Is Asimov profitable in 2026?

No revenue figures have been disclosed. Asimov is a 3-person, pre-seed startup with $500K raised on a convertible note as of June 15, 2026, per CB Insights. It has made no public profitability claims, and no independent source confirms paying customers.

Who is Asimov's biggest competitor?

Scale AI is Asimov's biggest threat: its Physical AI Data Engine was already shipping by April 29, 2026, and Scale AI holds hundreds of millions in capital versus Asimov's $500K, per Bouken Capital and Seedtable reporting.

Could AI replace what Asimov does?

Partly. Asimov's annotation layer (pose, depth, segmentation) is increasingly automatable by AI auto-labeling tools, and NVIDIA's June 2026 EgoScale dataset (20,000+ hours) shows open egocentric data can rival proprietary footage, threatening its core data-scarcity value.

Does Asimov have real paying customers?

Unconfirmed. Asimov claims on its YC launch page (March 9, 2026) to already supply 'the largest players in the space,' but no customer logos or contracts have been independently verified as of July 23, 2026, per tryasimov.ai review.

Who owns and runs Asimov?

Asimov is privately held and founder-led by CEO Lyem Ningthou (ex-USAF data pipelines) and CTO Anshul Verma (ex-Scale AI/Amazon, Berkeley ML), per YC's company page. It is backed by Y Combinator, Robinhood Ventures, and Bouken Capital as of April 29, 2026.

How much funding has Asimov raised?

Asimov has raised $500K total on a convertible note, with its last raise around January 2026, according to a CB Insights snapshot dated June 15, 2026. No priced seed or Series A round had been disclosed as of that date.

YC Winter 2026 · United States

Asimov

Startup paying people to film themselves moving, to train humanoid robot AI.

Losing its protection

Assessed 27 July 2026 · Evidence as of 27 July 2026

The structural protection is eroding.

On the recordSince 27 July 2026 · Checkable from Aug 2026 · Every call on the record Checkpoint: EU AI Act Article 10 provenance rules begin phasing in

The call

Asimov is chasing a real data need with no confirmed customers yet, while bigger, richer players already ship similar or larger datasets.

Asimov collects video of human movement from people's homes and sells it to companies building humanoid robots. The problem is that bigger companies like Scale AI are already selling the same thing, NVIDIA and Apple are giving away similar videos for free, and six other startups are chasing the same customers. The company's team knows how to build data pipelines, but no paying customer has been publicly named yet. The one thing that could save Asimov is if it becomes the trusted choice for companies that need proof the video data is legally and ethically collected (especially as Europe and US states tighten rules about recording people), but that only works if the company can sign up a real robot company and prove it, and if a major tech lab does not release free video data that makes Asimov's paid product unnecessary.

Why

AI pressure: NVIDIA EgoScale and Apple EgoDex open egocentric datasets. Frontier labs are already releasing large open human-movement datasets, cutting into the scarcity Asimov sells. Could AI do the core job? Yes, likely.

Bigger players: Scale AI's Physical AI Data Engine. Scale AI already sells the same kind of robotics data at far greater scale and funding, with no barrier stopping it from copying Asimov's approach. Could a giant copy it? Yes, easily.

What protects it. Nothing proven yet. A possible edge exists only if regulators force provenance-verified consent data that Scale and open datasets can't easily match.

What would change this call

Asimov names an independently confirmed paying lab customer with a multi-year contract by Q1 2027. Would show real recurring demand instead of vendor-only traction claims.
A frontier lab releases an open egocentric dataset of 10,000+ hours within 18 months. Would commoditize the data scarcity Asimov's pitch depends on.
Scale AI discloses new customer wins or pricing for its Physical AI Data Engine. Would show the better-funded incumbent capturing the same lab customers.

Key numbers

$500K raised on convertible note (2026-06-15)
3 employees total (2026-07-07)
NVIDIA EgoScale trained on 20,000+ egocentric hours (2026-06-18)

Named risks

No independently confirmed paying customer or contract despite vendor claims of frontier-lab trust. Severity: high
Open-sourced egocentric datasets from NVIDIA and Apple could commoditize data scarcity within 12-18 months. Severity: high
Massively outfunded by Scale AI, which has no structural barrier to copying the crowdsourced model. Severity: high
EU AI Act Article 10 provenance rules (phasing 2026-2027) and Illinois BIPA fines expose in-home biometric recording. Severity: medium
Annotation layer (pose, depth, segmentation) increasingly automatable, eroding processing-side value. Severity: medium

Rivals, as read in July 2026

Rival	Direction	Evidence
Scale AI	losing to	Physical AI Data Engine already shipping by 2026-04-29, holds hundreds of millions vs Asimov's $500K
NVIDIA (EgoScale) and Apple (EgoDex)	losing to	Open-sourced 20,000+ hr egocentric datasets as of 2026-06-18 threaten data scarcity value
Lightwheel, Build AI	contesting	Funded rivals racing same egocentric-data wedge, cited 2026-03-24
Human Archive, Humyn Labs, Egolab AI, Neocambrian	contesting	India-based entrants racing same space, reported 2026-06-11
Physical Intelligence, AgiBot World	contesting	Internal fleets and 1M+ open trajectories reduce need to buy third-party data, per 2026-07-08 catalog

Questions people ask

Is Asimov profitable in 2026?: No revenue figures have been disclosed. Asimov is a 3-person, pre-seed startup with $500K raised on a convertible note as of June 15, 2026, per CB Insights. It has made no public profitability claims, and no independent source confirms paying customers.
Who is Asimov's biggest competitor?: Scale AI is Asimov's biggest threat: its Physical AI Data Engine was already shipping by April 29, 2026, and Scale AI holds hundreds of millions in capital versus Asimov's $500K, per Bouken Capital and Seedtable reporting.
Could AI replace what Asimov does?: Partly. Asimov's annotation layer (pose, depth, segmentation) is increasingly automatable by AI auto-labeling tools, and NVIDIA's June 2026 EgoScale dataset (20,000+ hours) shows open egocentric data can rival proprietary footage, threatening its core data-scarcity value.
Does Asimov have real paying customers?: Unconfirmed. Asimov claims on its YC launch page (March 9, 2026) to already supply 'the largest players in the space,' but no customer logos or contracts have been independently verified as of July 23, 2026, per tryasimov.ai review.
Who owns and runs Asimov?: Asimov is privately held and founder-led by CEO Lyem Ningthou (ex-USAF data pipelines) and CTO Anshul Verma (ex-Scale AI/Amazon, Berkeley ML), per YC's company page. It is backed by Y Combinator, Robinhood Ventures, and Bouken Capital as of April 29, 2026.
How much funding has Asimov raised?: Asimov has raised $500K total on a convertible note, with its last raise around January 2026, according to a CB Insights snapshot dated June 15, 2026. No priced seed or Series A round had been disclosed as of that date.

Timeline

2026-07-07YC directory lists Asimov at 3 employees in San Francisco
2026-06-18NVIDIA EgoScale paper shows 20,000+ hr egocentric dataset, 54% gain
2026-06-15CB Insights snapshot: $500K raised total, convertible note, alive status
2026-06-11Times of India reports India-based egocentric-data rivals racing same niche
2026-05-31StartupHub reports founders ran side cleaning business for data access
2026-04-29Bouken Capital substack details investors and Scale AI comparison
2026-03-24Labellerr blog ranks Asimov among top egocentric data providers
2026-03-09Asimov launches publicly via YC Winter 2026 batch
2026-01Implied last funding raise prior to CB Insights June snapshot

Revision history

27 July 2026

First assessed. Losing its protection. The text above is fixed as of this date and is not edited afterwards.

In the same band

CellType OctaPulse The Token Company