Persona AI raises $27M for shipyard-purpose humanoids

Lead strip: Status: Research / MOU (shipyard welding path) · Availability: no public catalog SKU; RaaS framed · All-in cues: $27M pre-seed confirmed; list price and fleet hours — · Watchlist: Persona (no model page yet) · Scoring: methodology

Evidence brief only — not a sold ranking. Pre-seed capital and an MOU ≠ a warehouse tote fleet or a shipyard welding certificate.

Houston-based Persona AI closed an oversubscribed $27 million pre-seed in May 2025 and positioned its humanoid program for shipbuilding and heavy industrial work rather than the warehouse tote lane that dominates Digit-class coverage. For buyers, the useful split is capital + partner MOU versus checkable field hours — and how a purpose-built shipyard bet differs from Digit and Figure 03.

Robot strips (context, not a trophy board)

• Digit · Agility Robotics · Status: Digit 4 field / Digit 5 announced · Use: warehouse logistics · Evidence: named sites + published hours/totes · → /robots/digit/
• Figure 03 · Figure AI · Status: Pilot (BMW Spartanburg sequencing) · Use: factory logistics · Evidence: BMW Group partner statement · → /robots/figure-03/
• Persona (watchlist) · Persona AI · Status: Research / MOU · Use: shipyard welding / heavy industry (company framing) · Evidence: $27M pre-seed + HD Hyundai MOU; fleet hours — · → product page — (no /robots/persona/ yet)

What the $27M round documents

Persona’s 14 May 2025 release states it raised $27 million in an oversubscribed pre-seed co-led by Unity Growth and Tides Ventures, with a long participating list including Mirae Asset Group, Invicta Growth, Fathom Fund, Embedded Ventures, and others (PR Newswire / Persona).

The same materials frame the capital as accelerating a humanoid platform for shipbuilding and industrial manufacturing, cite a signed agreement with HD Hyundai for shipyard deployment with initial delivery expected within 18 months of that announcement, and describe a robotics-as-a-service (RaaS) commercial model rather than a retail CapEx SKU.

Independent coverage aligns on the same core facts and adds founder context: Nic Radford (CEO; formerly Nauticus / NASA dexterous robotics), Jerry Pratt (CTO; previously Figure AI), and Jide Akinyode (COO) founded Persona in 2024 (The Robot Report).

Documented: round size, co-leads, shipyard/industrial positioning, RaaS framing, HD Hyundai agreement language in company materials.
Not documented as a buyer receipt: public list price, warranty/SLA package, paid multi-site fleet hours, weld-cycle KPIs, or a walk-up order page. Those stay —.

HD Hyundai MOU: welding path, not a field fleet yet

A related 7 May 2025 release describes an MOU among HD Korea Shipbuilding & Offshore Engineering (HD KSOE), HD Hyundai Robotics, Persona AI, and Vazil Company to develop humanoids for complex shipyard welding. Stated roles: Persona on hardware and AI control; Vazil on welding tools and test environment; HD KSOE on live shipyard deployment support and field data; HD Hyundai Robotics on welding-path AI training data and validation (PR Newswire / HD Hyundai–Persona).

Timeline language in that MOU release: prototype completion targeted for end of 2026; field testing and commercialization framed to begin in 2027.

How to read this under our methodology:

Signal	Evidence class	Buyer takeaway
$27M pre-seed (company PR)	Capital / continuity	Program funded; not a throughput certificate
HD Hyundai / KSOE / Vazil MOU	Partner intent	Named industrial path; still pre-prototype
End-2026 prototype / 2027 field test language	Roadmap	Calendar cue — not logged weld hours
RaaS framing	Commercial model claim	Ask for rate card; public card —
Named shipyard weld KPIs / paid hours	—	Missing vs Digit-class logistics evidence

An MOU plus a delivery window is partner roadmap, not the same evidence lane as Digit’s published operational hours or BMW’s partner-stated Figure 03 sequencing project.

Purpose-built shipyard vs warehouse Digit-class platforms

Persona’s public story is deliberately vertical: welding and physically demanding shipyard work, labor shortages, and safety in heavy industry. That is a different design and evidence problem than tote moving in a structured fulfillment center.

Contrast the lanes carefully:

Platform	Typical 2026 evidence lane	Environment hardness (public framing)	Buy surface
Persona AI (watchlist)	Capital + shipyard welding MOU	Unstructured / hazardous heavy industry	RaaS claimed; catalog —
Digit · Agility Robotics	Named logistics sites; published hours/totes	Warehouse / industrial logistics	Enterprise / RaaS
Figure 03 · Figure AI	BMW Spartanburg sequencing (partner statement)	Automotive factory logistics	Enterprise partnership; consumer list —

Purpose-built ≠ automatically “ahead.” It means the company is optimizing for a different constraint set (welding precision, shipyard access, ruggedization) than Digit’s logistics stack or Figure’s automotive sequencing stack. Until Persona or HD Hyundai publish checkable field metrics, keep Persona in Research / MOU — useful watchlist, not a scored fleet peer.

We do not invent a product page at /robots/persona/ until there is enough public product evidence to support one. Treat Persona as a watchlist name beside Digit and Figure 03 strips.

Capital vs field evidence (same rule as other funding briefs)

Under methodology, funding informs continuity. It does not fill paid deployments, third-party record, or price transparency buckets by itself. The same discipline we apply to Sanctuary-style capital stories applies here: funding ≠ fleet.

Morgan Stanley-style total-addressable-market figures cited in Persona’s release are analyst market sizing, not Persona deployment proof. Interact Analysis-style caution in secondary coverage (slow unit uptake vs hype) is industry context, not a Persona score.

All-in cost cues

Public per-unit list price: —. RaaS rate card: —. Planners evaluating a future shipyard pilot should still budget integration, welding-tooling packages, safety case, supervision, spares, and facility access separately from any informal valuation headlines. Unknown commercials stay —.

Practical takeaway

1. Credit the $27M and the HD Hyundai welding MOU as capital + partner intent — not as logged shipyard hours.
2. Keep purpose-built shipyard evidence separate from warehouse Digit and factory Figure 03 lanes.
3. Watchlist Persona until a public product page, rate card, or partner field metrics appear (no /robots/persona/ yet).
4. Cross-read methodology and the 2026 humanoid evidence shortlist.

FAQ

Is Persona AI’s humanoid available to buy today?

No public catalog SKU, checkout, or list price appears in the May 2025 funding and MOU materials cited here. The company frames a RaaS model and a shipyard partnership path. Availability for buyers: enterprise outreach / future RaaS; public buy surface —.

Does the HD Hyundai deal mean robots are welding in shipyards now?

The MOU release targets prototypes by end of 2026 and field testing / commercialization from 2027. That is roadmap language. Published paid weld hours and cycle KPIs: —.

How is Persona different from Digit or Figure 03?

Digit carries a multi-site warehouse logistics operational narrative. Figure 03 has a BMW Group partner statement on Spartanburg sequencing (Figure 02 ≠ Figure 03). Persona’s public record centers on pre-seed capital and a shipyard welding MOU. Different evidence classes — not a trophy board.

Why is there no /robots/persona/ page?

We only stand up product pages when status, buy surface, and evidence density clear our bar. Persona is on the watchlist until those receipts exist. Funding alone does not create a product href.

Sources

• Persona AI / PR Newswire: $27M oversubscribed pre-seed (14 May 2025)
• Persona AI / PR Newswire: HD Hyundai welding humanoid MOU (7 May 2025)
• The Robot Report: Persona AI raises $27M…
• Robotscomparison: methodology, Digit, Figure 03, 2026 humanoid evidence shortlist

Persona list price, RaaS rate card, and shipyard fleet hours: —.