Sidewalk Delivery Robot Injuries: Legal Guide

When Delivery Robots Strike Pedestrians

#

Autonomous delivery robots have invaded American sidewalks. These six-wheeled machines—some weighing up to 100 pounds—now share pedestrian space with children, elderly walkers, wheelchair users, and the visually impaired. What started as a novelty on college campuses has expanded to neighborhoods across the country, and with that expansion has come a growing list of injuries.

From a robot striking an Arizona State University employee and offering her “promo codes” as compensation, to Chicago residents documenting over 350 incident reports in a single petition, sidewalk delivery robots are creating a new category of pedestrian injury claims. Unlike autonomous vehicles on roads, these robots operate in spaces designed for people—raising unique questions about liability, accessibility rights, and corporate accountability.

The Sidewalk Robot Landscape

#

Major Players and Deployments

#

Starship Technologies:

• Operates at 30+ university campuses across the United States
• Over 4 million autonomous deliveries completed
• 11 million robot miles logged globally
• Secured $90 million in funding (February 2024)
• Instructs campus police to make Starship “first point of contact” for incident investigations

Serve Robotics:

• Operating in California and Texas under Uber Eats partnership
• Targeting 2,000 units by end of 2025
• Raised $80 million (January 2025), total funding exceeds $247 million
• Aiming for breakeven on 2,000-unit fleet

Kiwibot:

• Active at multiple university campuses
• Acquired Auto Mobility Solutions (April 2024) for robotics chips
• First “High-Driving Automation Campus” at Loyola Marymount University (2022)
• Co-founder acknowledged sidewalk limitations affect both robots and wheelchairs

Other Operators:

• Coco (Chicago pilot program)
• Amazon Scout (testing suspended, potential restart)
• FedEx SameDay Bot (limited deployment)

Market Growth and Proliferation

#

The delivery robot market is experiencing explosive growth:

North America accounts for 32-42% of global deployments. As robots multiply on sidewalks, so do opportunities for pedestrian injuries.

Documented Incidents and Injuries

#

The ASU Starship Incident

#

Location: Arizona State University Campus Date: September 2023 Operator: Starship Technologies

An Arizona State University employee was waiting for a Starship delivery robot to pass so she could cross the sidewalk near a parking garage. After the robot drove past her, she began walking. Suddenly and without warning, the robot “abruptly” went into reverse and struck her, knocking her to the ground.

What Made It Worse:

• The robot reversed a second time after the initial collision, nearly striking her again
• Police video footage confirmed no apparent cause for the robot’s behavior
• Police report noted “there was not a vehicle attempting to enter the garage or any other thing appearing to be in front of it”
• Starship offered the injured victim promo codes for its food delivery service as compensation

This incident exemplifies both the unpredictable behavior of delivery robots and the inadequate corporate response to injuries they cause.

Chicago’s Growing Problem

#

A December 2025 petition urging Chicago to pause its sidewalk delivery robot pilot has gathered over 1,500 signatures, with 350+ “incident reports” from residents describing their interactions with robots operated by Serve and Coco.

Documented Issues:

• Robots blocking sidewalks
• Stopping in crosswalks
• Creating obstacles for wheelchairs, walkers, and crutches
• Near-misses with pedestrians
• At least one injury requiring stitches

Finland Hit-and-Run

#

In February 2024, a Starship delivery robot in Finland caused damage to a parked car—lurching “back and forth against the side of the car” and scratching both doors. The robot then fled the scene without stopping, its “safety flag happily waving.”

This incident demonstrates that delivery robots can cause property damage and effectively “flee” accident scenes, complicating evidence collection and liability determination.

ADA and Accessibility Concerns

#

Who Is Most Vulnerable

#

Sidewalk delivery robots create unique hazards for people with disabilities:

Wheelchair Users:

• Robots blocking sidewalk passages
• Stopping in accessible curb cut areas
• Occupying already-limited wheelchair-accessible routes
• 2019 Pittsburgh incident: robot blocked wheelchair user from accessing sidewalk after crossing busy intersection

Visually Impaired Pedestrians:

• Increased tripping hazard from stationary or slow-moving robots
• Risk of damaging canes on robot contact
• Guide dogs struggling to navigate around robots
• No audible warnings from most robot systems

Mobility Device Users:

• Viral West Hollywood incident: Serve Robotics robot collided with therapist Mark Chaney’s mobility scooter, damaging it
• Sparked “accessibility outrage” over autonomous technology design

Regulatory Responses to Accessibility Concerns

#

Toronto (December 2021):

• City council banned autonomous delivery robots from sidewalks and bike lanes
• City’s accessibility advisory committee voiced safety concerns
• Cited Ontarians With Disabilities Act

Ottawa:

• Restricted sidewalk robots following Toronto’s decision
• Similar concerns about disability access

Chicago (2025):

• Accessibility is “major theme” in 1,500+ signature petition
• Residents with wheelchairs, walkers, and crutches documenting obstacles

The ADA Problem

#

While companies like Kiwibot claim their robots “follow ADA guidance,” the reality is more complicated:

• Many sidewalks already fail ADA standards for width and accessibility
• Adding robots further reduces available space
• Robots may comply with ADA in optimal conditions but create barriers in narrow passages
• No federal guidance specifically addresses delivery robots and ADA compliance
• Kiwibot co-founder David Rodriguez acknowledged: “If it’s a problem for a robot then it’s also a problem for a wheelchair”

The Regulatory Patchwork

#

State Authorization Landscape

#

As of 2025, over 20 states have authorized sidewalk delivery robots, with dramatically varying regulations:

Municipal vs. State Authority Battles

#

A critical legal issue is emerging: state preemption of local regulation.

Corporate Lobbying: Amazon and FedEx have successfully lobbied state legislatures to:

• Allow bigger and faster delivery robots
• Bar municipalities from creating their own robot regulations
• Set statewide policies that override local control

States with Preemption:

• Virginia: State law restricts municipal regulation
• Wisconsin: Similar preemption provisions
• Six additional states passed preemption bills (2023-2024)

Municipal Restrictions Where Allowed:

• San Francisco: Permits required per robot, maximum 9 permits citywide, human operator must be present
• Knoxville, TN (April 2024): City council voted unanimously to prohibit “personal delivery devices” within city limits
• Toronto/Ottawa: Banned from sidewalks entirely

What This Means for Injury Claims

#

The regulatory patchwork affects liability:

• In preemption states: Municipalities may lack authority to require safety measures
• In restrictive cities: Operators may be violating local ordinances
• Insurance requirements: Vary dramatically by jurisdiction
• Reporting mandates: Inconsistent across states
• Operator oversight: Some require human monitoring, others don’t

Liability Framework

#

Multiple Potential Defendants

#

Sidewalk delivery robot injuries can involve several liable parties:

Robot Operator/Fleet Company:

• Starship, Kiwibot, Serve Robotics, Coco, etc.
• Negligent deployment and operation
• Inadequate monitoring and intervention
• Failure to respond to incident patterns
• Insufficient warning systems

Robot Manufacturer:

• Design defects in navigation systems
• Manufacturing defects in sensors or brakes
• Failure to warn of known limitations
• Inadequate safety features

Software Developer:

• Algorithm design defects
• Sensor fusion failures
• Navigation errors
• Failure to handle foreseeable scenarios

Property Owner (Premises Liability):

• If injury occurs on private property
• Universities, shopping centers, apartment complexes
• May be liable for allowing robot operations
• Failure to warn of robot presence

Service Provider/Partner:

• Uber Eats, DoorDash, etc.
• Contracting with robot operators
• May share liability depending on relationship

Product Liability Theories

#

Design Defect:

• Navigation system fails to detect pedestrians
• Inadequate obstacle avoidance capabilities
• Reverse motion without adequate warning
• Speed exceeds safe pedestrian-area operation

Manufacturing Defect:

• Specific sensors not meeting specifications
• Brake system failures
• Assembly errors affecting navigation

Failure to Warn:

• Inadequate warnings about robot presence
• No audible alerts for visually impaired
• Misleading “safety” claims
• Insufficient training for property owners

Negligence Theories

#

Negligent Operation:

• Deploying robots in congested pedestrian areas
• Operating in poor visibility conditions
• Inadequate real-time monitoring
• Failure to intervene when problems detected

Negligent Maintenance:

• Software not properly updated
• Sensors not calibrated
• Known defects not corrected
• Inadequate inspection protocols

Negligent Training:

• Inadequate training for remote operators
• Failure to train property staff on robot hazards
• No protocols for pedestrian interactions

Defensive Arguments to Expect

#

Comparative Negligence

#

Defendants may argue the pedestrian contributed to the injury:

• Not watching where they were walking
• Using phone while walking
• Not yielding to visible robot
• Walking outside designated pedestrian areas

Counter-arguments:

• Robots invade pedestrian space, not vice versa
• Pedestrians have right-of-way on sidewalks
• Robots should yield to humans, not demand humans yield
• ASU incident: victim was simply waiting, robot reversed into her

Assumption of Risk

#

Limited applicability, but defendants may argue:

• Pedestrian knew robots operated in area
• Warning signs were posted
• Pedestrian proceeded despite seeing robot

Counter-arguments:

• Using public sidewalks is not “voluntary assumption of risk”
• No meaningful consent to robot hazards
• Warning signs don’t waive liability for defective products

Federal Preemption

#

Defendants may argue federal regulation preempts state claims:

• Currently no federal sidewalk robot regulations
• Unlike autonomous vehicles, no NHTSA standards apply
• Preemption arguments generally fail without federal standards

Evidence Preservation for Victims

#

Immediate Steps After a Robot Injury

#

1. Photograph everything — The robot, injuries, location, any damage
2. Note robot identification — Company name, serial numbers, any visible markings
3. Record time and location — Precisely where and when
4. Identify witnesses — Names and contact information
5. Seek medical attention — Even for seemingly minor injuries
6. Don’t accept “promo codes” — Document but don’t accept inadequate offers
7. Contact an attorney — Before speaking with robot company representatives

Critical Evidence to Request

#

Spoliation Concerns

#

Delivery robot data can be overwritten quickly. Send preservation letters immediately through an attorney to:

• Robot operating company (Starship, Kiwibot, Serve, etc.)
• Robot manufacturer
• Property owner where incident occurred
• Any food delivery platform (Uber Eats, DoorDash, etc.)
• Municipal agency overseeing robot permits

Practical Guidance for Injured Pedestrians

#

What Makes These Cases Different

#

Challenges:

• Multiple potential defendants with shared liability
• Technical evidence controlled by robot companies
• Evolving regulatory framework
• Limited case law precedent
• Companies may attempt inadequate settlements (e.g., promo codes)

Opportunities:

• Documented pattern of incidents builds case
• Accessibility advocates increasingly vocal
• Municipal concerns creating pressure
• Companies well-funded for adequate compensation
• Product liability applies to defective robots

Building Your Case

#

Strengthen Your Claim:

• Document all injuries with medical records
• Preserve evidence of robot malfunction
• Research prior incidents with same company/robot
• Identify regulatory violations
• Connect with accessibility advocates if applicable
• Review municipal permit conditions

Finding Pattern Evidence:

• Chicago petition: 350+ documented incidents
• AIID (AI Incident Database) tracks robot incidents
• University safety reports
• Municipal hearing records
• News coverage of prior incidents

Frequently Asked Questions

#

Related Resources

#

• Autonomous Vehicle Accident Claims — Self-driving car liability
• AI Medical Diagnosis Liability — AI error patterns in healthcare
• Amazon Warehouse Injuries — Robotics in logistics
• Understanding Liability — Product liability frameworks
• Evidence Checklist — What to preserve after any injury