Nearhuman

Edge AI safety system for shared e-scooters and e-bikes, processing computer vision on-device.

Cover Block

From the public record

Field	Value
Company	Nearhuman
Tagline	Edge AI safety system for shared e-scooters and e-bikes, processing computer vision on-device [nearhuman.co]
Headquarters	Bristol, United Kingdom [GlobeNewswire, October 2026]
Founded	2023 [Crunchbase]
Stage	Pre-seed [GlobeNewswire, October 2026]
Business model	Hardware + Software
Technology	AI / Machine Learning
Geography	Western Europe
Growth profile	Venture Scale
Founding team	Solo Founder, Faizan Mir [RocketReach]
Funding label	Pre-seed
Total disclosed funding	£350,000, approximately $427,000 [GlobeNewswire, October 2026]

The Short Version

PUBLIC Nearhuman is building an on-vehicle AI safety system for shared e-scooters and e-bikes, and it merits investor attention now because it has paired a fresh pre-seed financing with a live Bristol pilot, giving the company an early chance to prove whether retrofit edge vision can translate into operator-grade safety telemetry in the field [GlobeNewswire, October 2026] [The Manila Times, October 2026] [nearhuman.co]. Founded in Bristol in 2023, the company is centered on Scootrr, a retrofit camera and embedded vision stack that Nearhuman says can detect pavement riding, pedestrian close calls, and road hazards in real time, with processing kept on the vehicle rather than pushed to the cloud [nearhuman.co] [GlobeNewswire, October 2026].

That product positioning is the clearest point of differentiation in the public record: Nearhuman says footage is redacted on board, the system runs at 30 fps at roughly 2 W, and the hardware is intended for vehicles already in service, which, if borne out in deployment, could lower integration friction for fleet operators relative to a full vehicle redesign [nearhuman.co]. The near-term commercial question is less about whether the technical story is legible and more about whether pilots convert into named operator relationships, repeat orders, and evidence that councils or insurers treat the data as decision-useful [GlobeNewswire, October 2026] [nearhuman.co].

The team story is founder-led and still concentrated. Public materials identify Faizan Mir as founder and CEO, describe a robotics background at UWE Bristol, and place him at the Bristol Robotics Laboratory Hardware Incubator, with company materials also referencing a filed patent and prototype units shipped for operator testing [nearhuman.co/about] [GlobeNewswire, October 2026] [RocketReach].

On financing, Nearhuman disclosed a £350,000 pre-seed led by SFC Capital in October 2026, equal to about $427,000 at the disclosed converted figure in the structured research, and the business model is best read as hardware plus software sold into shared micromobility operators, with a secondary analytics value proposition for councils and city authorities [GlobeNewswire, October 2026] [StartupMag, October 2026] [Crunchbase]. Over the next 12 to 18 months, the public milestones that matter are straightforward: pilot validation in Bristol, disclosure of at least one named fleet or OEM relationship, and evidence that on-device safety analytics can support a repeatable commercial motion rather than remain an interesting prototype [The Manila Times, October 2026] [nearhuman.co] [GlobeNewswire, October 2026].

Unconfirmed -- This section relies materially on company materials and company-originated funding coverage, with limited independent corroboration beyond round reporting.

Taxonomy Snapshot

Axis	Value
Stage	Pre-Seed
Business Model	Hardware + Software
Industry / Vertical	Other
Technology Type	AI / Machine Learning
Geography	Western Europe
Growth Profile	Venture Scale
Founding Team	Solo Founder
Funding	Pre-seed, total disclosed ~$427,000 [GlobeNewswire, October 2026]

The Company in Brief

PUBLIC

Nearhuman is a Bristol, United Kingdom startup building edge AI safety systems for micromobility, with its public materials centered on Scootrr, a computer vision product for e-scooters, e-bikes, and related vehicles [nearhuman.co] [Crunchbase]. The company presents its work in plain operational terms: safety intelligence that runs on the vehicle rather than relying on cloud processing, a positioning that matters because shared-fleet operators tend to care about latency, retrofitability, and privacy at the same time [nearhuman.co].

The public record on formation is still thin, but the available chronology is straightforward. Crunchbase lists Nearhuman as founded in 2023 and based in Bristol [Crunchbase]. By October 2026, the company had announced a £350,000 pre-seed round led by SFC Capital, which Nearhuman said would support product development, on-street testing, and commercial deployment [nearhuman.co] [Crunchbase].

What stands out at this stage is not breadth of disclosed company history, but the consistency of the early story across the limited primary sources. The website frames Nearhuman as a micromobility-first business, and public company materials do not point to a broader multi-vertical strategy yet [nearhuman.co]. That leaves investors with a compact but coherent picture: a young Bristol company, formed in 2023, focused on one applied edge AI use case and only recently entering the public funding record [nearhuman.co] [Crunchbase].

Single-source, plausible -- This section relies on the company website and Crunchbase, with chronology only partially corroborated by independent public records.

What They Have Built

MIXED The product story is straightforward, and the technical claim matters because Nearhuman is not pitching fleet software first. It is pitching a retrofit safety module called Scootrr for shared e-scooters and e-bikes, combining a camera, embedded hardware, computer vision, and on-device AI to detect events such as pavement riding, pedestrian close calls, and road hazards in real time [nearhuman.co] [GlobeNewswire, October 2026]. Nearhuman says the system is designed to fit vehicles already in service rather than require a new vehicle platform, which is a practical detail for shared micromobility operators that already manage mixed fleets and legacy hardware [nearhuman.co] [StartupMag, October 2026].

The company also leans heavily on edge processing as the core design choice. Nearhuman says processing happens on the vehicle, with footage redacted on board so sensitive video does not need to leave the vehicle, and its public materials state the system runs at 30 frames per second using roughly 2 watts of power [nearhuman.co] [nearhuman.co]. That framing suggests the product is being positioned as a privacy-conscious operational tool for operators and potentially city authorities, rather than as a cloud-first telematics layer, although public materials do not yet provide a verified demo with disclosed accuracy rates, false-positive rates, or named customer deployments [nearhuman.co] [LinkedIn].

Nearhuman has disclosed enough to show a working product surface, but less to prove production readiness at scale. Public reporting and company materials indicate a live pilot in Bristol, shipped prototypes for operator testing, and pilot discussions with operators and OEMs, yet the sources reviewed do not name the operator, OEM, or any commercial contract tied to the system [The Manila Times, October 2026] [nearhuman.co] [GlobeNewswire, October 2026]. For investors, the technical bet appears to rest on whether on-vehicle vision can deliver useful safety analytics within the power, privacy, and retrofit constraints the company is advertising, a claim that remains directionally credible but only partly corroborated in public.

Single-source, plausible -- Product description is corroborated by Nearhuman's website and October 2026 funding coverage, but key performance and deployment claims remain primarily company-sourced.

Market Size and Demand

PUBLIC The market matters now because shared micromobility has moved from pilot novelty to regulated urban infrastructure, and that shift raises the cost of safety failures for both operators and cities [nearhuman.co] [Road.cc, October 2026].

Nearhuman does not have a cited third-party market size benchmark in the available materials, so the cleanest public framing is to start from the problem set its product is trying to address rather than force a speculative TAM. The company positions Scootrr as a retrofit safety layer for shared e-scooters and e-bikes, aimed at operators and potentially city authorities, which places it at the intersection of micromobility fleet software, vehicle safety systems, and edge computer vision for transport [nearhuman.co] [GlobeNewswire, October 2026]. That is a narrower and more actionable market than consumer mobility hardware, because the purchase decision likely sits with fleet operators or public-sector stakeholders rather than individual riders, according to the company and coverage of its Bristol pilot [nearhuman.co] [The Manila Times, October 2026].

The available public evidence points to demand being driven by incident costs, compliance pressure, and retrofit economics. Nearhuman's homepage cites more than 30,000 injuries across the UK and EU, 1,300 casualties in the UK with roughly GBP 7 million of cost to the NHS and insurers, and 9,425 reported e-scooter accidents in Germany, up 14.1 percent year over year [nearhuman.co]. Those figures are company-presented and should be treated as directional rather than independently verified market sizing, but they help explain why an operator might pay for systems that detect pavement riding, pedestrian close calls, and road hazards in real time [nearhuman.co] [Road.cc, October 2026]. The retrofit angle also matters. If operators can add safety hardware to vehicles already in service, the sales motion is easier than waiting for a full OEM platform cycle, at least in principle [nearhuman.co] [GlobeNewswire, October 2026].

A small table is more honest than a market model here because the numeric claims in the public record describe the safety burden, not the addressable revenue pool.

Market signal	Figure	Source
Injuries across the UK and EU	30,000+	[nearhuman.co]
UK casualties	1,300	[nearhuman.co]
NHS and insurer cost in the UK	GBP 7M	[nearhuman.co]
Reported e-scooter accidents in Germany	9,425	[nearhuman.co]
Germany YoY accident growth	14.1%	[nearhuman.co]

The table does not size Nearhuman's market directly. It does show that the company's wedge is tied to a measurable safety problem, but the underlying figures come from the company website rather than an independent transport dataset, so they are better read as demand context than as a basis for revenue forecasting [nearhuman.co].

The adjacent markets are easier to identify than to quantify from the current record. One adjacency is OEM-integrated rider assistance and telematics for light electric vehicles, where safety sensing is built into the vehicle rather than sold as a retrofit module. Another is municipal safety and compliance tooling, since Nearhuman says its analytics are relevant to councils and city authorities, not only operators [Edward Stevenson - SFC Capital | LinkedIn, retrieved 2026] [nearhuman.co]. A third substitute path is cloud-based fleet monitoring or basic telematics, which may address operator oversight at lower cost but would not necessarily offer on-vehicle, real-time detection or onboard redaction, according to Nearhuman's product description [nearhuman.co].

Regulation is the clearest macro tailwind in the public record, even if the evidence is still early. Shared scooters and e-bikes operate in a politically sensitive category where pavement riding, pedestrian conflicts, and insurer scrutiny can shape permitting and fleet economics, which is the backdrop for Nearhuman's Bristol pilot and SFC Capital-backed safety pitch [GlobeNewswire, October 2026] [Road.cc, October 2026]. The company is effectively betting that cities and operators will prefer privacy-conscious edge processing over cloud video collection, especially where public acceptance depends on limiting sensitive footage transfer off the vehicle [nearhuman.co] [nearhuman.co]. That is a plausible market direction, but the current public evidence does not yet show whether procurement budgets will form around this category quickly enough to support venture-scale expansion.

Unconfirmed -- Market context relies materially on company website claims, with only limited corroboration from GlobeNewswire, Road.cc, and other coverage of the funding and pilot.

Who Else Is Fighting for This

MIXED Nearhuman is trying to occupy a narrow position inside micromobility safety: not fleet management in general, and not a new vehicle platform, but a retrofit edge-AI module that sits directly on shared e-scooters and e-bikes and processes safety events on the vehicle itself [nearhuman.co] [GlobeNewswire, October 2026].

The competitive map is easier to define by substitution than by named head-to-head peers, because the sourced public record here does not identify direct competitors by name. Incumbents in shared micromobility already have alternatives to buying a dedicated on-vehicle vision system: they can rely on rider education, manual operations teams, geofencing, insurance controls, and standard telematics stacks, all of which address safety or compliance from different angles [nearhuman.co]. Adjacent substitutes also include OEM-native hardware and broader fleet software providers, even when those products are not described in the available sources as computer-vision safety systems. That matters because Nearhuman is not only selling detection accuracy, according to its website and funding coverage, it is asking operators to add hardware to vehicles already in service, which raises the bar for installation effort, unit economics, and proof of operational benefit [nearhuman.co] [StartupMag, October 2026].

Where the company appears strongest today is product architecture rather than distribution. Nearhuman states that Scootrr is designed to retrofit existing e-scooters and e-bikes, process footage on-device, redact footage on board, and run at 30 frames per second using roughly 2 watts of power [nearhuman.co]. If those claims hold up in field conditions, the value proposition is fairly specific: lower latency than cloud-dependent analysis, a privacy posture that may be easier for operators and councils to defend, and deployment on installed fleets rather than waiting for a new hardware cycle [nearhuman.co] [The Manila Times, October 2026]. The question is durability. Architecture can be a real edge at pre-seed stage, but unless it compounds into exclusive fleet data, regulatory approvals, or deeply embedded operator workflows, it is perishable and can be matched by better-capitalized suppliers over time.

The company is more exposed on go-to-market than on concept. Public materials point to a live Bristol pilot and prototype testing with operators, but they do not name a customer, fleet partner, or OEM relationship, and the disclosed financing is modest at £350,000 pre-seed led by SFC Capital [GlobeNewswire, October 2026] [finsmes.com, October 2026]. In practice, that leaves Nearhuman competing not only against direct startups that may emerge with similar edge-compute claims, but also against the default decision to do nothing new until an operator sees a measurable reduction in claims, incidents, or regulatory friction. The absence of named operator wins does not disprove traction, but it does mean the competitive moat is still mostly technical and narrative rather than contractual.

The most plausible 18-month scenario is a sorting between companies that can turn pilot data into fleet-standard procurement and those that remain in evaluation mode. Nearhuman is the likely winner if retrofit installation proves straightforward and if on-device privacy handling becomes a meaningful purchasing criterion for operators or city stakeholders, because that would favor its stated architecture and narrow focus on shared micromobility safety [nearhuman.co] [GlobeNewswire, October 2026]. Nearhuman is also the likely loser if operators decide the problem is better solved through broader fleet software, OEM-integrated hardware, or policy tools that avoid adding another physical module to each vehicle. With no named public rival supported by the source set, the more serious competitive threat for now is substitution by incumbent workflows rather than a clearly identified startup taking the same budget line.

Unconfirmed -- This section relies primarily on company materials and republished funding coverage, with no named competitor set or independent public comparison data confirmed in the source record.

Opportunity

PUBLIC The prize here is not a better scooter camera, but a shot at becoming the safety and compliance layer that shared micromobility operators bolt onto fleets already in the field, a position that could matter if cities and operators begin to treat on-vehicle evidence and real-time intervention as standard operating infrastructure rather than optional hardware [nearhuman.co] [GlobeNewswire, October 2026].

The headline opportunity is straightforward. Nearhuman is trying to sell a retrofit, on-device computer vision system for shared e-scooters and e-bikes, which means it does not need the market to wait for new vehicle platforms or a full fleet replacement cycle before deployment [nearhuman.co] [GlobeNewswire, October 2026]. That makes the commercial path more reachable than many mobility hardware bets: the company says Scootrr is designed to fit existing vehicles, process video on the vehicle, redact footage on board, and detect events such as pavement riding, pedestrian close calls, and road hazards in real time, all of which maps directly to operator pain around safety incidents, insurance exposure, and municipal scrutiny [nearhuman.co] [The Manila Times, October 2026].

The upside case rests on whether this can move from pilot hardware to default operating infrastructure. Nearhuman has disclosed a live Bristol pilot and shipped prototypes for operator testing, while SFC Capital has backed the business at pre-seed stage with a £350,000 round announced in October 2026 [GlobeNewswire, October 2026] [StartupMag, October 2026]. None of that proves scaled adoption, but it does show the company has crossed the line from concept to field validation, which is the threshold that matters most in a fleet hardware sale.

Scenario	What happens	Catalyst	Why it's plausible
Retrofit fleet standard	Nearhuman becomes a preferred safety module for shared e-scooter and e-bike operators, sold as an upgrade across vehicles already on the road	A pilot converts into a named operator deployment and establishes a repeatable retrofit playbook	The system is explicitly designed to retrofit existing fleets rather than require new vehicle platforms, and the company says a live Bristol pilot is already running [nearhuman.co] [GlobeNewswire, October 2026]
Compliance infrastructure for cities	Operators adopt Scootrr because councils and city authorities increasingly want auditable safety analytics and privacy-aware incident data	A city or regulator begins requiring stronger evidence on pavement riding, close calls, or hazard patterns in permit processes	Nearhuman positions the product as privacy-first safety analytics for operators and councils, with on-board redaction intended to keep sensitive footage from leaving the vehicle [LinkedIn] [nearhuman.co]
OEM embedded path	A scooter or e-bike manufacturer integrates similar functionality upstream so Nearhuman moves from aftermarket module to built-in safety layer	A pilot with an OEM progresses to integration on new vehicle models	The company says pilots are progressing with operators and OEMs, and its product claim centers on embedded hardware plus on-device AI rather than a cloud-only software layer [nearhuman.co] [GlobeNewswire, October 2026]

What compounding looks like is less about a classic consumer network effect and more about an installed-base flywheel. If one operator can show that real-time detection and onboard redaction improve incident management or city relationships, the next sale should get easier because Nearhuman would have reference deployments, more edge-case video for model tuning, and a clearer proof case for retrofits across adjacent fleets [nearhuman.co] [GlobeNewswire, October 2026]. That flywheel strengthens if councils start to value standardized analytics and if OEM discussions turn into embedded distribution, since either route lowers customer acquisition friction for the next account.

There is also a narrower technical compounding effect. On-device processing at roughly 30 fps and about 2 W of power, if it holds up in production, would matter because low-power edge inference is unusually well matched to shared vehicles that cannot afford heavy compute, constant connectivity, or privacy-sensitive cloud uploads [nearhuman.co]. In practical terms, that means each successful deployment could improve both the model and the go-to-market message: safer operations, lower latency, and fewer data-handling objections in one package.

The size of the win is still hard to anchor because no public market-size figure or direct public comparable was provided in the source set. The cleanest public reference point is the problem statement on Nearhuman's homepage, which cites 30,000-plus injuries across the UK and EU, 1,300 casualties in the UK costing the NHS and insurers roughly £7 million, and 9,425 reported e-scooter accidents in Germany, up 14.1% year over year [nearhuman.co]. If regulators and operators increasingly respond to those pressures with mandatory or de facto standard safety instrumentation, the company could plausibly become acquisition-worthy infrastructure for a larger mobility, telematics, or vehicle-technology buyer, or a meaningful independent fleet-systems vendor, though any valuation attached to that outcome would be a scenario, not a forecast.

The evidence base remains early and mostly company-linked, so the upside case should be read as conditional rather than established. Still, the ingredients that usually need to appear at this stage are present in public view: a specific hardware-software wedge, a retrofit deployment model, a live pilot, operator testing, and initial institutional backing [nearhuman.co] [GlobeNewswire, October 2026] [StartupMag, October 2026]. For a company founded in 2023, that is enough to make the opportunity legible.

Unconfirmed -- This section relies materially on company website claims and company-linked funding coverage, with limited independent corroboration beyond the funding announcement and pilot references [nearhuman.co] [GlobeNewswire, October 2026] [StartupMag, October 2026].

Sources