Why Mark Zuckerberg and Maria Sharapova Are Betting on This AI Health Startup:

AI Meets Preventive Health: Neko's $700M Bet and Google DeepMind's Biosecurity Push:

Google DeepMind’s New Multi-Pillar Strategy to Fight AI Biosecurity Threats:

How AI-powered body scans and biosecurity safeguards are redrawing the line between consumer health tech and enterprise-grade AI governance.

$700M: Neko Health Series C round

~$7B: Reported post-round valuation

15+: DeepMind biosecurity partnerships

1: Neko Health's $700 Million Leap Into US Preventive Care:

Stockholm-based preventive health startup Neko Health has closed a $700 million Series C round to bring its AI-powered full-body scanning clinics to the United States, beginning with a location in New York City.

The round was led by Lightspeed Venture Partners and co-led by O.G. Venture Partners, with existing backers Atomico, General Catalyst, and Lakestar returning alongside new investors including Liberty City Ventures, Positive Sum, and BDT & MSD. O.G. Venture Partners' David Ofer is set to join Neko's board pending regulatory approval. The raise pushes Neko's disclosed funding since 2023 past $1 billion, following a $65 million Series A and a $260 million round in January 2025.

Neko's investor roster reads like a celebrity cap table: Meta CEO Mark Zuckerberg and Priscilla Chan, tennis legend Maria Sharapova, musician will.i.am, and former footballer Thierry Henry all participated, joining existing backers Alexis Ohanian and Zoë Saldaña. The Financial Times has reported, citing unnamed sources, that the company is now valued at roughly $7 billion, though Neko has not confirmed a figure publicly.

2: Inside the AI Body Scan: Technology, Claims, and Open Questions:

Neko's 60-minute, radiation-free assessment combines an electrocardiogram, arterial measurements, body-composition analysis, blood work processed on-site, and more than 2,000 high-resolution skin images, all reviewed with a clinician before the customer leaves the clinic.

The company currently runs eight clinics across the UK and Sweden, pricing scans at £299 and 2,750 Swedish kronor respectively — roughly $400 and $285. It says it has completed 100,000 scans since 2023 and that 75% of customers prepay for a repeat visit on the spot, a model that lets clinicians track changes in skin and vital signs over time.

In the US, two of Neko's proprietary devices — Derma-2 and Spectrum-2 — received FDA 510(k) clearance in May 2026, but those clearances cover the individual devices and their specific intended uses rather than the full Neko Health Scan as a single approved service. Neko positions its US clinics as preventive wellness providers, not full medical practices, and its own privacy notice tells customers to keep seeing their regular clinicians for diagnosis and treatment.

Neko's health-outcome data comes from 1,469 returning customers, and the company itself has been explicit that the analysis was not a controlled scientific study.

That data point is worth sitting with: no control group was used, participants may have changed behavior or started treatment between visits, and the group who returned for a second paid scan was not shown to be representative of Neko's broader customer base. US pricing, insurance coverage (currently none), and false-positive rates also remain undisclosed — open questions any enterprise evaluating consumer health AI should watch closely as the category scales.

3: Google DeepMind's Three-Pillar Bioresilience Strategy:

While Neko scales AI-driven diagnostics for individuals, Google DeepMind and Isomorphic Labs are tackling the opposite side of the AI-and-biology equation: keeping the same underlying capabilities out of the hands of bad actors.

The two organizations detailed a year-old bioresilience program built around three pillars — preventing misuse, detecting outbreaks faster, and responding once one is underway — now backed by more than 15 partnerships with government and biosecurity bodies, including Lawrence Livermore National Laboratory, the UK AI Security Institute, CEPI, and the Francis Crick Institute.

● Prevention: threat modeling, red-teaming, randomized controlled trials, and post-training refusal tuning for Gemini, plus an early-stage effort to adapt DeepMind's SynthID watermarking to flag AI-designed DNA sequences that current synthesis-screening tools might miss.

● Detection: pushing down the cost of metagenomic sequencing, including a Google–Pacific Biosciences collaboration that used the AlphaEvolve coding agent to improve sequencing accuracy.

● Response: leaning on more than 10,000 infectious-disease publications that have cited AlphaFold, plus a new Lawrence Livermore partnership using AlphaFold 3 for antibody design, including a pan-filovirus effort.

DeepMind is candid that none of this is a finished system — it describes the safeguards as an ongoing process, not a solved problem, which matters for any organization weighing how much to lean on today's AI safety tooling.

4: The Policy Gap and What Comes Next:

DeepMind's recommendations to US policymakers map directly onto its three pillars, backing specific pending bills such as the AI-Ready Bio-Data Standards Act, the Biosecurity Modernization and Innovation Act, and the SCALE Biology Act.

None of that legislation has been enacted yet, and DeepMind itself frames the next six to twelve months as the real test of whether policy catches up to the technology. The same tension shows up in Neko's expansion: the company has FDA clearance for individual components, celebrity-studded funding, and a fast-growing customer base, but no completed peer-reviewed validation of its full screening service and no disclosed US pricing or insurance strategy.

Together, the two stories capture where consumer and enterprise AI in health and biosecurity actually stand in mid-2026: real capability, real capital, and a governance layer that is still being built in parallel rather than finished in advance.