You can save an order-of-magnitude more lives

with good planning,” says co-founder Nicole Hu.

The company says it plans to introduce similar ser-

vices for wildires and loods later this year.

The One Concern software is already being

used in San Francisco to plan drills, says Michael

Dayton, deputy director for the city’s Department

of Emergency Management. Dayton says the soft-

ware can predict, for example, whether and where

an earthquake is likely to cause ires, based on

where that earthquake strikes and how strong it

is. It can also predict the safest routes for bringing

aid and other supplies into the city.

In Utah, the nonproit Field Innovation Team is

experimenting with AI software that can anticipate

what people in shelters will need based on the ages

and health of those most likely to lose their homes.

That information can guide how the shelters are

designed, what help they ofer, and even what

kinds of donations oicials solicit from the pub-

lic, says founder Desiree Matel-Anderson, former

chief innovation adviser at the Federal Emergency

Management Agency.

AI isn’t a panacea, especially if the relevant data-

bases aren’t kept current. “That has to be an ongo-

ing efort,” says Mark Ghilarducci, director of the

California Governor’s Oice of Emergency Services.

“The last thing you want to do is make decisions on

old or bad information.” Given the power of such

systems, privacy is another concern. Just as import-

ant, the software can often be a black box that’s

diicult to hold accountable, says Sarah Miller,

chair of the Emerging Technology Caucus for the

International Association of Emergency Managers.

“If the AI somehow accidentally decides that those

who have higher incomes are more worthy of sav-

ing, then it might redirect resources accordingly,

and we might not know that,” she says.

Hu says One Concern will never sell or share

any personal information, while Tuneberg says

Geospiza ensures that such data are available only

to people with “an appropriate and valid need to

know.” She says that in the unlikely event that AI

inadvertently privileges some groups over others,

as it can in the risk assessment software sometimes

used in criminal sentencing, developers would be

able to spot those outcomes through regular test-

ing. “Minorities and people with disabilities are

ignored by the system through human approaches

every single day,” she says. “This idea that AI is

going to do worse by them, I would say, is ridicu-

lous.” —Christopher Flavelle