H
Overview
hum.ai develops AI software to measure ocean health by fusing satellite imagery and ground-truth data with multimodal geospatial models. It trains on the SeeFar dataset to combine historical data with high-resolution imagery, enabling detailed analytics for conservation, carbon removal, and government monitoring. The models work across multiple satellite types and resolutions, using a large curated dataset to avoid being tied to a single satellite. The long-term goal is a nature-focused AGI that understands and helps manage the natural world, starting with ocean ecosystems to support conservation and environmental policy.
Work here?Claim your company
Simplify's Take
What believers are saying
- SeeFar's 2026 update shows active dataset expansion and ongoing model-training momentum.
- Open positions in 2026 show AI-research hiring despite the small team.
- Climate and AI backers plus government, conservation, and carbon-removal use cases broaden customer demand.
What critics are saying
- PitchBook still showed only $1.68 million raised through June 2024, starving compute runway.
- Built In says $10M+ raised, conflicting with PitchBook and signaling opaque disclosure.
- If pilots never convert to contracts by 2027, hum.ai's nature-AGI thesis dies.
What makes hum.ai unique
- SeeFar, updated July 19, 2026, trains satellite-agnostic geospatial foundation models.
- hum.ai targets ocean-health measurement, not generic chatbots, for conservation and blue-carbon workflows.
- Kitchener-founded team of PhDs builds multimodal models using public and commercial imagery.
Help us improve and share your feedback! Did you find this helpful?