The AI landscape doesn't move in one direction — it lurches. Some techniques leap from experiment to table stakes in a single quarter; others stall against regulatory walls, technical ceilings, or organisational inertia that no amount of hype can dislodge. Knowing which is which is the hard part. The State of Play cuts through the noise with a rigorously maintained index of AI techniques across every major business domain — classified by maturity, evidenced by real-world adoption, and updated daily so you always know where you stand relative to the field. Stop guessing. Start knowing.
A daily newsletter distilling the past two weeks of movement in a domain or two — delivered to your inbox while the index updates in the background.
Each dot marks the weighted maturity of practices within a domain — hover for a brief summary, click for more detail
Autonomous robots that deliver items within buildings and across last-mile outdoor environments. Includes sidewalk delivery robots and indoor courier bots; distinct from drone delivery which operates in airspace rather than on surfaces. Scope covers ML/AI-driven approaches; prior deterministic or rules-based automation is out of scope.
Autonomous delivery robots demonstrate a durable, profitable business in controlled campus environments—and have fundamentally failed to scale in contested urban sidewalks. Starship's 9 million cumulative deliveries across 60-plus US campuses, with 97% student approval and confirmed profitability, represent genuine leading-edge deployment: real value, real economics, repeatable. But this success is confined to closed environments that most cities and organisations have not and cannot replicate. Urban vendors like Coco and Serve have collectively deployed tens of thousands of units, yet that rapid scaling has triggered predictable regulatory hardening: Chicago banned commercial sidewalk robots in March 2026, Miami Beach imposed strict constraints, and multiple cities implemented moratoria. Safety incidents (bus shelter collisions documented in March 2026, accessibility failures, navigation errors in mixed-traffic environments) and municipal resistance have intensified faster than product maturity. The practice's structural tension remains unresolved: campus environments offer repeatable, profitable deployments to supportive institutions; public sidewalks present interlocking regulatory, safety, and social barriers with no evidence of relaxation. International evidence (Finland's 1,000+ robots achieving 10% food delivery GMV penetration) shows commercialisation is possible in permissive jurisdictions—but North America's fragmented regulatory landscape remains a durable constraint. Analyst projections remain bullish, but actual deployment geography has barely expanded beyond campuses and a handful of permissive cities since 2024.
Campus deployment has consolidated at production scale. Starship operates 2,700+ robots across 270+ global locations (including 60+ US campuses) with 9 million cumulative deliveries and confirmed profitability; fleet is expanding to 12,000 units by 2027. Oregon State University generated 265,000 orders in 2025, and Colorado State University demonstrates sustainable unit economics ($11,400 revenue in 6 weeks, 65% margins). DoorDash's Dot robot (20 mph, 30 lb payload) integrates sidewalk robots with drones and human couriers in a multi-modal platform strategy. Kiwibot maintains 500+ robots across 26+ US campuses through Sodexo partnership.
Urban scaling has triggered regulatory hardening. Serve Robotics operates 2,000+ confirmed robots (SEC filing, April 2026) across 20 US cities serving 4,500+ restaurants via Uber Eats and DoorDash integrations, completing hundreds of thousands of autonomous deliveries. Coco operates ~1,000 robots across six US/European cities and released next-generation hardware (upgraded cameras, LiDAR, snow tires) with claimed 50-60% cost advantage. However, March 2026 incident cluster in Chicago (Serve and Coco robots colliding into CTA bus shelters with pedestrian injuries) triggered Chicago City Council ban on commercial sidewalk robots. Glendale imposed moratorium pending safety review. Nuro's exit is definitive: California DMV data showed only 646 miles per disengagement in 2025 (down from 2,044 in 2024), validating strategic pivot to autonomous robotaxi licensing with Uber and Lucid.
Geographic differentiation has emerged. Finland demonstrates commercialisation viability with 1,000+ deployed sidewalk robots achieving ~10% food delivery GMV penetration—the clearest evidence of crossed adoption threshold outside campus environments. South Korea enforced robot safety certification system (July 2025) enabling international regulatory maturity. Starship expanded to 3,000 robots across UK/Europe (Sunderland, Leeds, Sheffield, pan-Europe rollout) demonstrating profitable international scaling. North American regulation remains fragmented: Pennsylvania legalised sidewalk robots without mandatory municipal input, while San Francisco, Toronto, and Santa Monica maintain bans or moratoria. Congressional attention emerged (April 2026) with Ranking Member Lofgren highlighting safety standards gaps in deployed robots. Market analysts project 27.2% CAGR growth ($921M 2025 to $6.6B 2034), but campus viability and international scaling mask persistent North American barriers: accessibility failures (Carnegie Mellon), low public support (40%, below majority), and documented teleoperation scalability constraints.
Recent May 2026 developments confirm the bifurcated market. Starship achieved 10 million cumulative deliveries operating 3,000+ robots across 300+ locations in 8 countries, with 125,000 daily autonomous road crossings demonstrating production-scale maturity and durable unit economics. Serve Robotics' $3M Q1 2026 revenue (581% YoY growth) and 2,000-unit deployment across 44 cities in 14 states serving 3,500+ restaurants shows vendor platform viability at scale. DoorDash's Dot robot launch with multi-modal orchestration platform signals deepening platform integration. However, Philadelphia's May 2026 delivery robot collision (May 4) exposed a critical regulatory gap: no federal agency officially tracks delivery robot incidents, requiring 10 phone calls to locate responsible state oversight. Glendale's moratorium draft (May 2026) cites worker displacement and accessibility concerns amid Serve's 25x geographic expansion (2 neighborhoods in 2023 to 40 by May 2026). Operationally, sidewalk robots increasingly collect municipal infrastructure and safety data (sidewalk condition, curb cuts, near-miss events), extending deployment value beyond logistics. International regulatory maturity advances: South Korea's April 2026 sandbox exemption enabling raw video AI training signals ecosystem refinement. The pattern persists: sustained campus viability with strong economics and approval ratings (97% at Starship), parallel urban scaling with regulatory friction and incident-related moratoria, and persistent incident-reporting gaps constraining federal oversight.
— Serve Robotics achieved production-scale deployment with 2,000 robots across 44 cities in 14 states, serving 3,500+ restaurants via Uber Eats and DoorDash, generating $3M Q1 2026 revenue (581% YoY growth) with targeting toward $1/delivery cost economics.
— Uber published comprehensive safety guidelines for autonomous delivery robots and drones effective May 2026, incorporating 3 years of real-world deployment experience, formalizing vendor compliance frameworks and signaling operational maturity across multiple autonomous modalities.
— Avride delivery robot collision with pedestrian (May 4, 2026, Philadelphia) exposed critical regulatory gap: no federal agency tracking delivery robot complaints, requiring 10 phone calls to locate responsible state agency, revealing incident-reporting weakness in distributed autonomous deployment.
— Serve Robotics scaled from 2 LA neighborhoods (2023) to 500+ units across 40 neighborhoods by May 2026, demonstrating rapid urban deployment growth; simultaneously triggered municipal moratoria (Glendale, Chicago) and California DMV citation authority for AV traffic violations starting July 2026.
— Glendale City Council drafted moratorium on commercial sidewalk delivery robots amid 25x geographic expansion (from 2 to 40 neighborhoods in 3 years), citing delivery worker displacement and pedestrian accessibility concerns, exemplifying municipal regulatory hardening as deployment scales.
— Serve Robotics robots operating in West Hollywood generate sidewalk condition data (width, curb cut damage) and real-time pedestrian safety metrics (near-miss events), demonstrating operational maturation into municipal infrastructure assets beyond logistics delivery.
— DoorDash released proprietary Dot robot (20 mph, 30 lb payload) and Autonomous Delivery Platform enabling multimodal orchestration between human couriers, robots, and drones, signaling platform-level vertical integration and strategic capability investment.
— Starship reached 10M cumulative deliveries operating 3,000+ robots across 300+ locations in 8 countries, with 22M kilometers driven and 125,000 daily road crossings operating autonomously, confirming production-scale maturity and profitable international scaling.
2018: Starship Technologies deployed autonomous delivery robots across 10+ cities including corporate campuses at Intuit (Mountain View) and university grounds, establishing viability of ground-based autonomous delivery in controlled environments. Nuro announced Kroger partnership for grocery delivery pilot in Scottsdale, Arizona, marking entry of a second major vendor and broadening use case from parcel to grocery.
2019: Campus adoption accelerated with deployments at Purdue (30+ robots, 100k+ cumulative deliveries globally), Northern Arizona University (12k deliveries in 7 weeks), George Mason, and University of Houston, establishing repeatable campus franchise model. Nuro expanded to public urban delivery in Houston with Kroger ($5.95 per delivery), but NYC banned FedEx robots in December citing safety and accessibility failures, exposing limits of urban readiness without geographic and regulatory tailoring.
2020: Starship expanded to Arizona State University (40 robots, 14 restaurants) and University of Pittsburgh (operational deployment with documented limitations), demonstrating pandemic-driven adoption for contactless delivery across multiple campuses. Kiwibot entered urban markets with 25 robots in San Jose beyond campus focus. Nuro achieved NHTSA exemption for public-road testing of R2 in February, advancing regulatory progress. Academic research (RWTH Aachen, Portland State) documented real-world deployment challenges including pedestrian interactions and efficiency trade-offs. Critical accessibility failures emerged (guide dogs blocked, curb cuts inaccessible), revealing design gaps and limiting broader public sidewalk deployment readiness.
2021: Campus deployment model consolidated with Starship expanding to N.C. A&T (20 robots, first HBCU partnership) and establishing first presence in North Carolina. Kiwibot emerged as credible second platform, announcing strategic partnership with food services giant Sodexo targeting 50+ locations and deploying to New Mexico State, Loyola Marymount, and Gonzaga universities; LMU completed 500+ deliveries in operational trial. However, accessibility barriers hardened into regulatory pushback: Ontario Disability Rights coalition formally petitioned Toronto City Council (December 2021) to ban sidewalk robots, citing persistent failures to accommodate wheelchair users, guide dogs, and elder pedestrians. Platforms remained narrowly focused on controlled campus environments; public urban sidewalk deployment blocked by cumulative design, safety, and regulatory constraints.
2022-H1: Starship extended campus dominance with University of Tennessee deployment (40 robots, 16 eateries, April 2022) and published national adoption survey (May 2022: 7,063 students across 20+ campuses, 98% positive sentiment, 61% weekly usage). Nuro launched third-generation commercial vehicle with Kroger in Houston (January 2022, doubled cargo capacity, $600M funding), advancing product maturity. However, regulatory barriers hardened: Toronto and Ottawa formalized sidewalk bans, Kirkland imposed moratorium on Amazon Scout robots (March 2022). Real-world deployments surfaced persistent challenges—public resistance (robots kicked, blocked, pranked), design gaps (curb cut and guide dog failures)—exposing limits of current systems for unrestricted urban deployment.
2022-H2: Campus adoption continued to expand with Kiwibot scaling to 500+ robots on 26 U.S. college campuses, competing directly with Starship's market dominance. Starship's Milton Keynes deployment in the UK demonstrated environmental and economic impact (70% car journey replacement, 400,000kg carbon reduction, double-digit business sales increases), showing international viability in supportive regulatory environments. However, empirical research (Northern Arizona University) and real-world incidents documented persistent pedestrian safety concerns and operational constraints (99% autonomy claims requiring significant human oversight and staffing). Canadian regulatory environment hardened further with continued prohibitions citing accessibility and congestion concerns; Serve Robotics' pilot in Vancouver achieved 95% customer satisfaction but faced barriers to widespread North American urban expansion. The practice demonstrated durability in controlled campus and select international environments, but municipal resistance to public sidewalk deployment remained the dominant barrier.
2023-H1: Campus deployment model continued unchanged with Kiwibot expanding to University of Maine (September 2023) and Shenandoah University (15 robots), extending Sodexo partnership across additional campuses. However, regulatory and operational constraints tightened further. Peer-reviewed research published in March 2023 documented persistent safety and comfort concerns in pedestrian interactions with sidewalk robots at NAU, providing empirical evidence of real-world deployment challenges. Early 2023 analysis revealed that at least 23 U.S. states had established specific regulations governing delivery robots by end of 2022, but pilot programs demonstrated minimal customer adoption: Pittsburgh achieved only 4 customer deliveries before discontinuation due to infrastructure challenges, while Detroit completed only 12 deliveries due to low consumer awareness and operational barriers. Regulatory barriers remained in place across major Canadian markets (Toronto, Ottawa bans intact) with no evidence of policy reversal. Campus environments remained the only viable operational domain, while broader urban sidewalk expansion remained blocked by persistent regulatory hostility, accessibility failures, and demonstrated low adoption in uncontrolled public settings.
2023-H2: Campus deployments continued with Starship expanding to Sam Houston State University (8 restaurant partners) and Kiwibot sustaining operations at University of Findlay despite adoption challenges from DoorDash competition and low student app uptake. Consumer acceptance research published in November 2023 identified significant public preference barriers—surveyed U.S. respondents demonstrated substantially lower acceptance of delivery robots compared to autonomous vehicles, indicating structural demand constraints. South Korea enacted first major regulatory framework enabling autonomous delivery robot operations on sidewalks with weight and speed limits, signalling potential regulatory maturity in supportive jurisdictions. However, North American regulatory resistance continued: pedestrian advocacy organizations opposed Massachusetts legislation, reflecting sustained public safety concerns in major U.S. markets. Market projections through 2030 forecast autonomous last-mile delivery growth from $0.9B (2023) to $4.2B (CAGR 22.7%), but adoption growth remained geographically concentrated in permissive campus environments rather than general urban deployment.
2024-Q1: Starship achieved profitability milestone with 6M cumulative deliveries across 80 global locations and $90M funding round (February 2024), confirming market leader viability amid industry consolidation (FedEx/Amazon pilots ended). South Korea expanded nationwide regulatory framework (January 2024) permitting raw video use for AI collision-avoidance training, advancing autonomous safety capabilities. Japan demonstrated in-building commercial deployments (Shinagawa Season Terrace, NTT Group trial) achieving 30% delivery-time reduction through optimized multi-robot coordination. However, ethnographic research (HRI 2024) documented persistent real-world friction: robots encounter 30-60 second navigation stops from environmental obstacles requiring human accommodation. North American regulatory barriers remained immovable (Toronto/Ottawa bans intact) and consumer demand constraints persisted. Campus environments continued as sole North American reliable operational domain; international expansion in supportive regulatory environments (South Korea, Japan) emerged as differentiated growth vector distinct from blocked North American urban sidewalk deployment.
2024-Q2: Serve Robotics emerged as third-platform competitor with $40M public offering (April 2024) and Magna International manufacturing partnership; reported $0.95M Q1 revenue with 124% growth and 2,000-robot Uber Eats deployment target. Empirical research on post-adoption behavior (550 users) identified hedonic and utilitarian factors driving reuse and recommendation. Finland survey documented positive/neutral consumer acceptance. However, Carnegie Mellon CHI 2024 research documented critical accessibility failures: robots blocking wheelchair-safe paths and unable to serve customers with mobility disabilities, highlighting persistent design barriers to urban sidewalk deployment. North American regulatory bans and accessibility gaps remained immovable constraints despite emerging platform competition and demonstrated international regulatory progress (South Korea, Japan).
2024-Q3: Campus consolidation accelerated with Starship expanding to 50+ North American university locations (Georgia Tech July, Cal Poly 40-robot September deployment) reaching 1.5M students; wireless charging infrastructure rolled out to 75% of locations, addressing operational constraints. Samsung C&T and Neubility launched South Korea apartment-building pilot with Level 4 autonomy under nationwide regulatory framework enabling raw video AI training. Nuro faced production challenges (tariffs, layoffs) while maintaining Houston-only operations. However, persistent safety concerns surfaced: September 2023 ASU incident (robot reversing into pedestrian, causing injury) highlighted unresolved accessibility and safety integration barriers constraining broader urban adoption.
2024-Q4: Campus consolidation persisted with Starship expanding to University of Minnesota (Minneapolis regulatory approval), Cal Poly (41 robots, 200+ daily orders, $30-40M Grubhub revenue), confirming production-scale deployment reaching 1.5M+ students across 50+ North American campuses. International regulatory progress accelerated: South Korea's Intelligent Robot Development Promotion Act amendment (effective October 2024) permitted autonomous delivery robots on public sidewalks; Robotis demonstrated GAEMI robot in Seoul showing advanced obstacle avoidance and adaptive speed control, though safety concerns persisted pending full commercialization. Nuro announced L4 driverless expansion to multiple cities (November 2024) despite continued production constraints from tariffs and workforce reductions. However, North American regulatory barriers hardened: Santa Monica imposed one-year moratorium (December 2024) due to safety and obstruction concerns, extending Toronto/Ottawa bans. Critical vulnerabilities emerged: peer-reviewed research on Finnish public sentiment (November 2024) documented persistent acceptance barriers across safety, operations, and security dimensions; December 2024 Serve Robotics collision with Waymo robotaxi highlighted navigation challenges in mixed-traffic urban environments; accessibility gaps (wheelchair path obstruction, inaccessible service) documented by Carnegie Mellon persisted as design-level constraints. The practice demonstrated durable campus viability with production-scale consolidation, modest international regulatory progress in supportive jurisdictions, but interlocking North American regulatory bans, accessibility failures, low consumer adoption intent, and safety concerns remain definitive barriers to urban sidewalk expansion.
2025-Q1: Campus deployment momentum continued with Starship expanding to Colorado State University (January 2025, 100-pound capacity, 30-60 min delivery, queue fully subscribed day one), establishing 55+ North American university locations with consistent high demand exceeding capacity. Global market projections signalled strong investor confidence: autonomous last-mile delivery market projected at $38.01B in 2025, growing 29.5% to $49.23B by 2026, though actual deployments remained concentrated in campus environments rather than broad urban adoption. However, research identified fundamental adoption barriers beyond technology: Vrije Universiteit Brussel PhD dissertation (March 2025) highlighted trust, design, and traffic integration as key constraints, emphasizing that real-world deployment experience remains limited; Urban Robotics Foundation analysis (March 2025) documented the 'regulation-innovation gap' and noted that sidewalk infrastructure itself creates inherent barriers requiring societal adaptation. Regulatory scrutiny began intensifying: Minneapolis City Council initiated mid-pilot review of Starship robots (February 2025) citing worker impact and safety incident concerns, signalling emerging municipal pushback as campus deployments scaled. North American urban expansion barriers persisted: regulatory bans (Toronto, Ottawa, Santa Monica moratorium) combined with design-level accessibility gaps (Carnegie Mellon findings) and research on adoption barriers indicated that broader deployment would require infrastructure and social changes, not technology advances alone.
2025-Q2: Nuro raised $106M in Series E funding (April) at $6B valuation and announced expansion of autonomous testing to Dallas, Miami, and San Diego (May), signalling continued vendor scaling efforts. Veho and RIVR completed Austin trial (May-June) with 95% delivery success rate over 400 autonomous deliveries, demonstrating real-world human-robot collaboration feasibility. However, regulatory barriers intensified: San Francisco Board approved ordinance (December 2024, reported April 2025) banning commercial sidewalk delivery robots with narrow R&D path only. Peer-reviewed research documented teleoperation challenges and human intervention requirements as critical scalability barriers; Urban Robotics Foundation analysis called for proactive regulations to address safety, accessibility, and enforcement gaps. The window reinforced bifurcated market: campus consolidation and vendor scaling on one side, regulatory hardening and municipal resistance on the other.
2025-Q3: Vendor strategic divergence emerged amid market consolidation: Nuro raised $203M Series E financing (July 2025) with partners Lucid and Uber for autonomous robotaxi development, signalling strategic pivot beyond delivery robot constraints; Starship confirmed $75M 2025 revenue, 80+ global locations, 8M+ cumulative deliveries at production scale. International regulatory progress: South Korea enforced 16-category robot safety certification system (July 2025) with first certified robots expected early 2026, signalling regulatory maturity. However, urban barriers intensified: Pennsylvania enacted autonomous robot law (August 2025) legalizing sidewalk operations without mandatory local municipal input; critical safety failure documented—Serve Robotics' West Hollywood collision (September 2025) revealed paradoxical design flaw where safety system designed to predict pedestrian intentions instead impeded pedestrian with cerebral palsy, exemplifying accessibility liability risks. Market evidence showed US autonomous last-mile market at $500M with continued growth projections, but adoption remained concentrated in campus consolidation (Cal Poly 200+ daily orders generating $30-40M dining revenue). Peer-reviewed evidence confirmed teleoperation as fundamental scalability barrier. Campus viability persisted alongside regulatory fragmentation and emerging safety-system accessibility failures constraining broader urban deployment.
2025-Q4: Campus consolidation and vendor platform diversification persisted: Starship expanded to 9M+ cumulative deliveries, 12M miles, 270+ global locations, 60+ US campuses; partnered with Veriff for biometric age-verification enabling regulated-goods delivery (alcohol, pharmaceuticals) with fleet expansion to 12,000 robots by 2027. DoorDash launched proprietary Dot robot (Sep 2025, 20 mph, 30 lbs) as multi-modal platform entry, integrating autonomous robots with human couriers, drones, and vehicles. Public sentiment data (YouGov, 27,553 respondents) showed 40% support, 35% opposition, 26% unsure, indicating measured adoption intent below majority. Emerging municipal resistance: Chicago pilot (Nov 2025) provoked 800+ resident petition citing safety and accessibility concerns with reported collision injury; legal liability analysis documented pedestrian injury risks. International regulatory progress sustained: South Korea's enforced safety certification system (July 2025) signalled maturity enabling first certified units by early 2026. However, North American barriers persisted: regulatory fragmentation (Pennsylvania legalization without municipal input; San Francisco, Toronto bans; Santa Monica moratorium), documented safety-system accessibility failures creating liability exposure, emerging municipal resistance from real-world incidents, and public support below 50% constraining broader urban expansion.
2026-Jan: Campus adoption metrics reinforced market leadership: Starship's 2025 Campus User Survey (5,000 students, 65 campuses) reported 97% approval with 13% reporting improved mental health; Oregon State University achieved highest-volume deployment with 265,000 orders in 2025 and 1.2M cumulative orders since program inception. Vendor scaling plans advanced: Starship announced fleet expansion from 2,700 to 12,000+ robots by 2027 and demonstrated extreme-weather capability through Lapland, Finland operations launch (Jan 2026). Market growth projections sustained industry confidence: autonomous last-mile delivery market projected $1.3B (2025) to $11.5B (2035) at 24.5% CAGR. However, real-world safety incidents persisted in urban operations: LAFD ambulance collision with Coco robot in Hollywood (Jan 2026) and prior train collision in Miami (Jan 2026) illustrated reliability and coordination challenges in mixed-traffic environments. Regulatory progress remained fragmented: Riverside, California advanced municipal framework for DoorDash Dot deployment while Serve Robotics' acquisition of Diligent Robotics signaled platform diversification into indoor healthcare logistics, distinct from outdoor last-mile scope.
2026-Feb: Vendor strategic divergence and regulatory hardening accelerated simultaneously. Nuro's pivot to autonomous robotaxis advanced with Uber/Lucid partnership targeting 20,000 vehicle deployment globally by 2032 using Nuro Driver (Level 4 DaaS platform); California DMV data (Feb 2026) documented reliability constraints underpinning the shift—Nuro's AVs averaged only 646 miles per disengagement (down from 2,044 in 2024), far below Waymo's 19,234 and Zoox's 60,682, validating the strategic abandonment of custom delivery robots as unprofitable. Rapid urban scaling triggered municipal resistance: Coco ~10,000 robots and Serve Robotics ~2,000 units deployed across US cities; Chicago's 1st Ward (Wicker Park/Logan Square) effectively banned expansion after community survey found 83% "strongly disagreed" with wider robot deployment; Miami Beach adopted tightened regulations (8 mph speed limits, safety flags, audible sounds, monitoring, $100,000 liability insurance, annual $100 permits) exemplifying shift from permissive to constraint-based municipal governance. Market analyst confidence remained high despite urban friction: 360iResearch projected USD 6.71B (2025) to USD 7.63B (2026) market growth at 15.27% CAGR; sidewalk robot market specifically forecast at USD 1.9B (2026) to USD 12.6B (2036) with Starship holding 31% competitive share. Starship expanded international operations with Just Eat Sunderland launch (Feb 2026) leveraging 2,700+ robot fleet and 9M+ cumulative deliveries; continued fleet expansion toward 12,000 units by 2027. Campus consolidation sustained: 97% user approval and $30-40M revenue from Cal Poly dining operations (85-90% transaction coverage). Urban deployment demonstrated bifurcated market: sustained campus viability with strong economics and approval, contrasted with rapid urban scaling (10k+ robots deployed) provoking community and regulatory resistance, reliability constraints driving vendor platform abandonment, and municipal governance increasingly favoring constraint-based regulation over permissive frameworks.
2026-Q2: Multi-vendor platform expansion and critical urban incident cluster. Starship-Uber Eats partnership expanded UK/Europe deployment (Leeds, Sheffield launch Dec 2025, pan-Europe rollout March 2026) with 3,000 robots across 270+ global locations and 9M+ cumulative deliveries confirming profitable international scaling. Coco released next-generation robot with upgraded hardware (cameras, LiDAR, submersible design, snow tires) operating 1,000+ units across LA/Chicago/Miami/Helsinki, ramping production to 1,000/month with claimed 50-60% cost advantage vs human couriers. DoorDash Dot completed first autonomous delivery in Fremont, CA (March 2026) with domestic manufacturing and municipal partnership; Florida Polytechnic became first US university to adopt Starship 360 POS platform integration (April 2026). Campus revenue metrics continued: Colorado State University generated $11,400 from robot deliveries in 6 weeks with 600 deliveries/week, mini-market accounting for 33% of volume at $15.67 check average with 65% margin. However, critical operational failures emerged in Chicago: March 23-26, 2026 incident cluster documented Serve and Coco robots crashing into CTA bus shelters causing glass shattering and pedestrian injuries; Chicago City Council voted to ban commercial sidewalk delivery robots March 2026 marking regulatory ceiling in major North American city. Glendale, CA imposed 4-week moratorium on Serve Robotics (April 2026) pending safety review. Temple University researchers (April 2026) documented labor shortage and cost-efficiency as primary deployment drivers alongside persistent human-robot interaction research gaps. Platform consolidation evidence: 24 US states had PDD legislation with per-delivery economics ranging $1.40-1.90, below profitability thresholds without scale. Serve Robotics' SEC DEF 14A filing (April 2026) confirmed 2,000+ robots deployed as of December 31 2025, serving 4,500+ restaurants via Uber Eats and DoorDash; Uber Eats disclosed hundreds of thousands of autonomous robot deliveries across 10+ cities with four vendor partners (Serve, Coco, Cartken, Avride). Finland crossed a mainstream adoption threshold with 1,000+ deployed sidewalk robots achieving ~10% food delivery GMV penetration—the clearest international benchmark of commercialisation viability. Coco deployed BlindSquare hazard-detection integration across ~1,000 US/European robots, sharing real-time obstacle data to a visually impaired navigation app in 26 languages. A US Congressional hearing (April 2026) with Ranking Member Lofgren highlighted the safety-standards gap in deployed robots, signalling federal policy attention. Analyst projections reached $6.6B by 2034 at 27.2% CAGR. The practice demonstrated sustained campus viability (Colorado State, Florida Poly) and international urban scaling (Starship-Uber UK/EU, Finland mainstream), but urban adoption barriers intensified through documented safety incidents, regulatory hardening, and emerging legislative scrutiny.
2026-May: Production scale milestones and safety governance gaps emerged simultaneously. Starship achieved 10 million cumulative deliveries operating 3,000+ robots across 300+ locations in 8 countries with 125,000 daily autonomous road crossings; Serve Robotics reported $3M Q1 2026 revenue (581% YoY growth) from 2,000 units across 44 cities in 14 states serving 3,500+ restaurants, and DoorDash launched its Dot robot with a multimodal Autonomous Delivery Platform orchestrating robots, drones, and human couriers—confirming vendor platform viability at scale. Against these scale signals, a Philadelphia Avride robot collision (May 4) exposed a critical governance gap: no federal agency tracks delivery robot incidents, and locating the responsible state authority required 10 phone calls; Glendale's moratorium proposal cited worker displacement and accessibility concerns amid Serve's 25x geographic expansion, while Uber published its first formal autonomous delivery safety framework formalising vendor compliance standards after three years of real-world deployment.