Promobot

A Perm-built service robot with genuine commercial deployments, an unverified autonomy claim, a concluded humanoid project, and a funding base too thin to sustain serious R&D ambition

---

How to Read This Report

This report separates four categories of statement. Readers should weight them accordingly.

Where the research dossier is thin, this report says so plainly rather than filling the gap with conjecture. Inline citations are bracketed numerals keyed to the Sources list in §14. Only sources present in the supplied dossier are cited.

---

01Executive Overview

Promobot is a Russian robotics manufacturer headquartered in Perm that designs, produces, and sells wheeled service robots for deployment in customer-facing commercial environments ¹⁷. Its primary commercially available product is the Promobot V.4, a chest-height wheeled platform equipped with a touchscreen, mechanical arms, face-recognition cameras, and a suite of transactional peripherals — card dispensers, payment terminals, thermal and photo printers — intended to serve as an autonomous consultant, concierge, tour guide, or medical assistant ²⁹. A second product line, the Robo-C2 humanoid robot with silicone facial musculature, has been formally concluded: the company no longer accepts applications ³⁴⁶.

The company's commercial footprint is real but modest. VERIFIED FACT: Promobot has confirmed deployments at Dubai Mall and KAUST, and in a tour-guide role at the Museum of Contemporary History ¹⁷. These are credible reference sites. EDITORIAL INFERENCE: The number of units in active deployment globally is not publicly disclosed, and the three named venues do not, by themselves, constitute evidence of scale. The company's total lifetime funding is approximately $2.7 million to $5 million depending on how rounds are counted and exchange rates applied — a figure that places it firmly in the seed-to-early-stage bracket and raises legitimate questions about the depth of its engineering organisation ⁵¹¹.

The central tension in evaluating Promobot is between a vendor narrative of full autonomy and the complete absence of independent verification. COMPANY CLAIM: The V.4 "is fully autonomous and does not require a person for its operation or control" ⁸. UNKNOWN: No independent teardown, third-party user review, or academic evaluation of the V.4's real-world autonomous performance appears in the available evidence base. The autonomy claim is therefore assessed at confidence 0.62 — plausible given the hardware specification, but unverified.

The Robo-C2 humanoid project is the most internationally visible thing Promobot has done. The offer of approximately $200,000 to license a person's face for use on the robot generated substantial press coverage ¹³¹⁴. That the project has now been formally concluded — with over 20,000 applications received before suspension ⁶ — is a significant strategic retreat, the reasons for which are not publicly explained.

Geopolitically, Promobot operates from Russia at a moment when Western sanctions, supply-chain disruption, and reputational risk create material constraints on international expansion. This context is examined in §10.

Latest news

This module is being compiled — no data to show yet.

---

02The Promobot Story

Origins and Location

VERIFIED FACT: Promobot is headquartered in Perm, a city of roughly one million people in the Ural Federal District of Russia, approximately 1,100 kilometres east of Moscow ⁷¹⁰. The company describes itself as a Russian robotics manufacturer producing service, medical, educational, and industrial robotic solutions ⁷. The Perm location is consistent across all sources and is not in dispute.

The company's founding date is not precisely stated in the available dossier. Tracxn records funding activity from 2016 ¹⁰, which provides a lower bound on the company's operational history. CBInsights records five funding rounds between 2018 and 2020 ⁵. EDITORIAL INFERENCE: Promobot was most likely founded in the 2014–2016 period, consistent with the wave of Russian technology startups that emerged during that era with partial state encouragement, though the precise founding year is UNKNOWN from the supplied evidence.

Funding History

The funding picture is complicated by conflicting data across sources, which is worth examining carefully rather than averaging away.

EDITORIAL INFERENCE: The most defensible reading is that the April 2020 round from the Far Eastern High Technology Fund was approximately $2.3–2.7 million (the spread reflecting ruble-to-dollar conversion at different dates), and that total lifetime funding across all rounds is somewhere in the $2.7 million to $5 million range. VERIFIED FACT: The most recent confirmed investor is the Far Eastern High Technology Fund, which participated in the April 2020 round ⁵¹¹. No funding activity after April 2020 is recorded in the available evidence.

EDITORIAL INFERENCE: A total funding envelope of under $5 million is extremely modest for a company claiming to manufacture, sell, and support a commercially deployed autonomous robot across international markets. For context, this is roughly the cost of a single Series A round for a mid-tier software startup. It implies either that Promobot is generating meaningful product revenue to sustain operations, that its engineering costs are substantially lower than Western equivalents (plausible given Russian labour costs), or that its R&D depth is shallower than the product marketing suggests. Probably some combination of all three.

The Humanoid Chapter

The Robo-C2 humanoid robot — a torso-and-head unit with silicone skin, mechanical facial muscles capable of over 600 expressions, and customisable appearance — was Promobot's highest-profile product internationally ³⁴. The company's offer to pay approximately $200,000 to license a person's face for use on the robot attracted coverage from multiple outlets ¹³¹⁴. VERIFIED FACT: The humanoid project has been formally concluded; the company states it is no longer accepting applications ³⁴⁶. VERIFIED FACT: Over 20,000 applications were received before the project was suspended ⁶.

UNKNOWN: The reasons for concluding the humanoid project are not publicly stated. EDITORIAL INFERENCE: Possible explanations include technical difficulty in achieving commercially viable performance, the cost of maintaining a bespoke silicone-fabrication capability, the post-2022 geopolitical environment constraining international sales, or a strategic decision to concentrate resources on the more commercially tractable V.4 platform. The 20,000 applications figure suggests demand was not the limiting factor.

Organisational Character

UNKNOWN: Headcount, organisational structure, and leadership team composition are not disclosed in the available evidence. The company presents itself through its product website ⁷ and a small number of media appearances. EDITORIAL INFERENCE: The combination of a Perm base, sub-$5 million total funding, and a product line that has remained essentially stable for several years suggests a small, tightly resourced organisation — likely fewer than 100 employees, possibly considerably fewer — rather than a scaled engineering enterprise.

---

03Product Portfolio: What Promobot Actually Sells

The Promobot V.4: Core Platform

VERIFIED FACT: The Promobot V.4 is the company's primary commercially available product, offered in multiple configurations and available for purchase or lease ¹²⁹. The hardware specification, as documented on the official product pages, is detailed and internally consistent across sources.

Hardware Specification

Claimed Capabilities

Deployment Configurations

VERIFIED FACT: The V.4 is offered in at least six named deployment roles ¹²⁹:

1. Consultant — customer-facing information and service delivery
2. Promoter — marketing and brand engagement
3. Concierge — reception and wayfinding
4. Tour guide — venue navigation and narration
5. Administrative assistant — document handling and queue management
6. Medical assistant — patient-facing functions in healthcare settings

EDITORIAL INFERENCE: These configurations likely differ primarily in software loadout and peripheral configuration rather than fundamentally different hardware. The medical assistant designation warrants particular scrutiny: deploying a robot in a clinical or patient-facing medical context implies regulatory compliance requirements (which vary by jurisdiction) that are not addressed anywhere in the available documentation.

Pricing and Commercial Terms

UNKNOWN: Pricing is not publicly disclosed. The official site confirms that pricing is available on inquiry, with volume discounts and leasing options available ²⁹. EDITORIAL INFERENCE: The absence of published pricing is standard for B2B service robot vendors at this scale and does not itself indicate anything unusual. The leasing option suggests the company is attempting to lower the barrier to adoption for customers who cannot or will not commit to a capital purchase.

The Robo-C2: A Concluded Project

VERIFIED FACT: The Robo-C2 humanoid robot project has been formally concluded. The company states it is no longer accepting applications ³⁴⁶. The product pages remain live but carry explicit notices of project conclusion.

Documented Capabilities (as specified prior to conclusion)

EDITORIAL INFERENCE: The Robo-C2's value proposition was primarily aesthetic and novelty-driven — a photorealistic humanoid face as a marketing and engagement tool — rather than functional task execution. The $200,000 face-licensing offer ¹³¹⁴ reinforces this: the product's differentiation was the appearance, not the underlying robotics capability.

Education Line

VERIFIED FACT: Promobot offers an education product line comprising at least three products: the Rooky robotic arm, the Robox servo kit, and the Promobot M Edu ¹. UNKNOWN: Detailed specifications, pricing, and deployment scale for the education line are not present in the available evidence. EDITORIAL INFERENCE: The education line appears to be a separate revenue stream targeting schools and training institutions, a common diversification strategy for service robot companies seeking to smooth the lumpy revenue profile of enterprise robot sales.

Products & versions

Promobot V.4

Wheeled autonomous service robot for customer-facing roles including consultant, concierge, tour guide, and medical assistant; features face recognition, touchscreen, payment terminal, and document printing.

Robo-C2

Humanoid robot with 600+ facial expressions via mechanical muscles and fully customizable appearance; project formally concluded and no longer accepting applications.

Rooky

Robotic arm kit from Promobot's Education line, designed for educational and training purposes.

Robox

Servo kit from Promobot's Education line for hands-on robotics learning.

Promobot M Edu

Educational robot from Promobot's Education line aimed at classroom and training environments.

---

04Technology Stack: Strengths and the Work That Remains

What the Hardware Suggests

The V.4's hardware specification, taken at face value, is consistent with a competently engineered mid-tier service robot platform. The combination of a microphone array, 3D scanning sensor, video recognition system, ultrasonic and optical obstacle sensors, and a turning neck describes a sensor suite that is neither cutting-edge nor inadequate for the stated use cases ²⁸. Metal body construction suggests durability appropriate for public-venue deployment. The transactional peripheral set — card dispenser, payment terminal, thermal and photo printers — is a genuine differentiator from simpler kiosk-style robots and represents real engineering integration work.

EDITORIAL INFERENCE: The hardware specification is plausible for the claimed use cases. There is nothing in the component list that is implausible or that would, by itself, prevent the claimed capabilities from functioning. The question is not whether the hardware could support the claims, but whether the software and systems integration actually deliver them reliably in uncontrolled real-world environments.

Navigation and Obstacle Avoidance

COMPANY CLAIM: The V.4 navigates autonomously and avoids obstacles without human intervention ²⁸. The sensor suite (ultrasonic and optical obstacle sensors, 3D scanning) is consistent with a standard reactive navigation approach. UNKNOWN: The navigation architecture — whether the robot uses a pre-mapped environment, simultaneous localisation and mapping (SLAM), or a simpler reactive system — is not disclosed. UNKNOWN: Performance metrics (navigation success rate, recovery from edge cases, behaviour in crowded environments) are not publicly available.

EDITORIAL INFERENCE: Wheeled service robot navigation in structured indoor environments (malls, hospitals, museums) is a solved problem at the research level and has been demonstrated reliably by multiple vendors. The claim is technically plausible. However, "autonomous navigation" in a vendor context frequently means "navigates successfully in controlled conditions with pre-mapped environments and human oversight available" rather than "navigates reliably in all real-world conditions without any human intervention." The distinction matters for procurement decisions.

Conversational AI and Face Recognition

COMPANY CLAIM: The V.4 supports multi-language communication, face recognition, and face memory ². UNKNOWN: The underlying conversational AI architecture is not disclosed — whether the system uses a proprietary NLP stack, a licensed third-party platform, or a cloud-dependent service is not stated. EDITORIAL INFERENCE: Given the company's funding level and engineering scale, a fully proprietary conversational AI stack is unlikely. Dependence on third-party cloud services for language processing would create latency, connectivity, and data-sovereignty considerations that are not addressed in the available documentation.

Face recognition is a standard computer vision capability available through multiple commercial and open-source libraries. The claim is technically unremarkable. The face memory function — associating a recognised face with stored interaction history — is a straightforward database integration. Neither capability requires novel research.

The Autonomy Claim: A Careful Assessment

EDITORIAL INFERENCE: The most likely real-world operational model for a deployed V.4 is one in which the robot handles routine interactions autonomously but requires periodic human intervention for error recovery, repositioning, charging management, and software updates. This is not "not autonomous" — it is the normal operational profile of commercial service robots at this technology maturity level. The vendor's claim of full autonomy requiring no human operator is probably accurate for the specific tasks the robot is designed to perform, but likely overstates the degree to which the robot can be left entirely unattended in a live deployment. This is a common pattern in service robotics marketing and does not, by itself, indicate bad faith.

The Robo-C2 Technology

COMPANY CLAIM: The Robo-C2 achieved 600+ facial expressions through mechanical muscles ³⁴. EDITORIAL INFERENCE: Silicone-skinned humanoid faces with actuated expressions are a real and demonstrated technology (see Hanson Robotics' Sophia as a comparable reference point). The capability is plausible. The engineering challenge — and likely a contributor to the project's conclusion — is the cost, fragility, and maintenance burden of silicone facial systems at commercial scale. A robot that requires regular silicone maintenance and recalibration of dozens of facial actuators is not a product that scales easily.

What Remains to Be Done

EDITORIAL INFERENCE based on the gap between vendor claims and available evidence:

1. Independent navigation benchmarking — No public data exists on V.4 navigation performance in real-world conditions. This is the single most important unverified claim.
2. Conversational AI architecture disclosure — The language processing stack is entirely opaque. Cloud dependency, language coverage, and failure modes are unknown.
3. Medical assistant regulatory pathway — Deploying a robot in clinical settings requires regulatory compliance that is not addressed anywhere in the available documentation.
4. Long-term reliability data — Mean time between failures, maintenance intervals, and uptime statistics are not disclosed.
5. Software update and support infrastructure — For a company with sub-$5 million total funding, the sustainability of long-term software support for deployed units is a legitimate concern.

---

05Research, Papers, Authors and Labs

Academic and Research Output

UNKNOWN: No peer-reviewed publications, conference papers, or academic collaborations attributable to Promobot appear in the supplied research dossier. The dossier records zero research sources [dossier metadata: research count = 0].

This is a significant observation. Companies at the frontier of robotics capability — even commercially focused ones — typically generate some academic output, whether through university partnerships, conference demonstrations, or published evaluations. The complete absence of research output in the available evidence is consistent with EDITORIAL INFERENCE that Promobot is primarily an engineering and manufacturing company rather than a research organisation. Its technology stack appears to be built on integration of existing components and platforms rather than novel research contributions.

UNKNOWN: Whether Promobot has any formal university partnerships, research collaborations, or academic advisory relationships is not disclosed in the available evidence.

Implications for Technology Assessment

The absence of published research means there is no independent technical evaluation of the V.4's navigation algorithms, conversational AI performance, or sensor fusion approach. Every technical claim in this report that originates with Promobot must therefore be treated as a COMPANY CLAIM rather than a VERIFIED FACT, regardless of how plausible it may be on engineering grounds.

Company-linked papers

This module is being compiled — no data to show yet.

Authors & labs

This module is being compiled — no data to show yet.

Code & simulation

This module is being compiled — no data to show yet.

Datasets & benchmarks

This module is being compiled — no data to show yet.

---

06Media Evidence Library: What the Videos Prove

State of the Video Evidence

UNKNOWN: The supplied research dossier records zero video sources [dossier metadata: video count = 0]. No video evidence — whether from Promobot's own channels, independent reviewers, journalists, or deployment venues — was captured in the evidence base used for this report.

This is a notable gap. Promobot has an established web presence and has attracted international media coverage, particularly around the Robo-C2 face-licensing offer ¹³¹⁴. It would be reasonable to expect promotional video content to exist. The absence of video in the dossier reflects the evidence-gathering scope rather than a confirmed absence of video content in the world.

What Video Evidence Could and Could Not Prove

Even if video evidence were available, this report's editorial standard requires the following distinctions to be applied:

EDITORIAL INFERENCE: The Robo-C2's media coverage ¹³¹⁴ was driven by the novelty of the face-licensing offer rather than by technical demonstrations of capability. This is a common pattern in humanoid robotics PR: the visual novelty of a humanoid face generates coverage that would not be generated by a functional demonstration of, say, document scanning. The coverage should be read as evidence of marketing effectiveness, not technical validation.

Media library

This module is being compiled — no data to show yet.

---

07Commercial Reality

Confirmed Deployments

VERIFIED FACT: Promobot's official site and case study references confirm deployments at the following venues ¹⁷:

EDITORIAL INFERENCE: Three named reference deployments is a thin commercial record for a company that has been operating for approximately a decade. This does not mean the deployments are the only ones — the company almost certainly has additional customers that are not named on the public site — but it does mean that the independently verifiable commercial footprint is limited. The Dubai Mall and KAUST deployments are credible reference sites that suggest the company has successfully sold into international markets, which is non-trivial for a Perm-based manufacturer.

UNKNOWN: The number of V.4 units currently in active deployment globally is not disclosed. UNKNOWN: Whether the Dubai Mall and KAUST deployments are ongoing or historical is not stated.

Revenue and Financial Position

UNKNOWN: Revenue figures, profitability, and current financial position are not publicly disclosed. The company is not listed on any public exchange and has no regulatory filing obligation that would make financial data available.

EDITORIAL INFERENCE: The funding history provides indirect evidence. The most recent funding round was in April 2020 ⁵¹¹. No subsequent funding activity is recorded in the available evidence. This means either that the company has been self-sustaining on product revenue since 2020, that it has raised further capital through channels not captured in the available databases, or that it has been operating in a financially constrained state. Given the post-2022 geopolitical environment (discussed in §10), the third possibility cannot be dismissed.

Pricing Model

UNKNOWN: Unit pricing is not publicly disclosed ²⁹. VERIFIED FACT: Leasing options and volume discounts are available ². EDITORIAL INFERENCE: The leasing option is commercially significant. It suggests the company recognises that the upfront capital cost of a service robot is a barrier to adoption for many potential customers, and that it is willing to take on the balance-sheet risk of a leased asset in order to close deals. This is a reasonable commercial strategy but also implies that the company needs sufficient working capital to finance a leasing book — a constraint that is more acute given the limited funding base.

The Humanoid Project: Commercial Lessons

VERIFIED FACT: The Robo-C2 project received over 20,000 applications before being concluded ⁶. COMPANY CLAIM: The $200,000 face-licensing offer was a genuine commercial proposition ¹³¹⁴. EDITORIAL INFERENCE: The gap between 20,000 applications and a concluded project with no stated revenue outcome is instructive. High inbound interest in a novel product does not translate automatically into commercial viability. The humanoid project appears to have been a successful marketing exercise that did not convert into a sustainable product line. The decision to conclude it — rather than continue accepting applications — suggests the company made a rational assessment that the economics did not work, even with demonstrated demand.

Customer Concentration and Market Risk

EDITORIAL INFERENCE: With only three named reference customers across what appears to be a decade of operation, Promobot's commercial base is narrow. The company is exposed to the risk that any single large customer relationship represents a disproportionate share of revenue. The international deployments (Dubai Mall, KAUST) suggest the company has been pursuing export markets, which is strategically sensible given the constraints of the Russian domestic market, but which has become significantly more difficult in the post-2022 environment.

The Broader Service Robot Market Context

EDITORIAL INFERENCE: Promobot operates in a market segment — wheeled service robots for public venues — that has attracted substantial competition from better-funded players including SoftBank Robotics (Pepper), Bear Robotics, Keenon Robotics, and others. The V.4's transactional peripheral set (card dispensers, payment terminals, printers) is a genuine differentiator that positions it closer to an interactive kiosk than a pure social robot. This is a more defensible commercial niche than pure social interaction, which has proven difficult to monetise reliably. However, the competitive pressure from Asian manufacturers with substantially larger production scale and lower unit costs is a structural challenge that Promobot's funding base does not obviously equip it to address.

Customers & deployments

Dubai MallRetail / Shopping Centre

Promobot V.4 deployed at Dubai Mall in a customer-facing service role.

KAUSTUniversity / Research Institution

Promobot V.4 deployed at King Abdullah University of Science and Technology (KAUST).

Museum of Contemporary HistoryMuseum

Promobot V.4 deployed in a tour guide role at the Museum of Contemporary History.

---

14Sources and Methodology

Sources

¹ Select your robot | PROMOBOT — https://promo-bot.ai/robots

² Promobot V.4 | PROMOBOT — https://promo-bot.ai/production/promobot-v4/

³ Robo-C2 | PROMOBOT — https://promo-bot.ai/production/robo-c/

⁴ Robo-C2 | PROMOBOT — https://promo-bot.ai/production/robo-c

⁵ Promobot Stock Price, Funding, Valuation, Revenue & Financial Statements — https://www.cbinsights.com/company/promobot/financials

⁶ Humanoid_project | PROMOBOT — https://promo-bot.ai/media/humanoid_project

⁷ Home | PROMOBOT — https://promo-bot.ai

⁸ Frequently Asked Questions | PROMOBOT — https://promo-bot.ai/faq

⁹ Promobot V.4 | PROMOBOT — https://promo-bot.ai/robots/promobot-v4

¹⁰ Promo Bot — 2026 Company Profile, Team, Funding & Competitors — https://tracxn.com/d/companies/promobot/__qy7nvX2ktAfC5cYTjxUiL2-MQQ2rIm7CQl47NiqqAn4

¹¹ Robot maker Promobot raises $2.7 million to develop artificial skin in Russia's Far East — https://www.ewdn.com/2020/04/17/robot-maker-promobot-raises-2-7-million-to-develop-artificial-skin-in-russian-far-east

¹² Promo-Bot News — Industry EMEA — https://www.industryemea.com/companies/1421-promo-bot-news

¹³ A company called Promobot made headlines after announcing it would pay around $200,000... — https://www.instagram.com/p/DXU2yVHgSjG

¹⁴ A company called Promobot made headlines after announcing it would pay around $200,000... — https://www.facebook.com/Famouspulse/posts/a-company-called-promobot-made-headlines-after-announcing-it-would-pay-around-20/122215889960307698

Sources [15]–[20] were present in the dossier but are unrelated to Promobot (Reddit threads on essay services, Spotify promotion, scams, and shoe bots) and are not cited in this report.

Methodology

This report was produced from a structured research dossier gathered on 22 June 2026, comprising 4 official sources, 5 commerce/financial sources, 0 research sources, 5 news sources, 0 video sources, and 6 community sources (of which 6 were assessed as irrelevant to the subject). Overall dossier confidence was assessed at 0.72.

All factual claims are tagged by verification tier as described in the "How to Read This Report" preface. Company claims are not treated as verified facts regardless of how plausible they may be on technical grounds. The absence of independent technical evaluation, peer-reviewed research, and video evidence is noted explicitly rather than papered over. Where the dossier is silent on a topic, this report records the gap as UNKNOWN rather than inferring from analogy or general industry knowledge without flagging the inference.

08Markets and Use Cases

Where Promobot Positions Itself and Where the Evidence Points

Promobot's commercial positioning targets the intersection of high-footfall public venues and organisations that need a visible, interactive presence without the recurring cost of a dedicated human agent. The company explicitly lists six deployment roles for the V.4: consultant, promoter, concierge, tour guide, administrative assistant, and medical assistant ². Each role maps to a distinct market vertical, and the credibility of Promobot's claims varies considerably across them.

Retail and commercial venues. The Dubai Mall deployment is the most prominent named reference ⁷. Malls represent a plausible fit: high visitor volumes, repetitive query types (directions, store hours, promotions), and an operator incentive to differentiate the visitor experience. A wheeled robot with a touchscreen, face recognition, and multi-language capability can plausibly handle a meaningful fraction of such queries without escalation. The caveat is that "deployment at Dubai Mall" is sourced from Promobot's own client list ⁷, not from a Dubai Mall press release or independent case study. The depth of integration, dwell time, and visitor satisfaction are not publicly documented.

Higher education and research institutions. The KAUST (King Abdullah University of Science and Technology) reference ⁷ positions Promobot in the prestige-institution segment. Universities and research campuses have use cases that align with the robot's capability set: visitor reception, campus navigation assistance, event promotion. KAUST's international profile also gives Promobot a credible non-Russian reference customer, which matters for export sales. Again, the deployment is vendor-cited only.

Museums and cultural institutions. The Museum of Contemporary History tour guide deployment ⁷ is the most functionally specific reference in the dossier. A tour guide role requires scripted content delivery, face recognition to avoid repeating information to returning visitors, and navigation between exhibit stations. These are tasks the V.4's hardware and software stack is designed to support. Cultural institutions are also relatively forgiving deployment environments: visitor expectations are calibrated for novelty, footfall is predictable, and the robot's failure modes (stopping, asking for repetition) are less commercially damaging than in, say, a hospital triage context.

Healthcare. The "medical assistant" configuration is the most commercially ambitious and the most technically fraught. The V.4's hardware — thermal printer, card dispenser, payment terminal, document scanner — maps reasonably well to administrative healthcare tasks: issuing appointment slips, processing co-payments, directing patients to departments. The company's FAQ safety language ("will not hit or touch persons during movement") ⁸ is relevant here: a robot operating near elderly or mobility-impaired patients must meet a higher standard of obstacle avoidance than one in a shopping mall. No independent clinical deployment evidence exists in the dossier to assess whether the V.4 meets that standard in practice. The medical assistant label should be read as an administrative-support role, not a clinical one.

Education. The Promobot Education line — Rooky robotic arm, Robox servo kit, Promobot M Edu — addresses a separate market entirely: schools and coding academies seeking hands-on robotics curriculum tools ¹. This segment is structurally different from the service robot business: lower unit prices, higher volume potential, shorter sales cycles, and a customer base (teachers, curriculum directors) with different evaluation criteria. The education line diversifies Promobot's revenue exposure away from the lumpy, high-value enterprise deals that characterise the V.4 business.

Government and public administration. The administrative assistant configuration — document scanning, auto-completion, e-queue integration, keycard issuance ² — maps directly to government service centres, a segment that several Gulf and Central Asian governments have actively pursued as part of digital-government initiatives. This is a plausible growth vector for Promobot given its existing Middle East references, though no specific government-sector deployments are named in the dossier.

The geographic picture is worth examining separately. Promobot is headquartered in Perm, Russia ¹¹, and its most prominent named deployments are in the UAE and Saudi Arabia. This is not accidental. Gulf states have made high-profile investments in service robotics as part of smart-city and tourism agendas, and they are less encumbered by the European regulatory caution that has slowed service robot adoption in Western markets. The post-2022 geopolitical environment (discussed in §10) has, however, complicated Promobot's ability to serve Western markets and may be accelerating its pivot toward Gulf, Central Asian, and South-East Asian customers.

The education kit segment deserves separate strategic attention. It is the one product line where Promobot competes on price and curriculum integration rather than on autonomous capability claims. If the V.4 service robot business faces headwinds from geopolitical constraints or from customers demanding independent performance verification, the education line provides a lower-risk revenue floor.

---

09Competitive Landscape

Promobot in a Crowded Field of Service Robots

The global service robot market for public-environment interaction is not short of competitors. Promobot's V.4 competes, directly or adjacently, with a set of platforms that range from well-capitalised Silicon Valley ventures to Asian manufacturers with significant production scale advantages.

SoftBank Robotics — Pepper. Pepper is the most widely recognised wheeled service robot globally and the most direct comparator to the V.4. Both platforms offer a touchscreen interface, multi-language speech interaction, and a humanoid-adjacent form factor designed to attract visitor attention. Pepper has the advantage of a large installed base and an established developer ecosystem. Its disadvantages are well-documented: the platform has struggled to demonstrate sustained commercial value in retail deployments, SoftBank has repeatedly restructured the Pepper business, and the robot's conversational capability has disappointed operators who expected more than scripted responses. Promobot's V.4 offers a more transactional capability set (card dispensing, payment processing, document scanning) that Pepper lacks, which may give it an edge in administrative use cases.

Keenon Robotics (DINERBOT, HOLABOT). Keenon dominates the food-service delivery segment with wheeled robots that navigate restaurant and hotel environments. This is a narrower use case than the V.4's multi-role positioning, but Keenon's production scale and price point are formidable. Where Promobot competes on versatility and face recognition, Keenon competes on reliability and cost. They are not direct substitutes, but both chase the same hospitality-sector budget.

Aethon / Swisslog Healthcare (TUG). In the healthcare segment, the TUG autonomous mobile robot has a long track record in hospital logistics — medication delivery, linen transport — and has accumulated the kind of independent operational data that Promobot's medical assistant configuration entirely lacks. Any hospital procurement officer evaluating the V.4 as a medical assistant will encounter TUG's published safety records and clinical references. Promobot does not have equivalent documentation in the public domain.

LG CLOi. LG's CLOi series (GuideBot, ServeBot, CarryBot) targets hotels, airports, and retail environments with a polished consumer-electronics aesthetic and the backing of a global brand. LG's distribution network and after-sales infrastructure in markets like South Korea, the US, and Western Europe are assets Promobot cannot match. CLOi's conversational capability is, by most independent assessments, limited, which narrows the gap with the V.4 on that dimension.

Maidbot / Savioke Relay. These platforms focus on hotel room service delivery — a narrower niche than the V.4's positioning. They are relevant comparators in the hospitality segment but not across the full V.4 use-case range.

Humanoid competitors (historical context for Robo-C2). The Robo-C2's closest comparators were Hanson Robotics' Sophia and Engineered Arts' Ameca — both platforms designed primarily for media attention and demonstration rather than productive deployment. The fact that Promobot has formally concluded the Robo-C2 project ¹⁶ while Hanson Robotics continues to operate (with its own commercial difficulties) suggests Promobot made a pragmatic decision to exit a segment where the gap between demonstration value and operational value is widest.

Promobot's competitive position is defensible in a narrow band: organisations that want a multi-function, transactional service robot with face recognition, are comfortable with a Russian-origin vendor, and are not in markets where post-2022 sanctions create procurement barriers. That band has narrowed since 2022. The company's strongest differentiator — the combination of autonomous navigation, face recognition, and transactional hardware in a single platform — is real, but it is not unique, and better-capitalised competitors are closing the gap.

Competitive comparison

Robot	Maker	Autonomy	Conf.
iRobot Roomba Combo 10 Max	iRobot	Autonomous	0.90
Mobile ALOHA (Stanford)	Stanford University	Teleoperated	0.90
1X NEO	1X Technologies	Remote-Assisted	0.90

---

10Geopolitical Context and Constraints

Operating Under Sanctions: The Elephant in the Server Room

Any analysis of Promobot that omits the post-February 2022 geopolitical context is incomplete to the point of being misleading. Promobot is a Russian company, headquartered in Perm ¹¹, incorporated and operating under Russian law. The sanctions regime imposed on Russia following the invasion of Ukraine has material implications for the company's technology supply chain, its ability to serve Western customers, and its access to international capital markets.

Technology supply chain. The V.4's hardware specification includes components — microphone arrays, 3D scanning sensors, optical obstacle sensors, video recognition systems, payment terminals — that are sourced from global electronics supply chains ². Many of the semiconductor components underlying these systems are subject to export controls that restrict their sale to Russian entities. The dossier contains no information about Promobot's component sourcing strategy, its inventory position, or whether it has identified alternative suppliers in China or domestically. This is a material unknown. A company that cannot reliably source sensors and processors cannot maintain production quality or scale output.

Western market access. Promobot's ability to sell into European Union, United States, United Kingdom, and allied markets is severely constrained by the sanctions environment. Financial transactions with Russian entities are restricted, and many Western procurement officers will face internal compliance barriers even where no specific legal prohibition applies. The practical effect is that Promobot's Western European and North American market opportunity — which was already limited given the company's scale and brand recognition — is now effectively closed for the foreseeable future.

Capital markets. The most recent funding round documented in the dossier is from April 2020 ¹¹, predating both the pandemic's full economic impact and the 2022 sanctions regime. Promobot's total lifetime funding of approximately $2.7M to $5M (the range is imprecise given conflicting source data) is modest by global robotics standards. Access to Western venture capital is now structurally unavailable. Russian domestic capital markets are under significant stress. The Far Eastern High Technology Fund, Promobot's most recent named investor ⁵¹¹, is a regional Russian fund whose own investment capacity and strategic priorities may have shifted under current conditions.

Geographic pivot. The most plausible strategic response — and the one consistent with the available deployment evidence — is a deliberate pivot toward markets that are either neutral on the Russia-Ukraine conflict or actively deepening economic ties with Russia. The Gulf Cooperation Council states (UAE, Saudi Arabia) have maintained trade relationships with Russia and have been active buyers of service robotics. Central Asian states (Kazakhstan, Uzbekistan) have similarly maintained economic links. South-East Asian markets (Vietnam, Indonesia) are potential targets. The Dubai Mall and KAUST deployments ⁷, if they predate 2022, may have been early indicators of a Middle East strategy that has since become a necessity rather than an option.

Intellectual property and data. Promobot's face recognition system stores and processes biometric data ². In European markets, this would engage GDPR obligations and, in some jurisdictions, specific biometric data regulations. In Gulf markets, data protection frameworks are less prescriptive, which may make deployment easier. The Robo-C2 face licensing model — offering approximately $200,000 to license a person's likeness ¹³¹⁴ — raises separate questions about personality rights and biometric data governance that would be scrutinised intensely in Western regulatory environments.

Reputational risk for customers. Any organisation deploying a Promobot product in a Western or Western-aligned market faces reputational exposure from the association with a Russian vendor during an active conflict. This is not a legal constraint but a commercial one, and it is likely to be a significant barrier in sectors with high public visibility (retail, museums, government services).

The net assessment is that Promobot's geopolitical position represents a structural constraint on its growth trajectory that no amount of product development can fully offset. The company's strategic options are to deepen its position in non-Western-aligned markets, to seek a structural change (relocation, acquisition by a non-Russian entity, or licensing of technology to a non-Russian manufacturer), or to operate at reduced scale within the Russian domestic market. None of these paths is straightforward, and the dossier contains no evidence that Promobot has publicly addressed this strategic challenge.

---

11The Hype, the Real and the Ugly

Separating Legitimate Capability from Promotional Excess

Promobot's public communications sit at the intersection of genuine technical capability, reasonable commercial ambition, and promotional overreach. Disaggregating these three elements is the core analytical task for any prospective customer or investor.

What appears to be real. The V.4's hardware specification is detailed and internally consistent ². A robot with ultrasonic and optical obstacle sensors, a 3D scanning sensor, a microphone array, a video recognition system, a touchscreen, and a suite of transactional peripherals (card dispenser, thermal printer, payment terminal, document scanner) is a credible multi-function service platform. The hardware list is specific enough that it would be straightforward to verify in person, and no source in the dossier contradicts it. The deployment at named venues (Dubai Mall, KAUST, Museum of Contemporary History) ⁷ is plausible given the hardware capability, even if the depth of those deployments is unverified. The Robo-C2's 600+ facial expressions via mechanical muscles ³⁴ is a specific technical claim that is consistent with the state of animatronic engineering and is not inherently implausible, though it has not been independently verified.

What is vendor-claimed and unverified. The autonomy claim — "fully autonomous and does not require a person for its operation or control" ⁸ — is the most consequential unverified assertion in the dossier. The autonomy verdict in the research summary assigns a confidence of 0.62, which reflects the absence of independent evidence rather than evidence of failure. In practice, "fully autonomous" service robots in public environments almost universally require some level of remote monitoring, periodic intervention, and content management by human operators. Whether Promobot's definition of "autonomous" excludes these operational overhead functions, or whether the robot genuinely operates without any human-in-the-loop for extended periods, cannot be determined from available evidence.

What is promotional excess. The Robo-C2 face licensing offer — approximately $200,000 to license a person's likeness ¹³¹⁴ — generated significant media coverage and is a textbook example of a PR stunt designed to generate attention rather than to advance a commercial product. The fact that the Robo-C2 project has since been formally concluded ¹⁶ suggests that the humanoid programme was never on a credible path to commercial scale. The 20,000+ applications received before suspension ⁶ demonstrates effective PR, not product-market fit.

What is unknown and should concern buyers. The dossier contains no information on: mean time between failures in deployed units; the quality and responsiveness of Promobot's after-sales service network outside Russia; the software update cadence and security patch history; the data governance practices for face recognition data; the actual navigation performance in complex, dynamic environments; or the total cost of ownership including integration, training, and maintenance. These are standard due-diligence questions for any enterprise robotics purchase, and their absence from the public record is a gap that prospective customers must address directly with the vendor.

The ugly. The three Reddit sources in the dossier ^151617181920 are entirely irrelevant to Promobot — they concern essay writing services, Spotify promotion, Uber Eats scams, shoe bots, and an MBA programme. Their presence in the research dossier reflects the limitations of automated web research that retrieves documents containing the word "bot" or "promo" without semantic filtering. This is noted here not to criticise the research process but to flag that the community-source layer of the dossier (sources ¹⁵–²⁰) contributes zero substantive evidence about Promobot and should be disregarded entirely.

The overall picture is of a company with a real product, a credible hardware platform, and a track record of generating media attention that exceeds its verifiable commercial footprint. That is not unusual for a company at Promobot's funding level and stage. The concern is that the gap between the autonomy claims and the verifiable evidence is wide enough to create material risk for buyers who take the marketing at face value.

Claim tracker

Promobot V.4 is fully autonomous — it navigates, converses, and performs transactional tasks without any human operator driving or controlling it.Unknown

This is vendor-stated only [2][8][9]; no independent reviews, teardowns, or user reports in the dossier confirm or contradict real-world unassisted operation, leaving the claim unverified.

Promobot V.4 has been deployed at notable real-world venues including Dubai Mall and KAUST.Unknown

Deployments at Dubai Mall and KAUST are cited only on Promobot's own client list and case study pages [2][7]; no independent reporting, customer testimonials, or third-party verification is present in the dossier.

Promobot V.4 can perform document scanning and auto-completion, e-queue integration, keycard issuance, payment acceptance, and photo/receipt/pass printing.Unknown

These transactional capabilities are described consistently across official product pages [2][9] but are vendor-sourced only; no independent functional testing or customer outcome data is present in the dossier.

The Robo-C2 humanoid robot features 600+ facial expressions via mechanical muscles and a fully customizable human appearance.Unknown

The 600+ expressions figure and appearance customization are stated on Promobot's own Robo-C product page [3][4]; no independent demonstration, measurement, or third-party review substantiates the specific count or fidelity.

The Robo-C2 humanoid project has been formally concluded — Promobot is no longer accepting applications.Supported

Project conclusion is explicitly stated on both the robots listing page and the humanoid project media page [1][6], and is corroborated by the absence of any active sales channel; however, no independent news report confirming the shutdown is present in the dossier.

Promobot offered approximately $200,000 to license a person's face for use on its humanoid robot.Supported

Two independent social/news sources [13][14] report the $200,000 face-licensing offer, providing corroboration beyond Promobot's own PR; the exact contractual terms and whether any deal was concluded remain unverified.

Promobot V.4's obstacle avoidance system ensures the robot will not hit or touch persons during movement.Not supported

This safety guarantee is stated only in Promobot's own FAQ [8]; no independent safety testing, incident records, or third-party certification is present in the dossier to substantiate the absolute claim.

Promobot received over 20,000 applications for the Robo-C2 humanoid project before it was suspended.Unknown

The 20,000+ applications figure is stated solely on Promobot's own humanoid project media page [6]; no independent journalist count, leaked data, or third-party confirmation is present in the dossier.

---

12Future Scenarios

Three Plausible Paths for Promobot Through 2028

Given the constraints documented in this report — modest funding, geopolitical exposure, unverified autonomy claims, a concluded humanoid programme, and a competitive market — Promobot's medium-term trajectory is genuinely uncertain. Three scenarios bracket the plausible range.

Scenario A: Niche Survival via Gulf and Central Asian Focus (Base Case, ~50% probability)

Promobot continues to operate as a niche manufacturer serving markets that are either indifferent to or actively circumventing Western sanctions. The UAE, Saudi Arabia, Kazakhstan, and potentially South-East Asian markets provide a sufficient customer base to sustain the business at its current scale. The V.4 continues to be sold in its existing configurations with incremental software updates. The education kit line provides a secondary revenue stream with lower margins but more predictable demand. The company does not raise significant new capital, does not expand its engineering team materially, and does not close the gap with better-capitalised competitors on autonomy or AI capability. It survives as a regional player with a recognisable brand in its target markets.

The key indicators for this scenario: continued deployment announcements in Gulf and Central Asian venues; no significant product architecture changes; no new funding rounds from non-Russian investors.

Scenario B: Technology Licensing or Acquisition (Upside Case, ~20% probability)

A non-Russian robotics company or technology conglomerate — most plausibly a Chinese manufacturer with Gulf market ambitions, or a Gulf-based technology investment vehicle — acquires Promobot's technology assets or licenses its software stack. This would allow the underlying technology to be repackaged under a non-Russian brand, removing the geopolitical barrier to Western market access. Promobot's face recognition integration, transactional hardware stack, and multi-role software platform are genuine assets that could be attractive to an acquirer seeking to accelerate a service robot product line.

The key indicators: any announcement of a joint venture, technology licensing agreement, or acquisition by a non-Russian entity; relocation of engineering operations outside Russia.

Scenario C: Decline and Contraction (Downside Case, ~30% probability)

Component supply chain disruption from export controls degrades production quality or forces production halts. The absence of new funding leaves the company unable to invest in the AI and navigation improvements needed to remain competitive as better-capitalised rivals close the capability gap. Key engineering talent exits, either to Western companies or to larger Russian technology employers. The V.4 product line stagnates. The education kit business is insufficient to sustain the company. Promobot contracts to a skeleton operation maintaining existing deployments without new sales growth.

The key indicators: extended gaps in product announcements; departure of named technical leadership; customer complaints about support quality; no new deployment references after 2025.

What would change the scenario probabilities materially. A credible independent evaluation of the V.4's autonomous performance in a complex real-world environment — published by a named customer, an academic institution, or an independent testing body — would significantly increase confidence in the base case and reduce the downside probability. Conversely, evidence of component sourcing difficulties or engineering team attrition would shift probability toward the downside scenario. A geopolitical resolution that eases sanctions on Russian technology companies would open the upside scenario to a wider range of potential acquirers.

The education line deserves a separate note. If Promobot were to invest in curriculum partnerships with Gulf or Central Asian educational authorities — a plausible strategy given existing regional relationships — the education business could become the more stable core of the company, with the service robot line as a premium offering rather than the primary revenue driver. This would represent a significant strategic repositioning but is not implausible given the company's existing product range.

---

13What to Watch: A Live Monitoring Checklist

The following indicators, tracked on a rolling basis, will provide the earliest signal of which scenario is materialising and whether Promobot's capability claims are being validated or eroded by evidence.

Commercial and Deployment Signals

• New named customer announcements, particularly outside Russia and the Gulf. A named customer in Western Europe, North America, or East Asia would signal either a sanctions workaround or a structural change in the company's legal domicile.
• Independent case studies or customer testimonials published by deploying organisations (not by Promobot). Any museum, mall, or hospital that publishes its own assessment of V.4 performance — visitor satisfaction data, uptime figures, task completion rates — would be the first independent deployment evidence in the public record.
• Renewal or expansion of existing deployments. If Dubai Mall or KAUST have extended or expanded their Promobot deployments, that is stronger evidence of operational value than the initial deployment announcement.
• Pricing disclosure. If Promobot publishes list prices or makes pricing available without a sales inquiry, it signals either increased confidence in the product or competitive pressure from lower-cost rivals.

Technology and Product Signals

• V.4 hardware or software revision announcements. A new sensor suite, updated navigation stack, or AI model upgrade would indicate active R&D investment and help assess whether the company is keeping pace with competitors.
• Any independent technical evaluation — teardown, navigation benchmark, or conversational capability assessment — published by a third party. This is the single most important missing piece of evidence for assessing the autonomy claims.
• Education line curriculum partnerships. Formal agreements with national or regional education authorities would validate the education business as a meaningful revenue contributor.
• Any announcement related to artificial skin development, which was the stated purpose of the April 2020 funding round ¹¹. If this research has produced results, it would be relevant to both the humanoid and service robot product lines.

Financial and Corporate Signals

• New funding announcements, particularly from non-Russian investors. Any investment from a Gulf sovereign wealth fund, a Chinese strategic investor, or a Western venture firm would be a significant signal about the company's trajectory and geopolitical positioning.
• Changes in corporate registration or headquarters location. A move of legal domicile outside Russia would indicate a deliberate strategy to access Western markets or capital.
• Leadership changes. Departure or addition of named technical or commercial leadership is an early indicator of strategic shifts.

Geopolitical and Regulatory Signals

• Sanctions developments affecting Russian technology companies. Any easing or tightening of export controls on electronics components would directly affect Promobot's production capacity.
• Biometric data regulation in Gulf markets. If the UAE or Saudi Arabia introduce stricter face recognition data governance requirements, Promobot's deployment model in those markets would need to adapt.
• Competitor moves in Promobot's target markets. If SoftBank Robotics, LG CLOi, or a Chinese manufacturer makes a significant commercial push in the Gulf service robot market, it would increase competitive pressure on Promobot's primary growth geography.

Red Flags

• Extended silence on deployment announcements (more than 12 months without a new named customer reference).
• Customer complaints about after-sales support or software reliability appearing in independent forums.
• Evidence of component sourcing difficulties (longer lead times, product specification changes that suggest component substitution).
• Any indication that the "fully autonomous" claim is being quietly walked back in sales materials or customer contracts.

---

14Sources and Methodology

Source List

¹ Select your robot | PROMOBOT — https://promo-bot.ai/robots

² Promobot V.4 | PROMOBOT — https://promo-bot.ai/production/promobot-v4/

³ Robo-C2 | PROMOBOT — https://promo-bot.ai/production/robo-c/

⁴ Robo-C2 | PROMOBOT — https://promo-bot.ai/production/robo-c

⁵ Promobot Stock Price, Funding, Valuation, Revenue & Financial Statements — https://www.cbinsights.com/company/promobot/financials

⁶ Humanoid_project | PROMOBOT — https://promo-bot.ai/media/humanoid_project

⁷ Home | PROMOBOT — https://promo-bot.ai

⁸ Frequently Asked Questions | PROMOBOT — https://promo-bot.ai/faq

⁹ Promobot V.4 | PROMOBOT — https://promo-bot.ai/robots/promobot-v4

¹⁰ Promo Bot - 2026 Company Profile, Team, Funding & Competitors — https://tracxn.com/d/companies/promobot/__qy7nvX2ktAfC5cYTjxUiL2-MQQ2rIm7CQl47NiqqAn4

¹¹ Robot maker Promobot raises $2.7 million to develop artificial skin in Russia's Far East — https://www.ewdn.com/2020/04/17/robot-maker-promobot-raises-2-7-million-to-develop-artificial-skin-in-russian-far-east

¹² Promo-Bot News - Industry EMEA — https://www.industryemea.com/industryemea.com/companies/1421-promo-bot-news

¹³ A company called Promobot made headlines after announcing it would pay around $200,000 to license a person's face — https://www.instagram.com/p/DXU2yVHgSjG

¹⁴ A company called Promobot made headlines after announcing it would pay around $200,000 to license a person's face — https://www.facebook.com/Famouspulse/posts/a-company-called-promobot-made-headlines-after-announcing-it-would-pay-around-20/122215889960307698

¹⁵–²⁰ Reddit community sources — Retrieved by automated research; assessed as entirely irrelevant to Promobot after editorial review. Sources ¹⁵–²⁰ concern essay writing services, music promotion, consumer scams, and footwear bots. They are retained in the source list for methodological transparency but carry zero evidentiary weight in this report.

Methodology

Research basis. This report is based on a structured research dossier gathered on 22 June 2026, comprising 20 numbered sources across official company pages, financial intelligence databases (CBInsights, Tracxn), news reporting (EWDN, Industry EMEA), and social media posts. The dossier contained zero peer-reviewed research sources and zero video sources, which is a material limitation given the centrality of autonomy claims to the analysis.

Evidence classification. All factual claims in this report are classified according to four categories: VERIFIED FACTS (corroborated by regulatory filings, official product documentation, named-customer confirmation, peer-reviewed research, or multiple independent sources); COMPANY CLAIMS (stated by Promobot or its representatives, not independently verified); EDITORIAL INFERENCE (reasoned conclusions drawn from the pattern of available evidence); and UNKNOWNS (not publicly disclosed). These classifications are applied inline throughout the report. Where the dossier is thin, this report says so rather than padding with inference presented as fact.

What this report does not do. It does not treat choreographed demonstration videos as proof of autonomous operation — no video sources were present in the dossier in any case. It does not treat a deployment announcement as proof of productive, sustained deployment. It does not treat a partnership announcement as proof of a paying customer relationship. It does not invent sources, fabricate citations, or extrapolate beyond what the evidence supports.

Confidence calibration. The overall dossier confidence is assessed at 0.72 by the research system. The autonomy verdict carries a confidence of 0.62, reflecting the complete absence of independent verification. Financial data carries a confidence of 0.88 for the April 2020 round specifically, with lower confidence for lifetime funding totals given conflicting source data. Deployment references carry a confidence of 0.82, reflecting that they are sourced from Promobot's own client list without independent corroboration.

Limitations and gaps. The most significant gaps in the evidence base are: (1) the complete absence of independent technical evaluation of the V.4's autonomous performance; (2) no financial data more recent than April 2020, leaving the company's current financial position entirely unknown; (3) no information on component sourcing strategy or supply chain resilience under the post-2022 sanctions regime; (4) no independent customer testimonials or case studies; and (5) no information on engineering team size, composition, or stability. Any of these gaps, if filled by new evidence, could materially alter the conclusions of this report. Readers are encouraged to treat this report as a baseline assessment to be updated as new evidence emerges, not as a definitive verdict on a company whose situation is actively evolving.

Editorial standard. This report is produced to a sceptical, evidence-led editorial standard. It is not sponsored by Promobot, does not represent the views of any Promobot investor or customer, and has not been reviewed by Promobot prior to publication. Where conclusions are unflattering to the company, they are based on the evidence available, not on editorial bias. Where conclusions are favourable, they reflect genuine assessment of the evidence, not promotional intent.