Artifact · v0 (real content)

Positioning Brief

Three positioning wedges, grounded in the competitive set. What SmartOne says, why it is structurally defensible, and where it falls apart if pushed.

1.3 · Positioning and Messaging · artifact id: positioning-brief-v0.html · 2026-05-28 · v0 draft

The articulation gap

"we're highly immature right now as to what we show to the market, and highly immature as to how we've evolved compared to other players in the market." Discovery call, May 15

The gap was named precisely on the call. SmartOne has not changed markets; it has outgrown the description it uses in market. The capabilities are real: temporal annotation, 3D spatial reasoning, trajectory prediction, sensor fusion at scale, ISO 27001, SOC 2 Type II. The vocabulary has not kept up. This brief does not create a new story. It names the one that already exists.

Positioning context: the Meta/Scale inflection

The strategic context for SmartOne's positioning shifted materially in June 2025 when Meta acquired a 49% stake in Scale AI. Four of the largest enterprise AI labs (Google, OpenAI, Microsoft, xAI) reduced or paused Scale relationships because any training data pipeline shared with Scale now flows through an infrastructure where Meta holds board-adjacent influence.

Several companies made the "Scale AI without the conflicts" claim. What separates a defensible claim from a marketing line is structural specificity. SmartOne's position needs to be grounded in verifiable facts, not aspirations. The three wedges below are all structural facts, not aspirations.

Wedge 1: Neutral Physical AI ground truth

The claim

Bootstrapped, conflict-free, Physical AI-depth already in production

SmartOne is the only player in the competitive set that simultaneously holds all four of the following: bootstrapped with no investor conflict, Physical AI capabilities already in production (not announced, not planned), SOC 2 Type II and ISO 27001, and a full-time institutional workforce rather than gig or crowd annotators.

What backs it
  • No investor with a competing AI platform in the shareholder structure
  • Robotics annotation, autonomous truck, drone, and sensor-fusion work confirmed in production portfolio
  • SOC 2 Type II certification led by Jeritiana Ravelojaona, Site Director Madagascar
  • ISO 27001 in place. Both pass the buyer gate that eliminates most emerging-market competitors
  • 1,000 full-time trained employees, not crowd or contractor network
Where it is fragile
  • Amazon at ~60% of revenue is an unresolved dependency that contradicts the "no concentration risk" version of this story
  • The Scale conflict window will not stay open indefinitely. New alternatives are scaling. The neutrality advantage has a 12-18 month half-life without new client evidence
  • Physical AI work exists in the portfolio but is not packaged as named case studies with permission to share. Buyers who ask "show me" cannot be satisfied today
The one-sentence version

"The data annotation firm your competitors cannot use, because they own it."

Wedge 2: Full-time institutional workforce

The claim

Fourteen years of trained full-time annotators, not gig or crowd

The majority of competitors in the data annotation market operate crowd or contractor networks: Appen (200,000+ crowd), Toloka (200,000+ annotators), Surge (50,000 contractors), Sama (contracting model post-crisis). SmartOne runs 1,000 full-time employees in Madagascar who have worked in annotation for years, building institutional knowledge and annotation discipline that crowd workers do not accumulate.

What backs it
  • Full-time employment model since founding. Workers are not gig workers who rotate in and out
  • 98%+ accuracy floor with structured reprocessing below threshold, documented since the first Amazon engagement
  • Ramp capacity: 50% headcount increase in two weeks while maintaining quality, demonstrated to Amazon
  • Digital skills academy in Madagascar building temporal annotation, 3D spatial reasoning, and world-modeling competency specifically for Physical AI demand
  • iMerit is the only other full-time institutional workforce player in the competitive set. iMerit is India-based and Series B-funded; SmartOne's Madagascar model is a structurally distinct cost and quality profile
Where it is fragile
  • Madagascar as a location raises questions from buyers unfamiliar with its labor market. The answer: a decade of proven delivery, not an explanation of why Madagascar is viable
  • The workforce model does not differentiate on price versus crowd competitors at small scale. It differentiates on consistency and accuracy at volume and in high-stakes domains
  • No published accuracy benchmarks or third-party validation. The "98%+" figure is founder-stated; it needs client-attributed backing for a sales context
The one-sentence version

"The annotators who labeled your training data will still be here in three years, remembering what they built."

Wedge 3: French-speaking EU data partner

The claim

ESSEC alumni, French entity, EU AI Act compliance, Palantir DGSI credential

The French enterprise AI market is at a structural inflection: DORA entered full enforcement in January 2025, NIS2 became enforceable in October 2024, and the EU AI Act reaches full application in August 2026. High-risk AI systems require documented data governance that flows directly into procurement: buyers need data vendors that can show provenance, accuracy rates, and audit trails. SmartOne's SOC 2 Type II and ISO 27001 certifications are procurement gates in 2026 that did not exist in 2022 when SmartOne last tried France.

What backs it
  • SMARTONE EU SAS already registered in Paris. Habib has been president of the entity since November 2021. It needs activation, not creation
  • Habib and Shahysta are ESSEC alumni with access to a 71,000-person alumni network, 3,200-member Entrepreneurs Club, and warm paths to BNP Paribas, LVMH, Airbus, and Schneider Electric
  • Palantir DGSI: Palantir renewed its multi-year contract with the French DGSI in December 2025. SmartOne's history annotating Palantir data means SmartOne's work passed Palantir's security and quality requirements. That chain of trust is credible with French defense primes
  • Safran is a current SmartOne client. Safran acquired Preligens (now Safran.AI) and operates a certified annotation center. A reference exists
  • No competitor in the B2 set (Scale AI, Surge, iMerit, Toloka) simultaneously holds a French entity, French-speaking leadership, and French defense client history
Where it is fragile
  • SMARTONE EU SAS has 1,500 euro registered capital, which reads poorly on a K-Bis review by French enterprise procurement teams. Capital increase needed before serious enterprise outreach
  • Website is English-only. French enterprise technology buyers at director level read English, but procurement documentation and RFP responses operate in French. English-only signals that the French market is not taken seriously
  • This wedge is Phase 2. It activates when Phase 1 survival-mode pressure has eased. Pushing it before Phase 1 stability creates a distraction with a 9-15 month time-to-revenue horizon
  • Safran reference needs explicit client permission before use in sales materials. Confirm NDA carve-outs first
The one-sentence version

"The only Physical AI data partner that can walk into a French prime's procurement meeting in French, with a French entity and a French defense credential already on file."

What the positioning work produces

The output from Phase 1 positioning and messaging work is a set of verified, buyer-ready assets: